diff --git "a/q-iql-models/antmaze-large-diverse-v0/0/debug.log" "b/q-iql-models/antmaze-large-diverse-v0/0/debug.log" new file mode 100644--- /dev/null +++ "b/q-iql-models/antmaze-large-diverse-v0/0/debug.log" @@ -0,0 +1,100000 @@ +2022-05-15 18:03:24.733047 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -1000 finished +------------------------------ ---------------- +epoch -1000 +replay_buffer/size 999047 +trainer/num train calls 1000 +trainer/QF1 Loss 1.00257 +trainer/QF2 Loss 0.989533 +trainer/Policy Loss 1025.02 +trainer/Q1 Predictions Mean 0.00207004 +trainer/Q1 Predictions Std 0.00210315 +trainer/Q1 Predictions Max 0.00931837 +trainer/Q1 Predictions Min -0.00224991 +trainer/Q2 Predictions Mean -0.00448099 +trainer/Q2 Predictions Std 0.00487994 +trainer/Q2 Predictions Max 0.00682312 +trainer/Q2 Predictions Min -0.0176247 +trainer/Q Targets Mean -0.997269 +trainer/Q Targets Std 0.0624953 +trainer/Q Targets Max 0 +trainer/Q Targets Min -1.00788 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000225342 +trainer/policy/mean Std 0.00227665 +trainer/policy/mean Max 0.00681801 +trainer/policy/mean Min -0.00836451 +trainer/policy/std Mean 0.0497871 +trainer/policy/std Std 3.72529e-09 +trainer/policy/std Max 0.0497871 +trainer/policy/std Min 0.0497871 +trainer/Advantage Weights Mean 0.955804 +trainer/Advantage Weights Std 0.0337305 +trainer/Advantage Weights Max 1.02027 +trainer/Advantage Weights Min 0.852412 +trainer/Advantage Score Mean -0.00458297 +trainer/Advantage Score Std 0.003549 +trainer/Advantage Score Max 0.00200714 +trainer/Advantage Score Min -0.0159685 +trainer/V1 Predictions Mean -0.00124363 +trainer/V1 Predictions Std 0.00220638 +trainer/V1 Predictions Max 0.00547463 +trainer/V1 Predictions Min -0.00693567 +trainer/VF Loss 3.43217e-06 +expl/num steps total 1000 +expl/num paths total 1 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.000316163 +expl/Actions Std 0.0494872 +expl/Actions Max 0.212291 +expl/Actions Min -0.165469 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 1000 +eval/num paths total 1 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 2.35406e-05 +eval/Actions Std 0.000146142 +eval/Actions Max 0.00173487 +eval/Actions Min -0.000817526 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.64521e-06 +time/evaluation sampling (s) 8.73114 +time/exploration sampling (s) 2.94519 +time/logging (s) 0.0109498 +time/saving (s) 0.0283455 +time/training (s) 14.7013 +time/epoch (s) 26.4169 +time/total (s) 59.0629 +Epoch -1000 +------------------------------ ---------------- +2022-05-15 18:03:46.984956 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -999 finished +------------------------------ ---------------- +epoch -999 +replay_buffer/size 999047 +trainer/num train calls 2000 +trainer/QF1 Loss 0.615876 +trainer/QF2 Loss 0.62767 +trainer/Policy Loss 192.961 +trainer/Q1 Predictions Mean -7.70818 +trainer/Q1 Predictions Std 0.813215 +trainer/Q1 Predictions Max -5.04476 +trainer/Q1 Predictions Min -9.948 +trainer/Q2 Predictions Mean -7.72284 +trainer/Q2 Predictions Std 0.813998 +trainer/Q2 Predictions Max -5.06266 +trainer/Q2 Predictions Min -9.86367 +trainer/Q Targets Mean -7.53493 +trainer/Q Targets Std 1.10076 +trainer/Q Targets Max 0 +trainer/Q Targets Min -10.8718 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0197955 +trainer/policy/mean Std 0.560424 +trainer/policy/mean Max 0.986774 +trainer/policy/mean Min -0.991378 +trainer/policy/std Mean 0.0669362 +trainer/policy/std Std 0.0010617 +trainer/policy/std Max 0.0682033 +trainer/policy/std Min 0.0649377 +trainer/Advantage Weights Mean 0.863956 +trainer/Advantage Weights Std 2.95913 +trainer/Advantage Weights Max 35.1823 +trainer/Advantage Weights Min 0.000125984 +trainer/Advantage Score Mean -0.172603 +trainer/Advantage Score Std 0.181415 +trainer/Advantage Score Max 0.356054 +trainer/Advantage Score Min -0.897936 +trainer/V1 Predictions Mean -6.69737 +trainer/V1 Predictions Std 0.709146 +trainer/V1 Predictions Max -4.8666 +trainer/V1 Predictions Min -8.43672 +trainer/VF Loss 0.00832547 +expl/num steps total 2000 +expl/num paths total 2 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0269915 +expl/Actions Std 0.337596 +expl/Actions Max 0.87078 +expl/Actions Min -0.920335 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 2000 +eval/num paths total 2 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0266157 +eval/Actions Std 0.331182 +eval/Actions Max 0.760732 +eval/Actions Min -0.88715 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.46503e-06 +time/evaluation sampling (s) 3.27025 +time/exploration sampling (s) 3.98622 +time/logging (s) 0.00905291 +time/saving (s) 0.0127765 +time/training (s) 14.9633 +time/epoch (s) 22.2416 +time/total (s) 81.3118 +Epoch -999 +------------------------------ ---------------- +2022-05-15 18:04:09.167596 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -998 finished +------------------------------ ---------------- +epoch -998 +replay_buffer/size 999047 +trainer/num train calls 3000 +trainer/QF1 Loss 0.280229 +trainer/QF2 Loss 0.287352 +trainer/Policy Loss 81.3122 +trainer/Q1 Predictions Mean -14.0065 +trainer/Q1 Predictions Std 1.86714 +trainer/Q1 Predictions Max -8.63443 +trainer/Q1 Predictions Min -18.8645 +trainer/Q2 Predictions Mean -13.9884 +trainer/Q2 Predictions Std 1.87737 +trainer/Q2 Predictions Max -8.59716 +trainer/Q2 Predictions Min -18.7119 +trainer/Q Targets Mean -13.7398 +trainer/Q Targets Std 1.79551 +trainer/Q Targets Max -9.02099 +trainer/Q Targets Min -18.8615 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0263206 +trainer/policy/mean Std 0.564821 +trainer/policy/mean Max 0.991258 +trainer/policy/mean Min -0.977446 +trainer/policy/std Mean 0.0946775 +trainer/policy/std Std 0.00460392 +trainer/policy/std Max 0.0990565 +trainer/policy/std Min 0.0868076 +trainer/Advantage Weights Mean 0.736781 +trainer/Advantage Weights Std 5.38174 +trainer/Advantage Weights Max 71.5831 +trainer/Advantage Weights Min 4.1669e-09 +trainer/Advantage Score Mean -0.500554 +trainer/Advantage Score Std 0.340109 +trainer/Advantage Score Max 0.427086 +trainer/Advantage Score Min -1.92961 +trainer/V1 Predictions Mean -12.8401 +trainer/V1 Predictions Std 1.87271 +trainer/V1 Predictions Max -7.8856 +trainer/V1 Predictions Min -17.8583 +trainer/VF Loss 0.0384078 +expl/num steps total 3000 +expl/num paths total 3 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.035766 +expl/Actions Std 0.33333 +expl/Actions Max 0.947338 +expl/Actions Min -0.806119 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 3000 +eval/num paths total 3 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0390711 +eval/Actions Std 0.320317 +eval/Actions Max 0.875001 +eval/Actions Min -0.849529 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78512e-06 +time/evaluation sampling (s) 2.94361 +time/exploration sampling (s) 4.08534 +time/logging (s) 0.00900041 +time/saving (s) 0.0117776 +time/training (s) 15.1284 +time/epoch (s) 22.1781 +time/total (s) 103.494 +Epoch -998 +------------------------------ ---------------- +2022-05-15 18:04:30.308833 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -997 finished +------------------------------ ---------------- +epoch -997 +replay_buffer/size 999047 +trainer/num train calls 4000 +trainer/QF1 Loss 1.04456 +trainer/QF2 Loss 0.989679 +trainer/Policy Loss 337.23 +trainer/Q1 Predictions Mean -18.8319 +trainer/Q1 Predictions Std 3.43731 +trainer/Q1 Predictions Max -8.59776 +trainer/Q1 Predictions Min -25.1828 +trainer/Q2 Predictions Mean -18.7826 +trainer/Q2 Predictions Std 3.43018 +trainer/Q2 Predictions Max -8.39671 +trainer/Q2 Predictions Min -25.2076 +trainer/Q Targets Mean -18.7358 +trainer/Q Targets Std 3.39834 +trainer/Q Targets Max 0 +trainer/Q Targets Min -25.1637 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0196048 +trainer/policy/mean Std 0.572457 +trainer/policy/mean Max 0.982793 +trainer/policy/mean Min -0.982721 +trainer/policy/std Mean 0.12658 +trainer/policy/std Std 0.00602946 +trainer/policy/std Max 0.132156 +trainer/policy/std Min 0.11656 +trainer/Advantage Weights Mean 6.17651 +trainer/Advantage Weights Std 20.9749 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.50219e-09 +trainer/Advantage Score Mean -0.249375 +trainer/Advantage Score Std 0.41421 +trainer/Advantage Score Max 1.00492 +trainer/Advantage Score Min -1.92187 +trainer/V1 Predictions Mean -17.9716 +trainer/V1 Predictions Std 3.27084 +trainer/V1 Predictions Max -8.52448 +trainer/V1 Predictions Min -24.4575 +trainer/VF Loss 0.0459341 +expl/num steps total 4000 +expl/num paths total 4 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0448729 +expl/Actions Std 0.350205 +expl/Actions Max 0.828904 +expl/Actions Min -0.923693 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 4000 +eval/num paths total 4 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0348088 +eval/Actions Std 0.318122 +eval/Actions Max 0.691621 +eval/Actions Min -0.726173 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.61702e-06 +time/evaluation sampling (s) 2.85676 +time/exploration sampling (s) 3.78197 +time/logging (s) 0.0084958 +time/saving (s) 0.00982122 +time/training (s) 14.4786 +time/epoch (s) 21.1356 +time/total (s) 124.634 +Epoch -997 +------------------------------ ---------------- +2022-05-15 18:04:52.439252 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -996 finished +------------------------------ ---------------- +epoch -996 +replay_buffer/size 999047 +trainer/num train calls 5000 +trainer/QF1 Loss 0.346142 +trainer/QF2 Loss 0.334509 +trainer/Policy Loss 65.6093 +trainer/Q1 Predictions Mean -22.9837 +trainer/Q1 Predictions Std 4.70125 +trainer/Q1 Predictions Max -7.42651 +trainer/Q1 Predictions Min -34.492 +trainer/Q2 Predictions Mean -22.9723 +trainer/Q2 Predictions Std 4.68988 +trainer/Q2 Predictions Max -7.0671 +trainer/Q2 Predictions Min -34.4184 +trainer/Q Targets Mean -22.968 +trainer/Q Targets Std 4.54665 +trainer/Q Targets Max -8.49027 +trainer/Q Targets Min -35.062 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000527946 +trainer/policy/mean Std 0.54675 +trainer/policy/mean Max 0.945632 +trainer/policy/mean Min -0.95507 +trainer/policy/std Mean 0.158789 +trainer/policy/std Std 0.00665132 +trainer/policy/std Max 0.166043 +trainer/policy/std Min 0.14794 +trainer/Advantage Weights Mean 2.5831 +trainer/Advantage Weights Std 12.8308 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.50766e-09 +trainer/Advantage Score Mean -0.49439 +trainer/Advantage Score Std 0.488982 +trainer/Advantage Score Max 1.25326 +trainer/Advantage Score Min -2.03127 +trainer/V1 Predictions Mean -22.1877 +trainer/V1 Predictions Std 4.60419 +trainer/V1 Predictions Max -7.86808 +trainer/V1 Predictions Min -34.5178 +trainer/VF Loss 0.0603215 +expl/num steps total 5000 +expl/num paths total 5 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0229775 +expl/Actions Std 0.340078 +expl/Actions Max 0.940704 +expl/Actions Min -1.07274 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 5000 +eval/num paths total 5 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0134713 +eval/Actions Std 0.297244 +eval/Actions Max 0.683559 +eval/Actions Min -0.684219 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.03425e-06 +time/evaluation sampling (s) 2.98237 +time/exploration sampling (s) 3.85803 +time/logging (s) 0.00775476 +time/saving (s) 0.0103334 +time/training (s) 15.2666 +time/epoch (s) 22.1251 +time/total (s) 146.763 +Epoch -996 +------------------------------ ---------------- +2022-05-15 18:05:14.215118 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -995 finished +------------------------------ ---------------- +epoch -995 +replay_buffer/size 999047 +trainer/num train calls 6000 +trainer/QF1 Loss 1.33151 +trainer/QF2 Loss 1.41493 +trainer/Policy Loss 245.075 +trainer/Q1 Predictions Mean -26.2367 +trainer/Q1 Predictions Std 6.31939 +trainer/Q1 Predictions Max -8.52483 +trainer/Q1 Predictions Min -43.032 +trainer/Q2 Predictions Mean -26.2212 +trainer/Q2 Predictions Std 6.26755 +trainer/Q2 Predictions Max -8.1524 +trainer/Q2 Predictions Min -42.8314 +trainer/Q Targets Mean -26.4565 +trainer/Q Targets Std 6.43569 +trainer/Q Targets Max 0 +trainer/Q Targets Min -43.7988 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0131185 +trainer/policy/mean Std 0.544686 +trainer/policy/mean Max 0.986688 +trainer/policy/mean Min -0.986694 +trainer/policy/std Mean 0.191915 +trainer/policy/std Std 0.00715731 +trainer/policy/std Max 0.200798 +trainer/policy/std Min 0.180456 +trainer/Advantage Weights Mean 9.66432 +trainer/Advantage Weights Std 26.6607 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.63264e-15 +trainer/Advantage Score Mean -0.283949 +trainer/Advantage Score Std 0.573792 +trainer/Advantage Score Max 2.11461 +trainer/Advantage Score Min -3.40486 +trainer/V1 Predictions Mean -25.8231 +trainer/V1 Predictions Std 6.208 +trainer/V1 Predictions Max -8.81223 +trainer/V1 Predictions Min -43.1424 +trainer/VF Loss 0.0913371 +expl/num steps total 6000 +expl/num paths total 6 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0113137 +expl/Actions Std 0.358541 +expl/Actions Max 1.01288 +expl/Actions Min -1.03876 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 6000 +eval/num paths total 6 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0177001 +eval/Actions Std 0.296143 +eval/Actions Max 0.57537 +eval/Actions Min -0.616091 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.88178e-06 +time/evaluation sampling (s) 2.91818 +time/exploration sampling (s) 3.86046 +time/logging (s) 0.00710124 +time/saving (s) 0.0101228 +time/training (s) 14.9745 +time/epoch (s) 21.7703 +time/total (s) 168.537 +Epoch -995 +------------------------------ ---------------- +2022-05-15 18:05:35.727326 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -994 finished +------------------------------ ---------------- +epoch -994 +replay_buffer/size 999047 +trainer/num train calls 7000 +trainer/QF1 Loss 2.33987 +trainer/QF2 Loss 2.42492 +trainer/Policy Loss 110.073 +trainer/Q1 Predictions Mean -30.3616 +trainer/Q1 Predictions Std 7.61007 +trainer/Q1 Predictions Max -1.76793 +trainer/Q1 Predictions Min -42.0508 +trainer/Q2 Predictions Mean -30.4217 +trainer/Q2 Predictions Std 7.64217 +trainer/Q2 Predictions Max -1.44802 +trainer/Q2 Predictions Min -42.2258 +trainer/Q Targets Mean -30.3007 +trainer/Q Targets Std 8.02847 +trainer/Q Targets Max 0 +trainer/Q Targets Min -42.461 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0303425 +trainer/policy/mean Std 0.573362 +trainer/policy/mean Max 0.965829 +trainer/policy/mean Min -0.977767 +trainer/policy/std Mean 0.224917 +trainer/policy/std Std 0.00741598 +trainer/policy/std Max 0.233981 +trainer/policy/std Min 0.212567 +trainer/Advantage Weights Mean 6.22 +trainer/Advantage Weights Std 21.1139 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.32841e-11 +trainer/Advantage Score Mean -0.333053 +trainer/Advantage Score Std 0.489436 +trainer/Advantage Score Max 1.1092 +trainer/Advantage Score Min -2.50445 +trainer/V1 Predictions Mean -29.7437 +trainer/V1 Predictions Std 7.58351 +trainer/V1 Predictions Max -1.38145 +trainer/V1 Predictions Min -41.8811 +trainer/VF Loss 0.0551472 +expl/num steps total 7000 +expl/num paths total 7 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00676186 +expl/Actions Std 0.379831 +expl/Actions Max 1.15308 +expl/Actions Min -1.36107 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 7000 +eval/num paths total 7 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0115342 +eval/Actions Std 0.305239 +eval/Actions Max 0.610941 +eval/Actions Min -0.749122 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.14183e-06 +time/evaluation sampling (s) 3.02499 +time/exploration sampling (s) 3.88987 +time/logging (s) 0.00979053 +time/saving (s) 0.0138836 +time/training (s) 14.5713 +time/epoch (s) 21.5099 +time/total (s) 190.052 +Epoch -994 +------------------------------ ---------------- +2022-05-15 18:05:57.372506 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -993 finished +------------------------------ ---------------- +epoch -993 +replay_buffer/size 999047 +trainer/num train calls 8000 +trainer/QF1 Loss 2.46118 +trainer/QF2 Loss 2.39195 +trainer/Policy Loss 197.518 +trainer/Q1 Predictions Mean -33.1033 +trainer/Q1 Predictions Std 8.78979 +trainer/Q1 Predictions Max -4.82908 +trainer/Q1 Predictions Min -50.8093 +trainer/Q2 Predictions Mean -33.1385 +trainer/Q2 Predictions Std 8.82804 +trainer/Q2 Predictions Max -4.45095 +trainer/Q2 Predictions Min -51.091 +trainer/Q Targets Mean -33.8289 +trainer/Q Targets Std 9.02954 +trainer/Q Targets Max 0 +trainer/Q Targets Min -53.202 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0042686 +trainer/policy/mean Std 0.554147 +trainer/policy/mean Max 0.972002 +trainer/policy/mean Min -0.984648 +trainer/policy/std Mean 0.258433 +trainer/policy/std Std 0.00808796 +trainer/policy/std Max 0.268244 +trainer/policy/std Min 0.245199 +trainer/Advantage Weights Mean 15.6058 +trainer/Advantage Weights Std 31.6541 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.93839e-09 +trainer/Advantage Score Mean -0.0715881 +trainer/Advantage Score Std 0.472535 +trainer/Advantage Score Max 1.34514 +trainer/Advantage Score Min -1.93525 +trainer/V1 Predictions Mean -33.3547 +trainer/V1 Predictions Std 8.81827 +trainer/V1 Predictions Max -5.3605 +trainer/V1 Predictions Min -52.4674 +trainer/VF Loss 0.0881348 +expl/num steps total 8000 +expl/num paths total 8 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0175386 +expl/Actions Std 0.395475 +expl/Actions Max 1.39142 +expl/Actions Min -1.66452 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 8000 +eval/num paths total 8 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0535722 +eval/Actions Std 0.290831 +eval/Actions Max 0.614129 +eval/Actions Min -0.595681 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.6729e-06 +time/evaluation sampling (s) 2.79103 +time/exploration sampling (s) 3.80609 +time/logging (s) 0.00772004 +time/saving (s) 0.01202 +time/training (s) 15.0184 +time/epoch (s) 21.6352 +time/total (s) 211.694 +Epoch -993 +------------------------------ ---------------- +2022-05-15 18:06:18.911496 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -992 finished +------------------------------ ---------------- +epoch -992 +replay_buffer/size 999047 +trainer/num train calls 9000 +trainer/QF1 Loss 2.87828 +trainer/QF2 Loss 2.8106 +trainer/Policy Loss 30.7785 +trainer/Q1 Predictions Mean -37.7051 +trainer/Q1 Predictions Std 8.63365 +trainer/Q1 Predictions Max -7.2998 +trainer/Q1 Predictions Min -50.5999 +trainer/Q2 Predictions Mean -37.6872 +trainer/Q2 Predictions Std 8.60891 +trainer/Q2 Predictions Max -7.97101 +trainer/Q2 Predictions Min -50.6815 +trainer/Q Targets Mean -37.4681 +trainer/Q Targets Std 8.77545 +trainer/Q Targets Max 0 +trainer/Q Targets Min -50.7191 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0167487 +trainer/policy/mean Std 0.560997 +trainer/policy/mean Max 0.977206 +trainer/policy/mean Min -0.967204 +trainer/policy/std Mean 0.292607 +trainer/policy/std Std 0.00878709 +trainer/policy/std Max 0.303832 +trainer/policy/std Min 0.278734 +trainer/Advantage Weights Mean 2.61595 +trainer/Advantage Weights Std 13.8398 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.69633e-11 +trainer/Advantage Score Mean -0.5383 +trainer/Advantage Score Std 0.493261 +trainer/Advantage Score Max 0.865158 +trainer/Advantage Score Min -2.30567 +trainer/V1 Predictions Mean -36.9277 +trainer/V1 Predictions Std 8.65335 +trainer/V1 Predictions Max -6.04016 +trainer/V1 Predictions Min -50.2477 +trainer/VF Loss 0.0624136 +expl/num steps total 9000 +expl/num paths total 9 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.012922 +expl/Actions Std 0.443017 +expl/Actions Max 1.52735 +expl/Actions Min -1.6758 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 9000 +eval/num paths total 9 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0384492 +eval/Actions Std 0.308525 +eval/Actions Max 0.548278 +eval/Actions Min -0.734377 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.48318e-06 +time/evaluation sampling (s) 2.82372 +time/exploration sampling (s) 3.78465 +time/logging (s) 0.0084051 +time/saving (s) 0.015081 +time/training (s) 14.9023 +time/epoch (s) 21.5342 +time/total (s) 233.232 +Epoch -992 +------------------------------ ---------------- +2022-05-15 18:06:41.025351 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -991 finished +------------------------------ ---------------- +epoch -991 +replay_buffer/size 999047 +trainer/num train calls 10000 +trainer/QF1 Loss 1.22292 +trainer/QF2 Loss 1.25702 +trainer/Policy Loss 76.5294 +trainer/Q1 Predictions Mean -39.6281 +trainer/Q1 Predictions Std 10.2922 +trainer/Q1 Predictions Max -6.54404 +trainer/Q1 Predictions Min -54.9219 +trainer/Q2 Predictions Mean -39.6725 +trainer/Q2 Predictions Std 10.2638 +trainer/Q2 Predictions Max -6.63178 +trainer/Q2 Predictions Min -54.8489 +trainer/Q Targets Mean -40.2153 +trainer/Q Targets Std 10.4116 +trainer/Q Targets Max 0 +trainer/Q Targets Min -55.7245 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00294554 +trainer/policy/mean Std 0.566184 +trainer/policy/mean Max 0.958016 +trainer/policy/mean Min -0.977254 +trainer/policy/std Mean 0.325952 +trainer/policy/std Std 0.00978878 +trainer/policy/std Max 0.338647 +trainer/policy/std Min 0.311256 +trainer/Advantage Weights Mean 9.23747 +trainer/Advantage Weights Std 24.1486 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.01543e-10 +trainer/Advantage Score Mean -0.192229 +trainer/Advantage Score Std 0.516262 +trainer/Advantage Score Max 1.30641 +trainer/Advantage Score Min -2.2325 +trainer/V1 Predictions Mean -39.6515 +trainer/V1 Predictions Std 10.3887 +trainer/V1 Predictions Max -5.92582 +trainer/V1 Predictions Min -55.0841 +trainer/VF Loss 0.065135 +expl/num steps total 10000 +expl/num paths total 10 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0138109 +expl/Actions Std 0.462724 +expl/Actions Max 1.60083 +expl/Actions Min -1.56682 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 10000 +eval/num paths total 10 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0220529 +eval/Actions Std 0.301595 +eval/Actions Max 0.611887 +eval/Actions Min -0.696864 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.59466e-06 +time/evaluation sampling (s) 2.997 +time/exploration sampling (s) 3.97427 +time/logging (s) 0.00674132 +time/saving (s) 0.0101141 +time/training (s) 15.1172 +time/epoch (s) 22.1053 +time/total (s) 255.344 +Epoch -991 +------------------------------ ---------------- +2022-05-15 18:07:02.952868 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -990 finished +------------------------------ ---------------- +epoch -990 +replay_buffer/size 999047 +trainer/num train calls 11000 +trainer/QF1 Loss 1.34354 +trainer/QF2 Loss 1.48341 +trainer/Policy Loss 50.6816 +trainer/Q1 Predictions Mean -43.326 +trainer/Q1 Predictions Std 10.9564 +trainer/Q1 Predictions Max -3.45411 +trainer/Q1 Predictions Min -63.9824 +trainer/Q2 Predictions Mean -43.3082 +trainer/Q2 Predictions Std 10.913 +trainer/Q2 Predictions Max -3.81989 +trainer/Q2 Predictions Min -64.2424 +trainer/Q Targets Mean -43.004 +trainer/Q Targets Std 11.1842 +trainer/Q Targets Max 0 +trainer/Q Targets Min -63.9846 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0288802 +trainer/policy/mean Std 0.573733 +trainer/policy/mean Max 0.967953 +trainer/policy/mean Min -0.985598 +trainer/policy/std Mean 0.360647 +trainer/policy/std Std 0.0112732 +trainer/policy/std Max 0.374666 +trainer/policy/std Min 0.344112 +trainer/Advantage Weights Mean 5.78606 +trainer/Advantage Weights Std 20.3851 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.56791e-11 +trainer/Advantage Score Mean -0.379656 +trainer/Advantage Score Std 0.478334 +trainer/Advantage Score Max 1.02271 +trainer/Advantage Score Min -2.48787 +trainer/V1 Predictions Mean -42.5952 +trainer/V1 Predictions Std 10.975 +trainer/V1 Predictions Max -3.46373 +trainer/V1 Predictions Min -63.4891 +trainer/VF Loss 0.0555057 +expl/num steps total 11000 +expl/num paths total 11 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0317081 +expl/Actions Std 0.501868 +expl/Actions Max 1.72408 +expl/Actions Min -2.20274 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 11000 +eval/num paths total 11 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0422527 +eval/Actions Std 0.373931 +eval/Actions Max 0.644048 +eval/Actions Min -0.773621 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.48315e-06 +time/evaluation sampling (s) 3.13499 +time/exploration sampling (s) 3.85483 +time/logging (s) 0.0107756 +time/saving (s) 0.0106979 +time/training (s) 14.9144 +time/epoch (s) 21.9257 +time/total (s) 277.275 +Epoch -990 +------------------------------ ---------------- +2022-05-15 18:07:25.267440 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -989 finished +------------------------------ ---------------- +epoch -989 +replay_buffer/size 999047 +trainer/num train calls 12000 +trainer/QF1 Loss 1.03313 +trainer/QF2 Loss 1.02419 +trainer/Policy Loss 22.4852 +trainer/Q1 Predictions Mean -44.8801 +trainer/Q1 Predictions Std 12.6916 +trainer/Q1 Predictions Max -5.17313 +trainer/Q1 Predictions Min -62.1833 +trainer/Q2 Predictions Mean -44.7817 +trainer/Q2 Predictions Std 12.6499 +trainer/Q2 Predictions Max -5.24959 +trainer/Q2 Predictions Min -61.9659 +trainer/Q Targets Mean -44.5201 +trainer/Q Targets Std 12.9262 +trainer/Q Targets Max 0 +trainer/Q Targets Min -61.2134 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0167279 +trainer/policy/mean Std 0.577513 +trainer/policy/mean Max 0.97127 +trainer/policy/mean Min -0.984004 +trainer/policy/std Mean 0.391728 +trainer/policy/std Std 0.0131284 +trainer/policy/std Max 0.407428 +trainer/policy/std Min 0.37261 +trainer/Advantage Weights Mean 3.705 +trainer/Advantage Weights Std 16.7809 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.6953e-12 +trainer/Advantage Score Mean -0.498698 +trainer/Advantage Score Std 0.507867 +trainer/Advantage Score Max 1.6622 +trainer/Advantage Score Min -2.58914 +trainer/V1 Predictions Mean -44.0171 +trainer/V1 Predictions Std 12.7319 +trainer/V1 Predictions Max -4.58833 +trainer/V1 Predictions Min -60.5098 +trainer/VF Loss 0.069295 +expl/num steps total 12000 +expl/num paths total 12 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0461321 +expl/Actions Std 0.556522 +expl/Actions Max 1.96185 +expl/Actions Min -1.99841 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 12000 +eval/num paths total 12 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.123376 +eval/Actions Std 0.454779 +eval/Actions Max 0.685275 +eval/Actions Min -0.77632 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.59606e-06 +time/evaluation sampling (s) 3.12656 +time/exploration sampling (s) 4.03072 +time/logging (s) 0.0114296 +time/saving (s) 0.0159527 +time/training (s) 15.1227 +time/epoch (s) 22.3074 +time/total (s) 299.589 +Epoch -989 +------------------------------ ---------------- +2022-05-15 18:07:47.679025 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -988 finished +------------------------------ ---------------- +epoch -988 +replay_buffer/size 999047 +trainer/num train calls 13000 +trainer/QF1 Loss 1.34815 +trainer/QF2 Loss 1.38845 +trainer/Policy Loss 34.7165 +trainer/Q1 Predictions Mean -46.6982 +trainer/Q1 Predictions Std 12.3764 +trainer/Q1 Predictions Max -6.06029 +trainer/Q1 Predictions Min -68.3835 +trainer/Q2 Predictions Mean -46.6746 +trainer/Q2 Predictions Std 12.3796 +trainer/Q2 Predictions Max -6.53872 +trainer/Q2 Predictions Min -67.8662 +trainer/Q Targets Mean -46.8719 +trainer/Q Targets Std 12.5224 +trainer/Q Targets Max 0 +trainer/Q Targets Min -68.169 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0180512 +trainer/policy/mean Std 0.578095 +trainer/policy/mean Max 0.979862 +trainer/policy/mean Min -0.978936 +trainer/policy/std Mean 0.420894 +trainer/policy/std Std 0.01545 +trainer/policy/std Max 0.438821 +trainer/policy/std Min 0.39945 +trainer/Advantage Weights Mean 5.62716 +trainer/Advantage Weights Std 20.3332 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.14748e-13 +trainer/Advantage Score Mean -0.405521 +trainer/Advantage Score Std 0.499182 +trainer/Advantage Score Max 0.798115 +trainer/Advantage Score Min -2.85111 +trainer/V1 Predictions Mean -46.4293 +trainer/V1 Predictions Std 12.3853 +trainer/V1 Predictions Max -5.52236 +trainer/V1 Predictions Min -67.5981 +trainer/VF Loss 0.0574824 +expl/num steps total 13000 +expl/num paths total 13 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0383436 +expl/Actions Std 0.559263 +expl/Actions Max 1.90344 +expl/Actions Min -2.0867 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 13000 +eval/num paths total 13 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.068835 +eval/Actions Std 0.384566 +eval/Actions Max 0.69292 +eval/Actions Min -0.793636 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.61609e-06 +time/evaluation sampling (s) 3.1929 +time/exploration sampling (s) 3.80255 +time/logging (s) 0.0122018 +time/saving (s) 0.0164402 +time/training (s) 15.3813 +time/epoch (s) 22.4054 +time/total (s) 322 +Epoch -988 +------------------------------ ---------------- +2022-05-15 18:08:09.467764 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -987 finished +------------------------------ ---------------- +epoch -987 +replay_buffer/size 999047 +trainer/num train calls 14000 +trainer/QF1 Loss 0.809979 +trainer/QF2 Loss 0.863356 +trainer/Policy Loss 32.7204 +trainer/Q1 Predictions Mean -49.0145 +trainer/Q1 Predictions Std 12.2687 +trainer/Q1 Predictions Max -5.02698 +trainer/Q1 Predictions Min -64.5602 +trainer/Q2 Predictions Mean -48.9947 +trainer/Q2 Predictions Std 12.2483 +trainer/Q2 Predictions Max -4.82833 +trainer/Q2 Predictions Min -64.1691 +trainer/Q Targets Mean -48.9966 +trainer/Q Targets Std 12.1621 +trainer/Q Targets Max 0 +trainer/Q Targets Min -63.8935 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00178558 +trainer/policy/mean Std 0.580825 +trainer/policy/mean Max 0.98561 +trainer/policy/mean Min -0.983971 +trainer/policy/std Mean 0.448121 +trainer/policy/std Std 0.0182518 +trainer/policy/std Max 0.468114 +trainer/policy/std Min 0.422297 +trainer/Advantage Weights Mean 5.69778 +trainer/Advantage Weights Std 20.4991 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.29029e-10 +trainer/Advantage Score Mean -0.350293 +trainer/Advantage Score Std 0.524942 +trainer/Advantage Score Max 3.31615 +trainer/Advantage Score Min -2.136 +trainer/V1 Predictions Mean -48.4991 +trainer/V1 Predictions Std 12.1393 +trainer/V1 Predictions Max -5.26323 +trainer/V1 Predictions Min -64.1107 +trainer/VF Loss 0.0964908 +expl/num steps total 14000 +expl/num paths total 14 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0193496 +expl/Actions Std 0.58096 +expl/Actions Max 1.82436 +expl/Actions Min -2.02871 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 14000 +eval/num paths total 14 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0722401 +eval/Actions Std 0.365842 +eval/Actions Max 0.646018 +eval/Actions Min -0.841875 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.60817e-06 +time/evaluation sampling (s) 3.22357 +time/exploration sampling (s) 3.84125 +time/logging (s) 0.00703009 +time/saving (s) 0.0096519 +time/training (s) 14.6953 +time/epoch (s) 21.7768 +time/total (s) 343.783 +Epoch -987 +------------------------------ ---------------- +2022-05-15 18:08:31.475247 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -986 finished +------------------------------ ---------------- +epoch -986 +replay_buffer/size 999047 +trainer/num train calls 15000 +trainer/QF1 Loss 0.94572 +trainer/QF2 Loss 0.981247 +trainer/Policy Loss 10.4143 +trainer/Q1 Predictions Mean -50.4736 +trainer/Q1 Predictions Std 14.1452 +trainer/Q1 Predictions Max -2.76885 +trainer/Q1 Predictions Min -71.9553 +trainer/Q2 Predictions Mean -50.5139 +trainer/Q2 Predictions Std 14.1321 +trainer/Q2 Predictions Max -2.59428 +trainer/Q2 Predictions Min -71.9336 +trainer/Q Targets Mean -50.2547 +trainer/Q Targets Std 14.332 +trainer/Q Targets Max 0 +trainer/Q Targets Min -71.8023 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00748196 +trainer/policy/mean Std 0.588619 +trainer/policy/mean Max 0.976623 +trainer/policy/mean Min -0.977167 +trainer/policy/std Mean 0.470759 +trainer/policy/std Std 0.0214464 +trainer/policy/std Max 0.492222 +trainer/policy/std Min 0.439499 +trainer/Advantage Weights Mean 1.90516 +trainer/Advantage Weights Std 11.6548 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.30664e-13 +trainer/Advantage Score Mean -0.542105 +trainer/Advantage Score Std 0.481749 +trainer/Advantage Score Max 0.888584 +trainer/Advantage Score Min -2.90978 +trainer/V1 Predictions Mean -49.828 +trainer/V1 Predictions Std 14.2192 +trainer/V1 Predictions Max -2.73669 +trainer/V1 Predictions Min -71.2648 +trainer/VF Loss 0.0596076 +expl/num steps total 15000 +expl/num paths total 15 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0570126 +expl/Actions Std 0.609715 +expl/Actions Max 2.42712 +expl/Actions Min -2.18294 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 15000 +eval/num paths total 15 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0922297 +eval/Actions Std 0.405952 +eval/Actions Max 0.730039 +eval/Actions Min -0.85792 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78279e-06 +time/evaluation sampling (s) 3.02204 +time/exploration sampling (s) 3.82855 +time/logging (s) 0.00818476 +time/saving (s) 0.0186557 +time/training (s) 15.126 +time/epoch (s) 22.0034 +time/total (s) 365.791 +Epoch -986 +------------------------------ ---------------- +2022-05-15 18:08:53.392530 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -985 finished +------------------------------ ---------------- +epoch -985 +replay_buffer/size 999047 +trainer/num train calls 16000 +trainer/QF1 Loss 1.93108 +trainer/QF2 Loss 1.89437 +trainer/Policy Loss 34.5013 +trainer/Q1 Predictions Mean -51.6401 +trainer/Q1 Predictions Std 14.4632 +trainer/Q1 Predictions Max -3.5732 +trainer/Q1 Predictions Min -73.4336 +trainer/Q2 Predictions Mean -51.6082 +trainer/Q2 Predictions Std 14.4607 +trainer/Q2 Predictions Max -3.7778 +trainer/Q2 Predictions Min -73.7971 +trainer/Q Targets Mean -51.8087 +trainer/Q Targets Std 14.5944 +trainer/Q Targets Max 0 +trainer/Q Targets Min -74.7918 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00236948 +trainer/policy/mean Std 0.597791 +trainer/policy/mean Max 0.99181 +trainer/policy/mean Min -0.975964 +trainer/policy/std Mean 0.4891 +trainer/policy/std Std 0.0254171 +trainer/policy/std Max 0.51606 +trainer/policy/std Min 0.453714 +trainer/Advantage Weights Mean 4.4609 +trainer/Advantage Weights Std 16.7232 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.54301e-10 +trainer/Advantage Score Mean -0.38196 +trainer/Advantage Score Std 0.490582 +trainer/Advantage Score Max 1.2431 +trainer/Advantage Score Min -2.08807 +trainer/V1 Predictions Mean -51.3575 +trainer/V1 Predictions Std 14.4352 +trainer/V1 Predictions Max -3.8571 +trainer/V1 Predictions Min -73.1527 +trainer/VF Loss 0.0557613 +expl/num steps total 16000 +expl/num paths total 16 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.018081 +expl/Actions Std 0.629457 +expl/Actions Max 2.02636 +expl/Actions Min -2.4235 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 16000 +eval/num paths total 16 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.129144 +eval/Actions Std 0.516562 +eval/Actions Max 0.775818 +eval/Actions Min -0.860287 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.83774e-06 +time/evaluation sampling (s) 2.99383 +time/exploration sampling (s) 3.9991 +time/logging (s) 0.00704199 +time/saving (s) 0.0118514 +time/training (s) 14.8987 +time/epoch (s) 21.9106 +time/total (s) 387.706 +Epoch -985 +------------------------------ ---------------- +2022-05-15 18:09:15.142182 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -984 finished +------------------------------ ---------------- +epoch -984 +replay_buffer/size 999047 +trainer/num train calls 17000 +trainer/QF1 Loss 1.23773 +trainer/QF2 Loss 1.20744 +trainer/Policy Loss 10.398 +trainer/Q1 Predictions Mean -51.6024 +trainer/Q1 Predictions Std 16.176 +trainer/Q1 Predictions Max -0.783328 +trainer/Q1 Predictions Min -78.2826 +trainer/Q2 Predictions Mean -51.4907 +trainer/Q2 Predictions Std 16.182 +trainer/Q2 Predictions Max -0.547502 +trainer/Q2 Predictions Min -78.3272 +trainer/Q Targets Mean -51.35 +trainer/Q Targets Std 16.3615 +trainer/Q Targets Max 0 +trainer/Q Targets Min -78.2661 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000478032 +trainer/policy/mean Std 0.604235 +trainer/policy/mean Max 0.99261 +trainer/policy/mean Min -0.97912 +trainer/policy/std Mean 0.498732 +trainer/policy/std Std 0.0308314 +trainer/policy/std Max 0.531742 +trainer/policy/std Min 0.455835 +trainer/Advantage Weights Mean 1.78814 +trainer/Advantage Weights Std 11.6335 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.33351e-12 +trainer/Advantage Score Mean -0.689 +trainer/Advantage Score Std 0.524789 +trainer/Advantage Score Max 0.486713 +trainer/Advantage Score Min -2.67836 +trainer/V1 Predictions Mean -50.9986 +trainer/V1 Predictions Std 16.3562 +trainer/V1 Predictions Max -0.412486 +trainer/V1 Predictions Min -78.5465 +trainer/VF Loss 0.0787478 +expl/num steps total 17000 +expl/num paths total 17 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0112626 +expl/Actions Std 0.650959 +expl/Actions Max 2.21671 +expl/Actions Min -2.37332 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 17000 +eval/num paths total 17 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.126976 +eval/Actions Std 0.419455 +eval/Actions Max 0.84672 +eval/Actions Min -0.888418 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.24779e-06 +time/evaluation sampling (s) 3.04165 +time/exploration sampling (s) 3.74148 +time/logging (s) 0.0113167 +time/saving (s) 0.0152888 +time/training (s) 14.9378 +time/epoch (s) 21.7475 +time/total (s) 409.459 +Epoch -984 +------------------------------ ---------------- +2022-05-15 18:09:36.893368 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -983 finished +------------------------------ ---------------- +epoch -983 +replay_buffer/size 999047 +trainer/num train calls 18000 +trainer/QF1 Loss 0.740961 +trainer/QF2 Loss 0.802045 +trainer/Policy Loss 34.9559 +trainer/Q1 Predictions Mean -53.0852 +trainer/Q1 Predictions Std 15.8284 +trainer/Q1 Predictions Max -0.634737 +trainer/Q1 Predictions Min -73.3917 +trainer/Q2 Predictions Mean -53.1196 +trainer/Q2 Predictions Std 15.8013 +trainer/Q2 Predictions Max -0.434447 +trainer/Q2 Predictions Min -72.7885 +trainer/Q Targets Mean -53.4105 +trainer/Q Targets Std 15.6269 +trainer/Q Targets Max 0 +trainer/Q Targets Min -73.3098 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0238181 +trainer/policy/mean Std 0.600755 +trainer/policy/mean Max 0.995296 +trainer/policy/mean Min -0.970247 +trainer/policy/std Mean 0.50484 +trainer/policy/std Std 0.0317363 +trainer/policy/std Max 0.539863 +trainer/policy/std Min 0.463088 +trainer/Advantage Weights Mean 7.2848 +trainer/Advantage Weights Std 23.4944 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.88774e-12 +trainer/Advantage Score Mean -0.369275 +trainer/Advantage Score Std 0.51779 +trainer/Advantage Score Max 1.23034 +trainer/Advantage Score Min -2.62732 +trainer/V1 Predictions Mean -53.0544 +trainer/V1 Predictions Std 15.591 +trainer/V1 Predictions Max 1.22889 +trainer/V1 Predictions Min -73.0508 +trainer/VF Loss 0.0707751 +expl/num steps total 18000 +expl/num paths total 18 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00222098 +expl/Actions Std 0.678609 +expl/Actions Max 2.55826 +expl/Actions Min -2.5061 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 18000 +eval/num paths total 18 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0285843 +eval/Actions Std 0.3659 +eval/Actions Max 0.757115 +eval/Actions Min -0.822245 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.26987e-06 +time/evaluation sampling (s) 2.94294 +time/exploration sampling (s) 3.65192 +time/logging (s) 0.00788714 +time/saving (s) 0.0109536 +time/training (s) 15.1269 +time/epoch (s) 21.7407 +time/total (s) 431.206 +Epoch -983 +------------------------------ ---------------- +2022-05-15 18:09:58.829557 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -982 finished +------------------------------ ---------------- +epoch -982 +replay_buffer/size 999047 +trainer/num train calls 19000 +trainer/QF1 Loss 0.770663 +trainer/QF2 Loss 0.761906 +trainer/Policy Loss 24.1949 +trainer/Q1 Predictions Mean -56.0489 +trainer/Q1 Predictions Std 16.3803 +trainer/Q1 Predictions Max -3.14289 +trainer/Q1 Predictions Min -73.7719 +trainer/Q2 Predictions Mean -56.1282 +trainer/Q2 Predictions Std 16.3595 +trainer/Q2 Predictions Max -2.45609 +trainer/Q2 Predictions Min -73.836 +trainer/Q Targets Mean -56.2151 +trainer/Q Targets Std 16.326 +trainer/Q Targets Max 0 +trainer/Q Targets Min -73.7678 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00836209 +trainer/policy/mean Std 0.609163 +trainer/policy/mean Max 0.984605 +trainer/policy/mean Min -0.982471 +trainer/policy/std Mean 0.507243 +trainer/policy/std Std 0.0339662 +trainer/policy/std Max 0.544383 +trainer/policy/std Min 0.458662 +trainer/Advantage Weights Mean 4.64509 +trainer/Advantage Weights Std 17.2446 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.25789e-08 +trainer/Advantage Score Mean -0.321713 +trainer/Advantage Score Std 0.447443 +trainer/Advantage Score Max 1.18876 +trainer/Advantage Score Min -1.81912 +trainer/V1 Predictions Mean -55.8047 +trainer/V1 Predictions Std 16.3985 +trainer/V1 Predictions Max -2.60684 +trainer/V1 Predictions Min -73.515 +trainer/VF Loss 0.0456925 +expl/num steps total 19000 +expl/num paths total 19 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0279587 +expl/Actions Std 0.67137 +expl/Actions Max 2.29182 +expl/Actions Min -2.47269 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 19000 +eval/num paths total 19 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.152109 +eval/Actions Std 0.487188 +eval/Actions Max 0.767964 +eval/Actions Min -0.864541 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.74833e-06 +time/evaluation sampling (s) 3.01859 +time/exploration sampling (s) 4.01879 +time/logging (s) 0.00678551 +time/saving (s) 0.00942925 +time/training (s) 14.8759 +time/epoch (s) 21.9295 +time/total (s) 453.14 +Epoch -982 +------------------------------ ---------------- +2022-05-15 18:10:20.668946 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -981 finished +------------------------------ ---------------- +epoch -981 +replay_buffer/size 999047 +trainer/num train calls 20000 +trainer/QF1 Loss 0.998929 +trainer/QF2 Loss 1.03659 +trainer/Policy Loss 73.2367 +trainer/Q1 Predictions Mean -57.8661 +trainer/Q1 Predictions Std 14.3049 +trainer/Q1 Predictions Max -2.3159 +trainer/Q1 Predictions Min -74.3814 +trainer/Q2 Predictions Mean -57.8388 +trainer/Q2 Predictions Std 14.3265 +trainer/Q2 Predictions Max -2.39545 +trainer/Q2 Predictions Min -74.3852 +trainer/Q Targets Mean -58.3135 +trainer/Q Targets Std 13.8892 +trainer/Q Targets Max 0 +trainer/Q Targets Min -74.3519 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0158522 +trainer/policy/mean Std 0.620265 +trainer/policy/mean Max 0.984414 +trainer/policy/mean Min -0.992898 +trainer/policy/std Mean 0.508164 +trainer/policy/std Std 0.0348333 +trainer/policy/std Max 0.543266 +trainer/policy/std Min 0.459763 +trainer/Advantage Weights Mean 11.7425 +trainer/Advantage Weights Std 28.6699 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.82843e-09 +trainer/Advantage Score Mean -0.123913 +trainer/Advantage Score Std 0.454657 +trainer/Advantage Score Max 1.31464 +trainer/Advantage Score Min -1.86655 +trainer/V1 Predictions Mean -57.924 +trainer/V1 Predictions Std 13.9876 +trainer/V1 Predictions Max -2.98011 +trainer/V1 Predictions Min -73.9788 +trainer/VF Loss 0.0712412 +expl/num steps total 20000 +expl/num paths total 20 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0130611 +expl/Actions Std 0.674562 +expl/Actions Max 2.47596 +expl/Actions Min -2.55223 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 20000 +eval/num paths total 20 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0539524 +eval/Actions Std 0.370436 +eval/Actions Max 0.837407 +eval/Actions Min -0.7938 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.59141e-06 +time/evaluation sampling (s) 3.04394 +time/exploration sampling (s) 3.92764 +time/logging (s) 0.00730945 +time/saving (s) 0.0103844 +time/training (s) 14.8459 +time/epoch (s) 21.8351 +time/total (s) 474.979 +Epoch -981 +------------------------------ ---------------- +2022-05-15 18:10:42.299260 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -980 finished +------------------------------ ---------------- +epoch -980 +replay_buffer/size 999047 +trainer/num train calls 21000 +trainer/QF1 Loss 0.904218 +trainer/QF2 Loss 0.891258 +trainer/Policy Loss 25.6498 +trainer/Q1 Predictions Mean -59.0856 +trainer/Q1 Predictions Std 15.4403 +trainer/Q1 Predictions Max -3.55147 +trainer/Q1 Predictions Min -77.775 +trainer/Q2 Predictions Mean -59.1585 +trainer/Q2 Predictions Std 15.4722 +trainer/Q2 Predictions Max -3.66544 +trainer/Q2 Predictions Min -78.173 +trainer/Q Targets Mean -59.2054 +trainer/Q Targets Std 15.7893 +trainer/Q Targets Max 0 +trainer/Q Targets Min -78.0024 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0147875 +trainer/policy/mean Std 0.606428 +trainer/policy/mean Max 0.983756 +trainer/policy/mean Min -0.98021 +trainer/policy/std Mean 0.508133 +trainer/policy/std Std 0.0345819 +trainer/policy/std Max 0.541677 +trainer/policy/std Min 0.462611 +trainer/Advantage Weights Mean 4.01785 +trainer/Advantage Weights Std 17.2821 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.50542e-13 +trainer/Advantage Score Mean -0.405572 +trainer/Advantage Score Std 0.470977 +trainer/Advantage Score Max 0.820141 +trainer/Advantage Score Min -2.84283 +trainer/V1 Predictions Mean -58.8435 +trainer/V1 Predictions Std 15.6688 +trainer/V1 Predictions Max -3.50179 +trainer/V1 Predictions Min -77.8563 +trainer/VF Loss 0.0505202 +expl/num steps total 21000 +expl/num paths total 21 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0367233 +expl/Actions Std 0.665836 +expl/Actions Max 2.39935 +expl/Actions Min -2.42562 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 21000 +eval/num paths total 21 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.110189 +eval/Actions Std 0.372363 +eval/Actions Max 0.8429 +eval/Actions Min -0.862578 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.75485e-06 +time/evaluation sampling (s) 2.89184 +time/exploration sampling (s) 3.78758 +time/logging (s) 0.00874549 +time/saving (s) 0.0111394 +time/training (s) 14.9272 +time/epoch (s) 21.6265 +time/total (s) 496.61 +Epoch -980 +------------------------------ ---------------- +2022-05-15 18:11:04.509121 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -979 finished +------------------------------ ---------------- +epoch -979 +replay_buffer/size 999047 +trainer/num train calls 22000 +trainer/QF1 Loss 1.25706 +trainer/QF2 Loss 1.27552 +trainer/Policy Loss 58.1352 +trainer/Q1 Predictions Mean -58.0199 +trainer/Q1 Predictions Std 18.227 +trainer/Q1 Predictions Max -3.1531 +trainer/Q1 Predictions Min -78.7913 +trainer/Q2 Predictions Mean -57.9204 +trainer/Q2 Predictions Std 18.2258 +trainer/Q2 Predictions Max -2.00201 +trainer/Q2 Predictions Min -78.7994 +trainer/Q Targets Mean -58.2457 +trainer/Q Targets Std 18.2299 +trainer/Q Targets Max 0 +trainer/Q Targets Min -78.4681 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0287465 +trainer/policy/mean Std 0.624053 +trainer/policy/mean Max 0.992988 +trainer/policy/mean Min -0.983208 +trainer/policy/std Mean 0.504612 +trainer/policy/std Std 0.0346318 +trainer/policy/std Max 0.540206 +trainer/policy/std Min 0.458196 +trainer/Advantage Weights Mean 9.1473 +trainer/Advantage Weights Std 25.1344 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.32549e-09 +trainer/Advantage Score Mean -0.154735 +trainer/Advantage Score Std 0.41845 +trainer/Advantage Score Max 1.08604 +trainer/Advantage Score Min -2.04415 +trainer/V1 Predictions Mean -57.8457 +trainer/V1 Predictions Std 18.3489 +trainer/V1 Predictions Max -0.812708 +trainer/V1 Predictions Min -78.2451 +trainer/VF Loss 0.0536144 +expl/num steps total 22000 +expl/num paths total 22 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.02649 +expl/Actions Std 0.674105 +expl/Actions Max 2.38151 +expl/Actions Min -2.48072 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 22000 +eval/num paths total 22 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.114207 +eval/Actions Std 0.39475 +eval/Actions Max 0.804091 +eval/Actions Min -0.831476 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.78722e-06 +time/evaluation sampling (s) 3.00426 +time/exploration sampling (s) 3.83946 +time/logging (s) 0.0110697 +time/saving (s) 0.0156228 +time/training (s) 15.3366 +time/epoch (s) 22.2071 +time/total (s) 518.821 +Epoch -979 +------------------------------ ---------------- +2022-05-15 18:11:26.363036 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -978 finished +------------------------------ ---------------- +epoch -978 +replay_buffer/size 999047 +trainer/num train calls 23000 +trainer/QF1 Loss 1.02799 +trainer/QF2 Loss 1.15337 +trainer/Policy Loss 3.78835 +trainer/Q1 Predictions Mean -59.6245 +trainer/Q1 Predictions Std 17.4644 +trainer/Q1 Predictions Max -0.852637 +trainer/Q1 Predictions Min -79.9435 +trainer/Q2 Predictions Mean -59.6666 +trainer/Q2 Predictions Std 17.4592 +trainer/Q2 Predictions Max -1.1607 +trainer/Q2 Predictions Min -79.3374 +trainer/Q Targets Mean -60.0324 +trainer/Q Targets Std 17.3493 +trainer/Q Targets Max 0 +trainer/Q Targets Min -79.6404 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00514894 +trainer/policy/mean Std 0.621842 +trainer/policy/mean Max 0.996408 +trainer/policy/mean Min -0.988764 +trainer/policy/std Mean 0.503915 +trainer/policy/std Std 0.0336065 +trainer/policy/std Max 0.540362 +trainer/policy/std Min 0.460822 +trainer/Advantage Weights Mean 0.861826 +trainer/Advantage Weights Std 6.30012 +trainer/Advantage Weights Max 63.4325 +trainer/Advantage Weights Min 3.98675e-10 +trainer/Advantage Score Mean -0.645692 +trainer/Advantage Score Std 0.422511 +trainer/Advantage Score Max 0.414998 +trainer/Advantage Score Min -2.16429 +trainer/V1 Predictions Mean -59.7054 +trainer/V1 Predictions Std 17.3622 +trainer/V1 Predictions Max -0.528242 +trainer/V1 Predictions Min -80.166 +trainer/VF Loss 0.0615486 +expl/num steps total 23000 +expl/num paths total 23 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0313136 +expl/Actions Std 0.688443 +expl/Actions Max 2.40146 +expl/Actions Min -2.34153 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 23000 +eval/num paths total 23 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.116478 +eval/Actions Std 0.392099 +eval/Actions Max 0.809134 +eval/Actions Min -0.852876 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.08687e-06 +time/evaluation sampling (s) 2.84034 +time/exploration sampling (s) 4.02869 +time/logging (s) 0.0104646 +time/saving (s) 0.0147386 +time/training (s) 14.9517 +time/epoch (s) 21.846 +time/total (s) 540.673 +Epoch -978 +------------------------------ ---------------- +2022-05-15 18:11:48.370905 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -977 finished +------------------------------ ---------------- +epoch -977 +replay_buffer/size 999047 +trainer/num train calls 24000 +trainer/QF1 Loss 1.50825 +trainer/QF2 Loss 1.37069 +trainer/Policy Loss 16.261 +trainer/Q1 Predictions Mean -61.2993 +trainer/Q1 Predictions Std 15.5849 +trainer/Q1 Predictions Max -6.70792 +trainer/Q1 Predictions Min -79.0442 +trainer/Q2 Predictions Mean -61.3027 +trainer/Q2 Predictions Std 15.5978 +trainer/Q2 Predictions Max -6.55128 +trainer/Q2 Predictions Min -79.7784 +trainer/Q Targets Mean -61.3754 +trainer/Q Targets Std 15.9011 +trainer/Q Targets Max 0 +trainer/Q Targets Min -78.94 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0381038 +trainer/policy/mean Std 0.641802 +trainer/policy/mean Max 0.995706 +trainer/policy/mean Min -0.988712 +trainer/policy/std Mean 0.502542 +trainer/policy/std Std 0.0338537 +trainer/policy/std Max 0.543215 +trainer/policy/std Min 0.458554 +trainer/Advantage Weights Mean 3.03966 +trainer/Advantage Weights Std 14.9568 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.72119e-12 +trainer/Advantage Score Mean -0.546396 +trainer/Advantage Score Std 0.511311 +trainer/Advantage Score Max 1.00586 +trainer/Advantage Score Min -2.6317 +trainer/V1 Predictions Mean -61.0628 +trainer/V1 Predictions Std 15.6871 +trainer/V1 Predictions Max -6.81919 +trainer/V1 Predictions Min -79.0171 +trainer/VF Loss 0.0662641 +expl/num steps total 24000 +expl/num paths total 24 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0154621 +expl/Actions Std 0.699856 +expl/Actions Max 2.49483 +expl/Actions Min -2.274 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 24000 +eval/num paths total 24 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.160385 +eval/Actions Std 0.539109 +eval/Actions Max 0.767913 +eval/Actions Min -0.78573 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.56998e-06 +time/evaluation sampling (s) 3.02516 +time/exploration sampling (s) 3.90168 +time/logging (s) 0.0116081 +time/saving (s) 0.0164481 +time/training (s) 15.0494 +time/epoch (s) 22.0043 +time/total (s) 562.681 +Epoch -977 +------------------------------ ---------------- +2022-05-15 18:12:10.898337 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -976 finished +------------------------------ ---------------- +epoch -976 +replay_buffer/size 999047 +trainer/num train calls 25000 +trainer/QF1 Loss 0.948641 +trainer/QF2 Loss 0.952568 +trainer/Policy Loss 37.9632 +trainer/Q1 Predictions Mean -62.9776 +trainer/Q1 Predictions Std 16.6231 +trainer/Q1 Predictions Max -4.48559 +trainer/Q1 Predictions Min -79.9957 +trainer/Q2 Predictions Mean -62.9185 +trainer/Q2 Predictions Std 16.6792 +trainer/Q2 Predictions Max -4.21058 +trainer/Q2 Predictions Min -80.2448 +trainer/Q Targets Mean -62.7038 +trainer/Q Targets Std 17.0255 +trainer/Q Targets Max 0 +trainer/Q Targets Min -80.2373 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0210781 +trainer/policy/mean Std 0.632891 +trainer/policy/mean Max 0.984627 +trainer/policy/mean Min -0.987071 +trainer/policy/std Mean 0.500015 +trainer/policy/std Std 0.0328427 +trainer/policy/std Max 0.538482 +trainer/policy/std Min 0.456075 +trainer/Advantage Weights Mean 6.49259 +trainer/Advantage Weights Std 20.4414 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.785e-12 +trainer/Advantage Score Mean -0.268377 +trainer/Advantage Score Std 0.48295 +trainer/Advantage Score Max 0.754568 +trainer/Advantage Score Min -2.53502 +trainer/V1 Predictions Mean -62.4345 +trainer/V1 Predictions Std 17.0015 +trainer/V1 Predictions Max -4.46069 +trainer/V1 Predictions Min -80.0143 +trainer/VF Loss 0.0489877 +expl/num steps total 25000 +expl/num paths total 25 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00380213 +expl/Actions Std 0.718224 +expl/Actions Max 2.20257 +expl/Actions Min -2.35816 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 25000 +eval/num paths total 25 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0139025 +eval/Actions Std 0.496707 +eval/Actions Max 0.979686 +eval/Actions Min -0.984136 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.43425e-06 +time/evaluation sampling (s) 3.2327 +time/exploration sampling (s) 3.93699 +time/logging (s) 0.00905249 +time/saving (s) 0.010779 +time/training (s) 15.3298 +time/epoch (s) 22.5193 +time/total (s) 585.205 +Epoch -976 +------------------------------ ---------------- +2022-05-15 18:12:33.435907 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -975 finished +------------------------------ ---------------- +epoch -975 +replay_buffer/size 999047 +trainer/num train calls 26000 +trainer/QF1 Loss 0.835642 +trainer/QF2 Loss 0.945394 +trainer/Policy Loss 31.4605 +trainer/Q1 Predictions Mean -60.949 +trainer/Q1 Predictions Std 18.7648 +trainer/Q1 Predictions Max -5.10799 +trainer/Q1 Predictions Min -82.2972 +trainer/Q2 Predictions Mean -60.9127 +trainer/Q2 Predictions Std 18.7731 +trainer/Q2 Predictions Max -5.14297 +trainer/Q2 Predictions Min -82.1637 +trainer/Q Targets Mean -61.0265 +trainer/Q Targets Std 18.6835 +trainer/Q Targets Max 0 +trainer/Q Targets Min -82.153 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00999197 +trainer/policy/mean Std 0.631204 +trainer/policy/mean Max 0.98488 +trainer/policy/mean Min -0.986805 +trainer/policy/std Mean 0.49623 +trainer/policy/std Std 0.0321479 +trainer/policy/std Max 0.534007 +trainer/policy/std Min 0.454432 +trainer/Advantage Weights Mean 5.40022 +trainer/Advantage Weights Std 17.6086 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.40021e-09 +trainer/Advantage Score Mean -0.252902 +trainer/Advantage Score Std 0.43306 +trainer/Advantage Score Max 1.27954 +trainer/Advantage Score Min -1.8595 +trainer/V1 Predictions Mean -60.6603 +trainer/V1 Predictions Std 18.7449 +trainer/V1 Predictions Max -4.7867 +trainer/V1 Predictions Min -82.29 +trainer/VF Loss 0.0455713 +expl/num steps total 26000 +expl/num paths total 26 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00162783 +expl/Actions Std 0.722105 +expl/Actions Max 2.44322 +expl/Actions Min -2.49542 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 26000 +eval/num paths total 26 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0156012 +eval/Actions Std 0.589461 +eval/Actions Max 0.994023 +eval/Actions Min -0.996136 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.73902e-06 +time/evaluation sampling (s) 3.12328 +time/exploration sampling (s) 3.83939 +time/logging (s) 0.0100008 +time/saving (s) 0.013267 +time/training (s) 15.547 +time/epoch (s) 22.5329 +time/total (s) 607.742 +Epoch -975 +------------------------------ ---------------- +2022-05-15 18:12:56.077565 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -974 finished +------------------------------ ---------------- +epoch -974 +replay_buffer/size 999047 +trainer/num train calls 27000 +trainer/QF1 Loss 1.22063 +trainer/QF2 Loss 1.18664 +trainer/Policy Loss 39.1754 +trainer/Q1 Predictions Mean -63.1063 +trainer/Q1 Predictions Std 16.8252 +trainer/Q1 Predictions Max -3.99521 +trainer/Q1 Predictions Min -83.6614 +trainer/Q2 Predictions Mean -63.0508 +trainer/Q2 Predictions Std 16.7547 +trainer/Q2 Predictions Max -4.73607 +trainer/Q2 Predictions Min -83.9421 +trainer/Q Targets Mean -63.0615 +trainer/Q Targets Std 16.651 +trainer/Q Targets Max 0 +trainer/Q Targets Min -83.7139 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0166503 +trainer/policy/mean Std 0.644472 +trainer/policy/mean Max 0.991464 +trainer/policy/mean Min -0.995058 +trainer/policy/std Mean 0.494682 +trainer/policy/std Std 0.0331712 +trainer/policy/std Max 0.531974 +trainer/policy/std Min 0.449756 +trainer/Advantage Weights Mean 7.23209 +trainer/Advantage Weights Std 22.352 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.29806e-08 +trainer/Advantage Score Mean -0.297324 +trainer/Advantage Score Std 0.466749 +trainer/Advantage Score Max 0.847916 +trainer/Advantage Score Min -1.75886 +trainer/V1 Predictions Mean -62.8337 +trainer/V1 Predictions Std 16.7465 +trainer/V1 Predictions Max -3.28964 +trainer/V1 Predictions Min -83.0742 +trainer/VF Loss 0.0527657 +expl/num steps total 27000 +expl/num paths total 27 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0133068 +expl/Actions Std 0.730995 +expl/Actions Max 2.4592 +expl/Actions Min -2.77731 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 27000 +eval/num paths total 27 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.159164 +eval/Actions Std 0.469454 +eval/Actions Max 0.970396 +eval/Actions Min -0.968772 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.95276e-06 +time/evaluation sampling (s) 3.70403 +time/exploration sampling (s) 4.14394 +time/logging (s) 0.00687119 +time/saving (s) 0.00943763 +time/training (s) 14.7687 +time/epoch (s) 22.633 +time/total (s) 630.38 +Epoch -974 +------------------------------ ---------------- +2022-05-15 18:13:17.545091 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -973 finished +------------------------------ ---------------- +epoch -973 +replay_buffer/size 999047 +trainer/num train calls 28000 +trainer/QF1 Loss 1.77679 +trainer/QF2 Loss 1.85995 +trainer/Policy Loss 77.5159 +trainer/Q1 Predictions Mean -63.4025 +trainer/Q1 Predictions Std 17.1502 +trainer/Q1 Predictions Max -0.524968 +trainer/Q1 Predictions Min -82.3642 +trainer/Q2 Predictions Mean -63.3584 +trainer/Q2 Predictions Std 17.1508 +trainer/Q2 Predictions Max -0.563121 +trainer/Q2 Predictions Min -82.2108 +trainer/Q Targets Mean -64.2311 +trainer/Q Targets Std 16.6488 +trainer/Q Targets Max 0 +trainer/Q Targets Min -82.9755 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.019482 +trainer/policy/mean Std 0.651731 +trainer/policy/mean Max 0.996672 +trainer/policy/mean Min -0.989779 +trainer/policy/std Mean 0.493657 +trainer/policy/std Std 0.0330789 +trainer/policy/std Max 0.529814 +trainer/policy/std Min 0.451578 +trainer/Advantage Weights Mean 15.273 +trainer/Advantage Weights Std 32.3961 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.84577e-08 +trainer/Advantage Score Mean -0.236984 +trainer/Advantage Score Std 0.556603 +trainer/Advantage Score Max 1.55508 +trainer/Advantage Score Min -1.68426 +trainer/V1 Predictions Mean -63.8881 +trainer/V1 Predictions Std 16.7996 +trainer/V1 Predictions Max -1.36698 +trainer/V1 Predictions Min -83.4243 +trainer/VF Loss 0.0961261 +expl/num steps total 28000 +expl/num paths total 28 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00395575 +expl/Actions Std 0.710749 +expl/Actions Max 2.39843 +expl/Actions Min -2.57725 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 28000 +eval/num paths total 28 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.114531 +eval/Actions Std 0.425328 +eval/Actions Max 0.868529 +eval/Actions Min -0.893548 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.11365e-06 +time/evaluation sampling (s) 2.94274 +time/exploration sampling (s) 3.77851 +time/logging (s) 0.0109653 +time/saving (s) 0.0147918 +time/training (s) 14.7196 +time/epoch (s) 21.4666 +time/total (s) 651.851 +Epoch -973 +------------------------------ ---------------- +2022-05-15 18:13:39.730805 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -972 finished +------------------------------ ---------------- +epoch -972 +replay_buffer/size 999047 +trainer/num train calls 29000 +trainer/QF1 Loss 0.629088 +trainer/QF2 Loss 0.604295 +trainer/Policy Loss 16.7475 +trainer/Q1 Predictions Mean -64.3059 +trainer/Q1 Predictions Std 17.5109 +trainer/Q1 Predictions Max -4.26699 +trainer/Q1 Predictions Min -86.9368 +trainer/Q2 Predictions Mean -64.375 +trainer/Q2 Predictions Std 17.5175 +trainer/Q2 Predictions Max -4.6399 +trainer/Q2 Predictions Min -86.2515 +trainer/Q Targets Mean -64.1428 +trainer/Q Targets Std 17.5967 +trainer/Q Targets Max -5.06065 +trainer/Q Targets Min -86.6893 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0151875 +trainer/policy/mean Std 0.63168 +trainer/policy/mean Max 0.991075 +trainer/policy/mean Min -0.989152 +trainer/policy/std Mean 0.492961 +trainer/policy/std Std 0.0316705 +trainer/policy/std Max 0.528997 +trainer/policy/std Min 0.453568 +trainer/Advantage Weights Mean 3.44285 +trainer/Advantage Weights Std 14.6234 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.30652e-14 +trainer/Advantage Score Mean -0.374298 +trainer/Advantage Score Std 0.50308 +trainer/Advantage Score Max 0.822875 +trainer/Advantage Score Min -3.14005 +trainer/V1 Predictions Mean -63.796 +trainer/V1 Predictions Std 17.7734 +trainer/V1 Predictions Max -4.07571 +trainer/V1 Predictions Min -86.1557 +trainer/VF Loss 0.0489949 +expl/num steps total 29000 +expl/num paths total 29 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00461022 +expl/Actions Std 0.702091 +expl/Actions Max 2.71369 +expl/Actions Min -2.38338 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 29000 +eval/num paths total 29 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0508608 +eval/Actions Std 0.543824 +eval/Actions Max 0.97718 +eval/Actions Min -0.983525 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.23732e-06 +time/evaluation sampling (s) 3.14072 +time/exploration sampling (s) 3.75681 +time/logging (s) 0.00934359 +time/saving (s) 0.0129435 +time/training (s) 15.2554 +time/epoch (s) 22.1752 +time/total (s) 674.033 +Epoch -972 +------------------------------ ---------------- +2022-05-15 18:14:01.903838 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -971 finished +------------------------------ ---------------- +epoch -971 +replay_buffer/size 999047 +trainer/num train calls 30000 +trainer/QF1 Loss 0.647579 +trainer/QF2 Loss 0.82534 +trainer/Policy Loss 12.7348 +trainer/Q1 Predictions Mean -65.0605 +trainer/Q1 Predictions Std 17.5373 +trainer/Q1 Predictions Max -5.91693 +trainer/Q1 Predictions Min -83.6557 +trainer/Q2 Predictions Mean -65.2151 +trainer/Q2 Predictions Std 17.5791 +trainer/Q2 Predictions Max -5.65083 +trainer/Q2 Predictions Min -84.0275 +trainer/Q Targets Mean -64.613 +trainer/Q Targets Std 17.4486 +trainer/Q Targets Max -6.58636 +trainer/Q Targets Min -83.5455 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00585176 +trainer/policy/mean Std 0.653159 +trainer/policy/mean Max 0.998029 +trainer/policy/mean Min -0.993164 +trainer/policy/std Mean 0.491247 +trainer/policy/std Std 0.0313636 +trainer/policy/std Max 0.525055 +trainer/policy/std Min 0.450754 +trainer/Advantage Weights Mean 2.29624 +trainer/Advantage Weights Std 12.9323 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.95229e-13 +trainer/Advantage Score Mean -0.504675 +trainer/Advantage Score Std 0.498247 +trainer/Advantage Score Max 0.919632 +trainer/Advantage Score Min -2.76358 +trainer/V1 Predictions Mean -64.3077 +trainer/V1 Predictions Std 17.6246 +trainer/V1 Predictions Max -5.74091 +trainer/V1 Predictions Min -83.7084 +trainer/VF Loss 0.0596109 +expl/num steps total 30000 +expl/num paths total 30 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00764978 +expl/Actions Std 0.761435 +expl/Actions Max 2.5258 +expl/Actions Min -2.36504 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 30000 +eval/num paths total 30 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.222128 +eval/Actions Std 0.610682 +eval/Actions Max 0.979054 +eval/Actions Min -0.972143 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.92296e-06 +time/evaluation sampling (s) 3.30462 +time/exploration sampling (s) 3.66879 +time/logging (s) 0.00782203 +time/saving (s) 0.0139975 +time/training (s) 15.1713 +time/epoch (s) 22.1666 +time/total (s) 696.204 +Epoch -971 +------------------------------ ---------------- +2022-05-15 18:14:23.916086 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -970 finished +------------------------------ ---------------- +epoch -970 +replay_buffer/size 999047 +trainer/num train calls 31000 +trainer/QF1 Loss 0.572955 +trainer/QF2 Loss 0.612288 +trainer/Policy Loss 25.1994 +trainer/Q1 Predictions Mean -64.4817 +trainer/Q1 Predictions Std 16.571 +trainer/Q1 Predictions Max -3.70625 +trainer/Q1 Predictions Min -84.3871 +trainer/Q2 Predictions Mean -64.4998 +trainer/Q2 Predictions Std 16.5706 +trainer/Q2 Predictions Max -3.64097 +trainer/Q2 Predictions Min -84.2046 +trainer/Q Targets Mean -64.1689 +trainer/Q Targets Std 16.6518 +trainer/Q Targets Max -3.85515 +trainer/Q Targets Min -84.7863 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00680804 +trainer/policy/mean Std 0.644988 +trainer/policy/mean Max 0.993678 +trainer/policy/mean Min -0.987832 +trainer/policy/std Mean 0.489687 +trainer/policy/std Std 0.0304852 +trainer/policy/std Max 0.522331 +trainer/policy/std Min 0.44482 +trainer/Advantage Weights Mean 3.44618 +trainer/Advantage Weights Std 15.9729 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.03039e-11 +trainer/Advantage Score Mean -0.483534 +trainer/Advantage Score Std 0.489964 +trainer/Advantage Score Max 0.778259 +trainer/Advantage Score Min -2.46202 +trainer/V1 Predictions Mean -63.8396 +trainer/V1 Predictions Std 16.7781 +trainer/V1 Predictions Max -2.99083 +trainer/V1 Predictions Min -83.9866 +trainer/VF Loss 0.0570464 +expl/num steps total 31000 +expl/num paths total 31 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0109 +expl/Actions Std 0.761004 +expl/Actions Max 2.48835 +expl/Actions Min -2.8068 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 31000 +eval/num paths total 31 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0594286 +eval/Actions Std 0.577697 +eval/Actions Max 0.985313 +eval/Actions Min -0.990818 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.50506e-06 +time/evaluation sampling (s) 3.09137 +time/exploration sampling (s) 3.74434 +time/logging (s) 0.00881935 +time/saving (s) 0.013514 +time/training (s) 15.1478 +time/epoch (s) 22.0059 +time/total (s) 718.216 +Epoch -970 +------------------------------ ---------------- +2022-05-15 18:14:46.048722 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -969 finished +------------------------------ ---------------- +epoch -969 +replay_buffer/size 999047 +trainer/num train calls 32000 +trainer/QF1 Loss 1.07133 +trainer/QF2 Loss 1.18626 +trainer/Policy Loss 15.6737 +trainer/Q1 Predictions Mean -65.7482 +trainer/Q1 Predictions Std 16.8489 +trainer/Q1 Predictions Max -2.78805 +trainer/Q1 Predictions Min -84.2026 +trainer/Q2 Predictions Mean -65.7394 +trainer/Q2 Predictions Std 16.8644 +trainer/Q2 Predictions Max -3.71325 +trainer/Q2 Predictions Min -83.9767 +trainer/Q Targets Mean -65.7519 +trainer/Q Targets Std 17.1096 +trainer/Q Targets Max 0 +trainer/Q Targets Min -83.6779 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0083917 +trainer/policy/mean Std 0.620102 +trainer/policy/mean Max 0.98524 +trainer/policy/mean Min -0.994276 +trainer/policy/std Mean 0.48994 +trainer/policy/std Std 0.0303839 +trainer/policy/std Max 0.524331 +trainer/policy/std Min 0.445756 +trainer/Advantage Weights Mean 3.10937 +trainer/Advantage Weights Std 12.2338 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.5473e-10 +trainer/Advantage Score Mean -0.333799 +trainer/Advantage Score Std 0.448189 +trainer/Advantage Score Max 0.749237 +trainer/Advantage Score Min -2.20908 +trainer/V1 Predictions Mean -65.3894 +trainer/V1 Predictions Std 17.118 +trainer/V1 Predictions Max -3.14028 +trainer/V1 Predictions Min -83.6132 +trainer/VF Loss 0.0412067 +expl/num steps total 32000 +expl/num paths total 32 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.000227705 +expl/Actions Std 0.744414 +expl/Actions Max 2.38316 +expl/Actions Min -2.50829 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 32000 +eval/num paths total 32 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.141991 +eval/Actions Std 0.615396 +eval/Actions Max 0.984797 +eval/Actions Min -0.980788 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.65334e-06 +time/evaluation sampling (s) 3.39679 +time/exploration sampling (s) 3.93652 +time/logging (s) 0.00816589 +time/saving (s) 0.0115696 +time/training (s) 14.7731 +time/epoch (s) 22.1262 +time/total (s) 740.347 +Epoch -969 +------------------------------ ---------------- +2022-05-15 18:15:08.349512 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -968 finished +------------------------------ ---------------- +epoch -968 +replay_buffer/size 999047 +trainer/num train calls 33000 +trainer/QF1 Loss 1.35694 +trainer/QF2 Loss 1.45925 +trainer/Policy Loss 72.7772 +trainer/Q1 Predictions Mean -64.958 +trainer/Q1 Predictions Std 18.6848 +trainer/Q1 Predictions Max -2.31729 +trainer/Q1 Predictions Min -85.7989 +trainer/Q2 Predictions Mean -64.9742 +trainer/Q2 Predictions Std 18.697 +trainer/Q2 Predictions Max -2.74798 +trainer/Q2 Predictions Min -85.9232 +trainer/Q Targets Mean -65.5379 +trainer/Q Targets Std 18.4104 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.2602 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00935922 +trainer/policy/mean Std 0.654298 +trainer/policy/mean Max 0.992553 +trainer/policy/mean Min -0.985493 +trainer/policy/std Mean 0.486942 +trainer/policy/std Std 0.0301525 +trainer/policy/std Max 0.520418 +trainer/policy/std Min 0.441881 +trainer/Advantage Weights Mean 13.3501 +trainer/Advantage Weights Std 29.8233 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.31619e-07 +trainer/Advantage Score Mean -0.119096 +trainer/Advantage Score Std 0.460327 +trainer/Advantage Score Max 1.43549 +trainer/Advantage Score Min -1.58434 +trainer/V1 Predictions Mean -65.2228 +trainer/V1 Predictions Std 18.4634 +trainer/V1 Predictions Max -3.77518 +trainer/V1 Predictions Min -84.7729 +trainer/VF Loss 0.0775553 +expl/num steps total 33000 +expl/num paths total 33 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00497235 +expl/Actions Std 0.692814 +expl/Actions Max 2.19739 +expl/Actions Min -2.54143 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 33000 +eval/num paths total 33 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0243703 +eval/Actions Std 0.623744 +eval/Actions Max 0.988841 +eval/Actions Min -0.992466 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.73483e-06 +time/evaluation sampling (s) 3.19614 +time/exploration sampling (s) 3.92214 +time/logging (s) 0.00681861 +time/saving (s) 0.00957815 +time/training (s) 15.1584 +time/epoch (s) 22.2931 +time/total (s) 762.645 +Epoch -968 +------------------------------ ---------------- +2022-05-15 18:15:29.601114 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -967 finished +------------------------------ ---------------- +epoch -967 +replay_buffer/size 999047 +trainer/num train calls 34000 +trainer/QF1 Loss 0.9723 +trainer/QF2 Loss 0.954175 +trainer/Policy Loss 21.8427 +trainer/Q1 Predictions Mean -65.6874 +trainer/Q1 Predictions Std 17.9788 +trainer/Q1 Predictions Max -4.08358 +trainer/Q1 Predictions Min -85.6944 +trainer/Q2 Predictions Mean -65.7416 +trainer/Q2 Predictions Std 17.9116 +trainer/Q2 Predictions Max -4.52138 +trainer/Q2 Predictions Min -85.502 +trainer/Q Targets Mean -65.5614 +trainer/Q Targets Std 18.0257 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.3075 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00324442 +trainer/policy/mean Std 0.645041 +trainer/policy/mean Max 0.991153 +trainer/policy/mean Min -0.992133 +trainer/policy/std Mean 0.485213 +trainer/policy/std Std 0.0289082 +trainer/policy/std Max 0.518857 +trainer/policy/std Min 0.441688 +trainer/Advantage Weights Mean 4.01407 +trainer/Advantage Weights Std 17.5293 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.5294e-17 +trainer/Advantage Score Mean -0.408489 +trainer/Advantage Score Std 0.547998 +trainer/Advantage Score Max 1.31278 +trainer/Advantage Score Min -3.70004 +trainer/V1 Predictions Mean -65.2562 +trainer/V1 Predictions Std 18.0957 +trainer/V1 Predictions Max -3.54117 +trainer/V1 Predictions Min -85.0014 +trainer/VF Loss 0.0639649 +expl/num steps total 34000 +expl/num paths total 34 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00126444 +expl/Actions Std 0.702444 +expl/Actions Max 2.52915 +expl/Actions Min -2.51201 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 34000 +eval/num paths total 34 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.218002 +eval/Actions Std 0.654881 +eval/Actions Max 0.988852 +eval/Actions Min -0.988977 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.47383e-06 +time/evaluation sampling (s) 3.06901 +time/exploration sampling (s) 3.8414 +time/logging (s) 0.00734861 +time/saving (s) 0.0125246 +time/training (s) 14.3169 +time/epoch (s) 21.2472 +time/total (s) 783.897 +Epoch -967 +------------------------------ ---------------- +2022-05-15 18:15:51.235940 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -966 finished +------------------------------ ---------------- +epoch -966 +replay_buffer/size 999047 +trainer/num train calls 35000 +trainer/QF1 Loss 1.3511 +trainer/QF2 Loss 1.41629 +trainer/Policy Loss 12.4342 +trainer/Q1 Predictions Mean -66.3839 +trainer/Q1 Predictions Std 18.0285 +trainer/Q1 Predictions Max -4.72041 +trainer/Q1 Predictions Min -85.1215 +trainer/Q2 Predictions Mean -66.4303 +trainer/Q2 Predictions Std 18.0597 +trainer/Q2 Predictions Max -4.71866 +trainer/Q2 Predictions Min -85.3009 +trainer/Q Targets Mean -66.1807 +trainer/Q Targets Std 18.343 +trainer/Q Targets Max 0 +trainer/Q Targets Min -84.5782 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0221751 +trainer/policy/mean Std 0.640291 +trainer/policy/mean Max 0.995753 +trainer/policy/mean Min -0.992276 +trainer/policy/std Mean 0.483478 +trainer/policy/std Std 0.0294746 +trainer/policy/std Max 0.517452 +trainer/policy/std Min 0.441637 +trainer/Advantage Weights Mean 1.94013 +trainer/Advantage Weights Std 9.52446 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.44631e-11 +trainer/Advantage Score Mean -0.412963 +trainer/Advantage Score Std 0.475636 +trainer/Advantage Score Max 1.62767 +trainer/Advantage Score Min -2.34649 +trainer/V1 Predictions Mean -65.8526 +trainer/V1 Predictions Std 18.3078 +trainer/V1 Predictions Max -4.09943 +trainer/V1 Predictions Min -84.6864 +trainer/VF Loss 0.0523809 +expl/num steps total 35000 +expl/num paths total 35 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0217648 +expl/Actions Std 0.711789 +expl/Actions Max 2.34909 +expl/Actions Min -2.49403 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 35000 +eval/num paths total 35 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.186898 +eval/Actions Std 0.644351 +eval/Actions Max 0.991057 +eval/Actions Min -0.988555 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.39467e-06 +time/evaluation sampling (s) 3.07092 +time/exploration sampling (s) 3.84341 +time/logging (s) 0.00731643 +time/saving (s) 0.0135953 +time/training (s) 14.6942 +time/epoch (s) 21.6294 +time/total (s) 805.53 +Epoch -966 +------------------------------ ---------------- +2022-05-15 18:16:13.277910 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -965 finished +------------------------------ ---------------- +epoch -965 +replay_buffer/size 999047 +trainer/num train calls 36000 +trainer/QF1 Loss 1.17468 +trainer/QF2 Loss 1.19602 +trainer/Policy Loss 39.5976 +trainer/Q1 Predictions Mean -64.943 +trainer/Q1 Predictions Std 19.0704 +trainer/Q1 Predictions Max -1.94938 +trainer/Q1 Predictions Min -86.3484 +trainer/Q2 Predictions Mean -64.9473 +trainer/Q2 Predictions Std 19.1192 +trainer/Q2 Predictions Max -2.06403 +trainer/Q2 Predictions Min -86.8687 +trainer/Q Targets Mean -65.0166 +trainer/Q Targets Std 19.2978 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.4731 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0128945 +trainer/policy/mean Std 0.645971 +trainer/policy/mean Max 0.994758 +trainer/policy/mean Min -0.992784 +trainer/policy/std Mean 0.483721 +trainer/policy/std Std 0.0287958 +trainer/policy/std Max 0.515524 +trainer/policy/std Min 0.443968 +trainer/Advantage Weights Mean 6.41809 +trainer/Advantage Weights Std 21.3963 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.54861e-09 +trainer/Advantage Score Mean -0.34531 +trainer/Advantage Score Std 0.494931 +trainer/Advantage Score Max 0.920067 +trainer/Advantage Score Min -1.94567 +trainer/V1 Predictions Mean -64.7331 +trainer/V1 Predictions Std 19.2539 +trainer/V1 Predictions Max -2.10748 +trainer/V1 Predictions Min -85.7874 +trainer/VF Loss 0.0575473 +expl/num steps total 36000 +expl/num paths total 36 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0357819 +expl/Actions Std 0.764347 +expl/Actions Max 2.41065 +expl/Actions Min -2.3466 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 36000 +eval/num paths total 36 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0210608 +eval/Actions Std 0.641947 +eval/Actions Max 0.996102 +eval/Actions Min -0.99658 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.44403e-06 +time/evaluation sampling (s) 2.94558 +time/exploration sampling (s) 3.89117 +time/logging (s) 0.0111269 +time/saving (s) 0.0150731 +time/training (s) 15.1771 +time/epoch (s) 22.04 +time/total (s) 827.575 +Epoch -965 +------------------------------ ---------------- +2022-05-15 18:16:35.762597 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -964 finished +------------------------------ ---------------- +epoch -964 +replay_buffer/size 999047 +trainer/num train calls 37000 +trainer/QF1 Loss 0.57251 +trainer/QF2 Loss 0.60397 +trainer/Policy Loss 14.1631 +trainer/Q1 Predictions Mean -68.8303 +trainer/Q1 Predictions Std 16.1702 +trainer/Q1 Predictions Max -4.76206 +trainer/Q1 Predictions Min -85.3295 +trainer/Q2 Predictions Mean -68.8233 +trainer/Q2 Predictions Std 16.2097 +trainer/Q2 Predictions Max -4.42403 +trainer/Q2 Predictions Min -85.4583 +trainer/Q Targets Mean -68.6568 +trainer/Q Targets Std 16.2955 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.176 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00456486 +trainer/policy/mean Std 0.661187 +trainer/policy/mean Max 0.991062 +trainer/policy/mean Min -0.992999 +trainer/policy/std Mean 0.481266 +trainer/policy/std Std 0.0288304 +trainer/policy/std Max 0.512359 +trainer/policy/std Min 0.439651 +trainer/Advantage Weights Mean 2.50892 +trainer/Advantage Weights Std 11.9764 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.07289e-10 +trainer/Advantage Score Mean -0.372458 +trainer/Advantage Score Std 0.397315 +trainer/Advantage Score Max 0.734253 +trainer/Advantage Score Min -2.14019 +trainer/V1 Predictions Mean -68.3735 +trainer/V1 Predictions Std 16.3193 +trainer/V1 Predictions Max -4.63019 +trainer/V1 Predictions Min -85.3128 +trainer/VF Loss 0.0366056 +expl/num steps total 37000 +expl/num paths total 37 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00168024 +expl/Actions Std 0.710433 +expl/Actions Max 2.70648 +expl/Actions Min -2.61064 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 37000 +eval/num paths total 37 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00858149 +eval/Actions Std 0.608727 +eval/Actions Max 0.996587 +eval/Actions Min -0.99591 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.59792e-06 +time/evaluation sampling (s) 3.02763 +time/exploration sampling (s) 3.90542 +time/logging (s) 0.00929428 +time/saving (s) 0.0111108 +time/training (s) 15.5218 +time/epoch (s) 22.4752 +time/total (s) 850.056 +Epoch -964 +------------------------------ ---------------- +2022-05-15 18:16:57.579491 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -963 finished +------------------------------ ---------------- +epoch -963 +replay_buffer/size 999047 +trainer/num train calls 38000 +trainer/QF1 Loss 0.682641 +trainer/QF2 Loss 0.640796 +trainer/Policy Loss 11.3829 +trainer/Q1 Predictions Mean -67.1035 +trainer/Q1 Predictions Std 17.688 +trainer/Q1 Predictions Max -2.68373 +trainer/Q1 Predictions Min -85.4043 +trainer/Q2 Predictions Mean -67.2133 +trainer/Q2 Predictions Std 17.6752 +trainer/Q2 Predictions Max -2.77017 +trainer/Q2 Predictions Min -85.5795 +trainer/Q Targets Mean -67.5183 +trainer/Q Targets Std 17.6044 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.1748 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0165722 +trainer/policy/mean Std 0.652698 +trainer/policy/mean Max 0.992573 +trainer/policy/mean Min -0.989979 +trainer/policy/std Mean 0.478587 +trainer/policy/std Std 0.0285021 +trainer/policy/std Max 0.508769 +trainer/policy/std Min 0.438139 +trainer/Advantage Weights Mean 2.72837 +trainer/Advantage Weights Std 13.9767 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.14391e-09 +trainer/Advantage Score Mean -0.387959 +trainer/Advantage Score Std 0.415035 +trainer/Advantage Score Max 0.966125 +trainer/Advantage Score Min -1.89078 +trainer/V1 Predictions Mean -67.2136 +trainer/V1 Predictions Std 17.7169 +trainer/V1 Predictions Max -3.44734 +trainer/V1 Predictions Min -86.0531 +trainer/VF Loss 0.0431162 +expl/num steps total 38000 +expl/num paths total 38 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0126785 +expl/Actions Std 0.74931 +expl/Actions Max 2.43053 +expl/Actions Min -2.20027 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 38000 +eval/num paths total 38 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0213989 +eval/Actions Std 0.559508 +eval/Actions Max 0.993713 +eval/Actions Min -0.987245 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.83029e-06 +time/evaluation sampling (s) 3.02819 +time/exploration sampling (s) 3.85995 +time/logging (s) 0.00832247 +time/saving (s) 0.0158383 +time/training (s) 14.898 +time/epoch (s) 21.8103 +time/total (s) 871.871 +Epoch -963 +------------------------------ ---------------- +2022-05-15 18:17:19.799297 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -962 finished +------------------------------ ---------------- +epoch -962 +replay_buffer/size 999047 +trainer/num train calls 39000 +trainer/QF1 Loss 4.99736 +trainer/QF2 Loss 5.01232 +trainer/Policy Loss 106.824 +trainer/Q1 Predictions Mean -67.1079 +trainer/Q1 Predictions Std 18.0591 +trainer/Q1 Predictions Max -2.17538 +trainer/Q1 Predictions Min -87.7264 +trainer/Q2 Predictions Mean -67.0167 +trainer/Q2 Predictions Std 18.0754 +trainer/Q2 Predictions Max -2.2595 +trainer/Q2 Predictions Min -87.7269 +trainer/Q Targets Mean -68.0532 +trainer/Q Targets Std 17.7312 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.7341 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00917009 +trainer/policy/mean Std 0.649561 +trainer/policy/mean Max 0.997153 +trainer/policy/mean Min -0.997925 +trainer/policy/std Mean 0.477918 +trainer/policy/std Std 0.0271091 +trainer/policy/std Max 0.507887 +trainer/policy/std Min 0.439399 +trainer/Advantage Weights Mean 20.7503 +trainer/Advantage Weights Std 35.7899 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.63773e-08 +trainer/Advantage Score Mean 0.10991 +trainer/Advantage Score Std 0.472917 +trainer/Advantage Score Max 2.19226 +trainer/Advantage Score Min -1.74508 +trainer/V1 Predictions Mean -67.7785 +trainer/V1 Predictions Std 17.7241 +trainer/V1 Predictions Max -4.89891 +trainer/V1 Predictions Min -88.353 +trainer/VF Loss 0.165252 +expl/num steps total 39000 +expl/num paths total 39 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00714094 +expl/Actions Std 0.76305 +expl/Actions Max 2.44654 +expl/Actions Min -2.29906 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 39000 +eval/num paths total 39 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0305851 +eval/Actions Std 0.631943 +eval/Actions Max 0.994106 +eval/Actions Min -0.992363 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.69199e-06 +time/evaluation sampling (s) 3.27921 +time/exploration sampling (s) 3.84831 +time/logging (s) 0.00745108 +time/saving (s) 0.0111735 +time/training (s) 15.0665 +time/epoch (s) 22.2126 +time/total (s) 894.089 +Epoch -962 +------------------------------ ---------------- +2022-05-15 18:17:41.641758 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -961 finished +------------------------------ ---------------- +epoch -961 +replay_buffer/size 999047 +trainer/num train calls 40000 +trainer/QF1 Loss 0.787359 +trainer/QF2 Loss 0.801161 +trainer/Policy Loss 62.2989 +trainer/Q1 Predictions Mean -68.1781 +trainer/Q1 Predictions Std 17.5183 +trainer/Q1 Predictions Max -0.4991 +trainer/Q1 Predictions Min -85.7751 +trainer/Q2 Predictions Mean -68.1721 +trainer/Q2 Predictions Std 17.4952 +trainer/Q2 Predictions Max -0.620864 +trainer/Q2 Predictions Min -85.8384 +trainer/Q Targets Mean -68.5866 +trainer/Q Targets Std 17.4077 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2707 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0012913 +trainer/policy/mean Std 0.657498 +trainer/policy/mean Max 0.997692 +trainer/policy/mean Min -0.988612 +trainer/policy/std Mean 0.475618 +trainer/policy/std Std 0.0275814 +trainer/policy/std Max 0.508117 +trainer/policy/std Min 0.435242 +trainer/Advantage Weights Mean 12.6232 +trainer/Advantage Weights Std 28.4274 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.17638e-11 +trainer/Advantage Score Mean -0.136406 +trainer/Advantage Score Std 0.465198 +trainer/Advantage Score Max 1.15582 +trainer/Advantage Score Min -2.36843 +trainer/V1 Predictions Mean -68.3086 +trainer/V1 Predictions Std 17.5 +trainer/V1 Predictions Max -1.00769 +trainer/V1 Predictions Min -86.1851 +trainer/VF Loss 0.0703524 +expl/num steps total 40000 +expl/num paths total 40 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00239599 +expl/Actions Std 0.755352 +expl/Actions Max 2.14348 +expl/Actions Min -2.56472 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 40000 +eval/num paths total 40 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0116398 +eval/Actions Std 0.647756 +eval/Actions Max 0.994808 +eval/Actions Min -0.996088 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.14973e-06 +time/evaluation sampling (s) 3.26884 +time/exploration sampling (s) 3.80091 +time/logging (s) 0.00796008 +time/saving (s) 0.00982105 +time/training (s) 14.7494 +time/epoch (s) 21.8369 +time/total (s) 915.931 +Epoch -961 +------------------------------ ---------------- +2022-05-15 18:18:03.120541 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -960 finished +------------------------------ ---------------- +epoch -960 +replay_buffer/size 999047 +trainer/num train calls 41000 +trainer/QF1 Loss 0.62832 +trainer/QF2 Loss 0.640848 +trainer/Policy Loss 30.872 +trainer/Q1 Predictions Mean -67.603 +trainer/Q1 Predictions Std 18.0077 +trainer/Q1 Predictions Max -2.59017 +trainer/Q1 Predictions Min -88.6025 +trainer/Q2 Predictions Mean -67.6149 +trainer/Q2 Predictions Std 18.0312 +trainer/Q2 Predictions Max -3.59631 +trainer/Q2 Predictions Min -88.7551 +trainer/Q Targets Mean -67.636 +trainer/Q Targets Std 17.9058 +trainer/Q Targets Max -3.14831 +trainer/Q Targets Min -88.0156 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000451161 +trainer/policy/mean Std 0.651693 +trainer/policy/mean Max 0.993233 +trainer/policy/mean Min -0.988213 +trainer/policy/std Mean 0.474946 +trainer/policy/std Std 0.0273154 +trainer/policy/std Max 0.50634 +trainer/policy/std Min 0.435808 +trainer/Advantage Weights Mean 5.24157 +trainer/Advantage Weights Std 19.3189 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.28068e-09 +trainer/Advantage Score Mean -0.34483 +trainer/Advantage Score Std 0.428808 +trainer/Advantage Score Max 0.829399 +trainer/Advantage Score Min -1.8738 +trainer/V1 Predictions Mean -67.3166 +trainer/V1 Predictions Std 18.0183 +trainer/V1 Predictions Max -2.09233 +trainer/V1 Predictions Min -87.8875 +trainer/VF Loss 0.0448566 +expl/num steps total 41000 +expl/num paths total 41 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.000238329 +expl/Actions Std 0.752986 +expl/Actions Max 2.62424 +expl/Actions Min -2.58561 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 41000 +eval/num paths total 41 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0334927 +eval/Actions Std 0.681955 +eval/Actions Max 0.997381 +eval/Actions Min -0.998359 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.72762e-06 +time/evaluation sampling (s) 2.96144 +time/exploration sampling (s) 3.68196 +time/logging (s) 0.00746358 +time/saving (s) 0.0104326 +time/training (s) 14.8117 +time/epoch (s) 21.473 +time/total (s) 937.408 +Epoch -960 +------------------------------ ---------------- +2022-05-15 18:18:25.144272 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -959 finished +------------------------------ ---------------- +epoch -959 +replay_buffer/size 999047 +trainer/num train calls 42000 +trainer/QF1 Loss 1.13404 +trainer/QF2 Loss 1.21644 +trainer/Policy Loss 10.9998 +trainer/Q1 Predictions Mean -67.3633 +trainer/Q1 Predictions Std 20.0961 +trainer/Q1 Predictions Max -1.27689 +trainer/Q1 Predictions Min -88.8895 +trainer/Q2 Predictions Mean -67.4441 +trainer/Q2 Predictions Std 20.1137 +trainer/Q2 Predictions Max -2.13325 +trainer/Q2 Predictions Min -88.8409 +trainer/Q Targets Mean -67.216 +trainer/Q Targets Std 20.2969 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.9346 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0135627 +trainer/policy/mean Std 0.663628 +trainer/policy/mean Max 0.9972 +trainer/policy/mean Min -0.995484 +trainer/policy/std Mean 0.473228 +trainer/policy/std Std 0.027072 +trainer/policy/std Max 0.503364 +trainer/policy/std Min 0.431479 +trainer/Advantage Weights Mean 2.13689 +trainer/Advantage Weights Std 11.4989 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.63304e-16 +trainer/Advantage Score Mean -0.522377 +trainer/Advantage Score Std 0.541373 +trainer/Advantage Score Max 0.761883 +trainer/Advantage Score Min -3.63509 +trainer/V1 Predictions Mean -66.9862 +trainer/V1 Predictions Std 20.3473 +trainer/V1 Predictions Max -1.13208 +trainer/V1 Predictions Min -89.2879 +trainer/VF Loss 0.0633901 +expl/num steps total 42000 +expl/num paths total 42 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.182906 +expl/Actions Std 0.841558 +expl/Actions Max 2.46197 +expl/Actions Min -2.36965 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 42000 +eval/num paths total 42 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0135373 +eval/Actions Std 0.660475 +eval/Actions Max 0.997871 +eval/Actions Min -0.991586 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.89408e-06 +time/evaluation sampling (s) 2.97052 +time/exploration sampling (s) 3.76817 +time/logging (s) 0.0111834 +time/saving (s) 0.0162889 +time/training (s) 15.2546 +time/epoch (s) 22.0207 +time/total (s) 959.434 +Epoch -959 +------------------------------ ---------------- +2022-05-15 18:18:47.404924 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -958 finished +------------------------------ ---------------- +epoch -958 +replay_buffer/size 999047 +trainer/num train calls 43000 +trainer/QF1 Loss 0.799339 +trainer/QF2 Loss 0.81877 +trainer/Policy Loss 11.7198 +trainer/Q1 Predictions Mean -68.088 +trainer/Q1 Predictions Std 18.8076 +trainer/Q1 Predictions Max -2.72189 +trainer/Q1 Predictions Min -86.5468 +trainer/Q2 Predictions Mean -68.0961 +trainer/Q2 Predictions Std 18.8241 +trainer/Q2 Predictions Max -2.92037 +trainer/Q2 Predictions Min -86.801 +trainer/Q Targets Mean -68.4342 +trainer/Q Targets Std 18.7714 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6793 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0173677 +trainer/policy/mean Std 0.656857 +trainer/policy/mean Max 0.99636 +trainer/policy/mean Min -0.99461 +trainer/policy/std Mean 0.472587 +trainer/policy/std Std 0.0274129 +trainer/policy/std Max 0.502512 +trainer/policy/std Min 0.429518 +trainer/Advantage Weights Mean 2.41904 +trainer/Advantage Weights Std 11.7384 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.76801e-19 +trainer/Advantage Score Mean -0.358878 +trainer/Advantage Score Std 0.498331 +trainer/Advantage Score Max 1.06476 +trainer/Advantage Score Min -4.21872 +trainer/V1 Predictions Mean -68.1631 +trainer/V1 Predictions Std 18.8619 +trainer/V1 Predictions Max -2.96805 +trainer/V1 Predictions Min -86.4489 +trainer/VF Loss 0.0472399 +expl/num steps total 43000 +expl/num paths total 43 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0323778 +expl/Actions Std 0.808616 +expl/Actions Max 2.59509 +expl/Actions Min -2.46955 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 43000 +eval/num paths total 43 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0832383 +eval/Actions Std 0.642472 +eval/Actions Max 0.994801 +eval/Actions Min -0.996646 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0075e-05 +time/evaluation sampling (s) 3.21564 +time/exploration sampling (s) 3.72567 +time/logging (s) 0.0114527 +time/saving (s) 0.0156686 +time/training (s) 15.2831 +time/epoch (s) 22.2515 +time/total (s) 981.693 +Epoch -958 +------------------------------ ---------------- +2022-05-15 18:19:09.264307 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -957 finished +------------------------------ ---------------- +epoch -957 +replay_buffer/size 999047 +trainer/num train calls 44000 +trainer/QF1 Loss 0.958653 +trainer/QF2 Loss 0.960245 +trainer/Policy Loss 30.8362 +trainer/Q1 Predictions Mean -67.6908 +trainer/Q1 Predictions Std 18.7683 +trainer/Q1 Predictions Max -3.55679 +trainer/Q1 Predictions Min -87.4674 +trainer/Q2 Predictions Mean -67.7144 +trainer/Q2 Predictions Std 18.8151 +trainer/Q2 Predictions Max -3.85221 +trainer/Q2 Predictions Min -87.4404 +trainer/Q Targets Mean -67.8824 +trainer/Q Targets Std 19.0445 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9016 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0302399 +trainer/policy/mean Std 0.655591 +trainer/policy/mean Max 0.993119 +trainer/policy/mean Min -0.987503 +trainer/policy/std Mean 0.47193 +trainer/policy/std Std 0.0279308 +trainer/policy/std Max 0.504879 +trainer/policy/std Min 0.432634 +trainer/Advantage Weights Mean 5.76242 +trainer/Advantage Weights Std 20.3721 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.25205e-12 +trainer/Advantage Score Mean -0.366824 +trainer/Advantage Score Std 0.574757 +trainer/Advantage Score Max 0.807808 +trainer/Advantage Score Min -2.57981 +trainer/V1 Predictions Mean -67.581 +trainer/V1 Predictions Std 19.0767 +trainer/V1 Predictions Max -2.87003 +trainer/V1 Predictions Min -87.5311 +trainer/VF Loss 0.0625597 +expl/num steps total 44000 +expl/num paths total 44 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0153782 +expl/Actions Std 0.780183 +expl/Actions Max 2.34342 +expl/Actions Min -2.43481 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 44000 +eval/num paths total 44 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0523501 +eval/Actions Std 0.686507 +eval/Actions Max 0.997807 +eval/Actions Min -0.995698 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.59792e-06 +time/evaluation sampling (s) 3.00591 +time/exploration sampling (s) 3.86893 +time/logging (s) 0.00739376 +time/saving (s) 0.0104896 +time/training (s) 14.9563 +time/epoch (s) 21.849 +time/total (s) 1003.55 +Epoch -957 +------------------------------ ---------------- +2022-05-15 18:19:30.714265 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -956 finished +------------------------------ ---------------- +epoch -956 +replay_buffer/size 999047 +trainer/num train calls 45000 +trainer/QF1 Loss 0.947179 +trainer/QF2 Loss 0.811437 +trainer/Policy Loss 26.6337 +trainer/Q1 Predictions Mean -68.7819 +trainer/Q1 Predictions Std 17.9747 +trainer/Q1 Predictions Max -3.78122 +trainer/Q1 Predictions Min -88.8826 +trainer/Q2 Predictions Mean -68.8496 +trainer/Q2 Predictions Std 17.9832 +trainer/Q2 Predictions Max -4.14577 +trainer/Q2 Predictions Min -88.421 +trainer/Q Targets Mean -68.9922 +trainer/Q Targets Std 17.6363 +trainer/Q Targets Max -4.26146 +trainer/Q Targets Min -87.4427 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00636772 +trainer/policy/mean Std 0.677445 +trainer/policy/mean Max 0.995349 +trainer/policy/mean Min -0.9973 +trainer/policy/std Mean 0.47088 +trainer/policy/std Std 0.0272365 +trainer/policy/std Max 0.503806 +trainer/policy/std Min 0.435362 +trainer/Advantage Weights Mean 4.99924 +trainer/Advantage Weights Std 18.5771 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.27752e-15 +trainer/Advantage Score Mean -0.333219 +trainer/Advantage Score Std 0.479832 +trainer/Advantage Score Max 1.42555 +trainer/Advantage Score Min -3.30854 +trainer/V1 Predictions Mean -68.6001 +trainer/V1 Predictions Std 17.999 +trainer/V1 Predictions Max -3.97992 +trainer/V1 Predictions Min -88.216 +trainer/VF Loss 0.0573185 +expl/num steps total 45000 +expl/num paths total 45 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00810269 +expl/Actions Std 0.7975 +expl/Actions Max 2.28001 +expl/Actions Min -2.47658 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 45000 +eval/num paths total 45 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00752646 +eval/Actions Std 0.6658 +eval/Actions Max 0.998267 +eval/Actions Min -0.993953 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.50322e-06 +time/evaluation sampling (s) 2.94567 +time/exploration sampling (s) 3.75638 +time/logging (s) 0.0100703 +time/saving (s) 0.0150799 +time/training (s) 14.7195 +time/epoch (s) 21.4467 +time/total (s) 1025 +Epoch -956 +------------------------------ ---------------- +2022-05-15 18:19:52.976907 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -955 finished +------------------------------ ---------------- +epoch -955 +replay_buffer/size 999047 +trainer/num train calls 46000 +trainer/QF1 Loss 0.855811 +trainer/QF2 Loss 0.928835 +trainer/Policy Loss 16.0261 +trainer/Q1 Predictions Mean -69.0953 +trainer/Q1 Predictions Std 17.2668 +trainer/Q1 Predictions Max -4.66633 +trainer/Q1 Predictions Min -86.6133 +trainer/Q2 Predictions Mean -68.9546 +trainer/Q2 Predictions Std 17.3286 +trainer/Q2 Predictions Max -5.08871 +trainer/Q2 Predictions Min -86.4199 +trainer/Q Targets Mean -69.132 +trainer/Q Targets Std 17.5653 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5706 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00545233 +trainer/policy/mean Std 0.675647 +trainer/policy/mean Max 0.994086 +trainer/policy/mean Min -0.991214 +trainer/policy/std Mean 0.471797 +trainer/policy/std Std 0.0263356 +trainer/policy/std Max 0.501923 +trainer/policy/std Min 0.435396 +trainer/Advantage Weights Mean 3.11791 +trainer/Advantage Weights Std 14.343 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.89164e-12 +trainer/Advantage Score Mean -0.402867 +trainer/Advantage Score Std 0.501409 +trainer/Advantage Score Max 1.08801 +trainer/Advantage Score Min -2.58575 +trainer/V1 Predictions Mean -68.857 +trainer/V1 Predictions Std 17.6813 +trainer/V1 Predictions Max -4.43274 +trainer/V1 Predictions Min -87.4098 +trainer/VF Loss 0.0517677 +expl/num steps total 46000 +expl/num paths total 46 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.083104 +expl/Actions Std 0.786451 +expl/Actions Max 2.71025 +expl/Actions Min -2.27617 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 46000 +eval/num paths total 46 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00968704 +eval/Actions Std 0.600839 +eval/Actions Max 0.997286 +eval/Actions Min -0.995488 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.85171e-06 +time/evaluation sampling (s) 3.26141 +time/exploration sampling (s) 3.69601 +time/logging (s) 0.00780364 +time/saving (s) 0.0132061 +time/training (s) 15.2735 +time/epoch (s) 22.2519 +time/total (s) 1047.26 +Epoch -955 +------------------------------ ---------------- +2022-05-15 18:20:15.342007 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -954 finished +------------------------------ ---------------- +epoch -954 +replay_buffer/size 999047 +trainer/num train calls 47000 +trainer/QF1 Loss 0.44301 +trainer/QF2 Loss 0.522905 +trainer/Policy Loss 17.4493 +trainer/Q1 Predictions Mean -69.8987 +trainer/Q1 Predictions Std 17.3621 +trainer/Q1 Predictions Max -1.91429 +trainer/Q1 Predictions Min -87.9033 +trainer/Q2 Predictions Mean -69.8675 +trainer/Q2 Predictions Std 17.3972 +trainer/Q2 Predictions Max -2.73026 +trainer/Q2 Predictions Min -87.9163 +trainer/Q Targets Mean -69.7841 +trainer/Q Targets Std 17.3181 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4961 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0140009 +trainer/policy/mean Std 0.653149 +trainer/policy/mean Max 0.996042 +trainer/policy/mean Min -0.993029 +trainer/policy/std Mean 0.473037 +trainer/policy/std Std 0.0267332 +trainer/policy/std Max 0.503333 +trainer/policy/std Min 0.4357 +trainer/Advantage Weights Mean 2.88306 +trainer/Advantage Weights Std 13.8158 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.83474e-08 +trainer/Advantage Score Mean -0.337596 +trainer/Advantage Score Std 0.385513 +trainer/Advantage Score Max 0.887436 +trainer/Advantage Score Min -1.73787 +trainer/V1 Predictions Mean -69.4776 +trainer/V1 Predictions Std 17.3439 +trainer/V1 Predictions Max -2.89341 +trainer/V1 Predictions Min -87.3533 +trainer/VF Loss 0.0356465 +expl/num steps total 47000 +expl/num paths total 47 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00315295 +expl/Actions Std 0.780389 +expl/Actions Max 2.45539 +expl/Actions Min -2.34772 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 47000 +eval/num paths total 47 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00665316 +eval/Actions Std 0.676806 +eval/Actions Max 0.994583 +eval/Actions Min -0.996731 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.65172e-06 +time/evaluation sampling (s) 3.11606 +time/exploration sampling (s) 3.80917 +time/logging (s) 0.0107904 +time/saving (s) 0.0150168 +time/training (s) 15.4093 +time/epoch (s) 22.3603 +time/total (s) 1069.62 +Epoch -954 +------------------------------ ---------------- +2022-05-15 18:20:37.370635 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -953 finished +------------------------------ ---------------- +epoch -953 +replay_buffer/size 999047 +trainer/num train calls 48000 +trainer/QF1 Loss 0.661001 +trainer/QF2 Loss 0.735502 +trainer/Policy Loss 2.72758 +trainer/Q1 Predictions Mean -68.8057 +trainer/Q1 Predictions Std 17.4909 +trainer/Q1 Predictions Max -3.08606 +trainer/Q1 Predictions Min -86.7223 +trainer/Q2 Predictions Mean -68.7479 +trainer/Q2 Predictions Std 17.5215 +trainer/Q2 Predictions Max -3.55859 +trainer/Q2 Predictions Min -86.5216 +trainer/Q Targets Mean -69.0511 +trainer/Q Targets Std 17.6358 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6575 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000589841 +trainer/policy/mean Std 0.667621 +trainer/policy/mean Max 0.993803 +trainer/policy/mean Min -0.995252 +trainer/policy/std Mean 0.47082 +trainer/policy/std Std 0.0280079 +trainer/policy/std Max 0.501711 +trainer/policy/std Min 0.43281 +trainer/Advantage Weights Mean 0.450092 +trainer/Advantage Weights Std 3.15204 +trainer/Advantage Weights Max 41.0923 +trainer/Advantage Weights Min 8.43064e-13 +trainer/Advantage Score Mean -0.76999 +trainer/Advantage Score Std 0.529096 +trainer/Advantage Score Max 0.371582 +trainer/Advantage Score Min -2.78017 +trainer/V1 Predictions Mean -68.7096 +trainer/V1 Predictions Std 17.6715 +trainer/V1 Predictions Max -2.15884 +trainer/V1 Predictions Min -87.0592 +trainer/VF Loss 0.0885913 +expl/num steps total 48000 +expl/num paths total 48 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00877141 +expl/Actions Std 0.912569 +expl/Actions Max 2.4697 +expl/Actions Min -2.35266 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 48000 +eval/num paths total 48 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0552004 +eval/Actions Std 0.654671 +eval/Actions Max 0.997192 +eval/Actions Min -0.994228 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.01329e-06 +time/evaluation sampling (s) 3.18396 +time/exploration sampling (s) 3.98565 +time/logging (s) 0.0098667 +time/saving (s) 0.0142603 +time/training (s) 14.8254 +time/epoch (s) 22.0191 +time/total (s) 1091.65 +Epoch -953 +------------------------------ ---------------- +2022-05-15 18:20:59.244876 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -952 finished +------------------------------ ---------------- +epoch -952 +replay_buffer/size 999047 +trainer/num train calls 49000 +trainer/QF1 Loss 1.02722 +trainer/QF2 Loss 0.961243 +trainer/Policy Loss 8.73744 +trainer/Q1 Predictions Mean -68.3909 +trainer/Q1 Predictions Std 18.4049 +trainer/Q1 Predictions Max -1.55536 +trainer/Q1 Predictions Min -86.2134 +trainer/Q2 Predictions Mean -68.3712 +trainer/Q2 Predictions Std 18.3898 +trainer/Q2 Predictions Max -1.41981 +trainer/Q2 Predictions Min -86.3002 +trainer/Q Targets Mean -67.9716 +trainer/Q Targets Std 18.5707 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.3408 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00737593 +trainer/policy/mean Std 0.682295 +trainer/policy/mean Max 0.998711 +trainer/policy/mean Min -0.999416 +trainer/policy/std Mean 0.46928 +trainer/policy/std Std 0.0262093 +trainer/policy/std Max 0.498167 +trainer/policy/std Min 0.43249 +trainer/Advantage Weights Mean 1.26466 +trainer/Advantage Weights Std 9.55955 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.38499e-17 +trainer/Advantage Score Mean -0.586581 +trainer/Advantage Score Std 0.488619 +trainer/Advantage Score Max 0.559498 +trainer/Advantage Score Min -3.79246 +trainer/V1 Predictions Mean -67.7609 +trainer/V1 Predictions Std 18.5587 +trainer/V1 Predictions Max -1.17683 +trainer/V1 Predictions Min -85.284 +trainer/VF Loss 0.0611836 +expl/num steps total 49000 +expl/num paths total 49 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0180836 +expl/Actions Std 0.769475 +expl/Actions Max 2.29033 +expl/Actions Min -2.40758 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 49000 +eval/num paths total 49 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0346667 +eval/Actions Std 0.647526 +eval/Actions Max 0.996782 +eval/Actions Min -0.99701 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.80328e-06 +time/evaluation sampling (s) 2.95482 +time/exploration sampling (s) 3.75338 +time/logging (s) 0.00693629 +time/saving (s) 0.00976296 +time/training (s) 15.1383 +time/epoch (s) 21.8632 +time/total (s) 1113.52 +Epoch -952 +------------------------------ ---------------- +2022-05-15 18:21:20.926793 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -951 finished +------------------------------ ---------------- +epoch -951 +replay_buffer/size 999047 +trainer/num train calls 50000 +trainer/QF1 Loss 0.719014 +trainer/QF2 Loss 0.758693 +trainer/Policy Loss 3.23446 +trainer/Q1 Predictions Mean -67.9826 +trainer/Q1 Predictions Std 19.4368 +trainer/Q1 Predictions Max -4.91193 +trainer/Q1 Predictions Min -88.8631 +trainer/Q2 Predictions Mean -68.0195 +trainer/Q2 Predictions Std 19.4782 +trainer/Q2 Predictions Max -5.12528 +trainer/Q2 Predictions Min -89.0612 +trainer/Q Targets Mean -67.9582 +trainer/Q Targets Std 19.5764 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9064 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0067003 +trainer/policy/mean Std 0.673667 +trainer/policy/mean Max 0.996764 +trainer/policy/mean Min -0.999181 +trainer/policy/std Mean 0.466318 +trainer/policy/std Std 0.0272158 +trainer/policy/std Max 0.495196 +trainer/policy/std Min 0.428239 +trainer/Advantage Weights Mean 1.17189 +trainer/Advantage Weights Std 7.49321 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.80531e-12 +trainer/Advantage Score Mean -0.454572 +trainer/Advantage Score Std 0.425883 +trainer/Advantage Score Max 0.521973 +trainer/Advantage Score Min -2.70403 +trainer/V1 Predictions Mean -67.6764 +trainer/V1 Predictions Std 19.7158 +trainer/V1 Predictions Max -3.59527 +trainer/V1 Predictions Min -88.6782 +trainer/VF Loss 0.0418203 +expl/num steps total 50000 +expl/num paths total 50 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0425051 +expl/Actions Std 0.772286 +expl/Actions Max 2.82089 +expl/Actions Min -2.29954 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 50000 +eval/num paths total 50 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00296297 +eval/Actions Std 0.646273 +eval/Actions Max 0.996796 +eval/Actions Min -0.997288 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.77767e-06 +time/evaluation sampling (s) 3.08047 +time/exploration sampling (s) 3.77401 +time/logging (s) 0.0120829 +time/saving (s) 0.0152054 +time/training (s) 14.7996 +time/epoch (s) 21.6814 +time/total (s) 1135.21 +Epoch -951 +------------------------------ ---------------- +2022-05-15 18:21:42.195347 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -950 finished +------------------------------ ---------------- +epoch -950 +replay_buffer/size 999047 +trainer/num train calls 51000 +trainer/QF1 Loss 0.9363 +trainer/QF2 Loss 1.03406 +trainer/Policy Loss 18.4615 +trainer/Q1 Predictions Mean -68.3022 +trainer/Q1 Predictions Std 19.6845 +trainer/Q1 Predictions Max -2.07561 +trainer/Q1 Predictions Min -88.3753 +trainer/Q2 Predictions Mean -68.3272 +trainer/Q2 Predictions Std 19.6816 +trainer/Q2 Predictions Max -1.76605 +trainer/Q2 Predictions Min -88.1524 +trainer/Q Targets Mean -68.3595 +trainer/Q Targets Std 19.5385 +trainer/Q Targets Max -2.41214 +trainer/Q Targets Min -88.2086 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0084892 +trainer/policy/mean Std 0.672264 +trainer/policy/mean Max 0.997257 +trainer/policy/mean Min -0.994704 +trainer/policy/std Mean 0.467678 +trainer/policy/std Std 0.0256643 +trainer/policy/std Max 0.497187 +trainer/policy/std Min 0.432188 +trainer/Advantage Weights Mean 2.84486 +trainer/Advantage Weights Std 14.2364 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.33561e-15 +trainer/Advantage Score Mean -0.40839 +trainer/Advantage Score Std 0.443868 +trainer/Advantage Score Max 1.04109 +trainer/Advantage Score Min -3.33341 +trainer/V1 Predictions Mean -68.0287 +trainer/V1 Predictions Std 19.7273 +trainer/V1 Predictions Max -1.77084 +trainer/V1 Predictions Min -87.9935 +trainer/VF Loss 0.0461634 +expl/num steps total 51000 +expl/num paths total 51 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00343678 +expl/Actions Std 0.781202 +expl/Actions Max 2.3806 +expl/Actions Min -2.52343 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 51000 +eval/num paths total 51 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0249598 +eval/Actions Std 0.603886 +eval/Actions Max 0.99694 +eval/Actions Min -0.992907 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.69664e-06 +time/evaluation sampling (s) 3.01888 +time/exploration sampling (s) 3.68789 +time/logging (s) 0.00923894 +time/saving (s) 0.0225612 +time/training (s) 14.5197 +time/epoch (s) 21.2583 +time/total (s) 1156.47 +Epoch -950 +------------------------------ ---------------- +2022-05-15 18:22:04.486672 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -949 finished +------------------------------ ---------------- +epoch -949 +replay_buffer/size 999047 +trainer/num train calls 52000 +trainer/QF1 Loss 0.92674 +trainer/QF2 Loss 0.943808 +trainer/Policy Loss 27.6762 +trainer/Q1 Predictions Mean -67.6479 +trainer/Q1 Predictions Std 19.3614 +trainer/Q1 Predictions Max -2.65245 +trainer/Q1 Predictions Min -87.8462 +trainer/Q2 Predictions Mean -67.5068 +trainer/Q2 Predictions Std 19.4073 +trainer/Q2 Predictions Max -3.65451 +trainer/Q2 Predictions Min -87.0416 +trainer/Q Targets Mean -67.6718 +trainer/Q Targets Std 19.4535 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0551 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00547495 +trainer/policy/mean Std 0.667412 +trainer/policy/mean Max 0.997525 +trainer/policy/mean Min -0.991764 +trainer/policy/std Mean 0.468345 +trainer/policy/std Std 0.0250052 +trainer/policy/std Max 0.495523 +trainer/policy/std Min 0.430899 +trainer/Advantage Weights Mean 5.35783 +trainer/Advantage Weights Std 18.9653 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.62974e-16 +trainer/Advantage Score Mean -0.288546 +trainer/Advantage Score Std 0.479465 +trainer/Advantage Score Max 0.87413 +trainer/Advantage Score Min -3.63529 +trainer/V1 Predictions Mean -67.4176 +trainer/V1 Predictions Std 19.6552 +trainer/V1 Predictions Max -2.99598 +trainer/V1 Predictions Min -88.0781 +trainer/VF Loss 0.0483192 +expl/num steps total 52000 +expl/num paths total 52 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0234995 +expl/Actions Std 0.805293 +expl/Actions Max 2.63088 +expl/Actions Min -2.45504 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 52000 +eval/num paths total 52 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0335536 +eval/Actions Std 0.670028 +eval/Actions Max 0.997196 +eval/Actions Min -0.994049 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.41795e-06 +time/evaluation sampling (s) 3.48227 +time/exploration sampling (s) 4.14911 +time/logging (s) 0.0123016 +time/saving (s) 0.017967 +time/training (s) 14.6279 +time/epoch (s) 22.2896 +time/total (s) 1178.76 +Epoch -949 +------------------------------ ---------------- +2022-05-15 18:22:26.311395 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -948 finished +------------------------------ ---------------- +epoch -948 +replay_buffer/size 999047 +trainer/num train calls 53000 +trainer/QF1 Loss 0.834706 +trainer/QF2 Loss 0.815671 +trainer/Policy Loss 26.2223 +trainer/Q1 Predictions Mean -68.7017 +trainer/Q1 Predictions Std 19.1522 +trainer/Q1 Predictions Max -2.93676 +trainer/Q1 Predictions Min -87.0383 +trainer/Q2 Predictions Mean -68.7736 +trainer/Q2 Predictions Std 19.0896 +trainer/Q2 Predictions Max -3.45726 +trainer/Q2 Predictions Min -87.17 +trainer/Q Targets Mean -68.8507 +trainer/Q Targets Std 18.8961 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1535 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.010861 +trainer/policy/mean Std 0.671437 +trainer/policy/mean Max 0.994712 +trainer/policy/mean Min -0.993558 +trainer/policy/std Mean 0.465807 +trainer/policy/std Std 0.0265724 +trainer/policy/std Max 0.494796 +trainer/policy/std Min 0.421815 +trainer/Advantage Weights Mean 4.59303 +trainer/Advantage Weights Std 19.7697 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.01807e-08 +trainer/Advantage Score Mean -0.354627 +trainer/Advantage Score Std 0.408174 +trainer/Advantage Score Max 1.13721 +trainer/Advantage Score Min -1.84028 +trainer/V1 Predictions Mean -68.5706 +trainer/V1 Predictions Std 19.0719 +trainer/V1 Predictions Max -3.25156 +trainer/V1 Predictions Min -87.0873 +trainer/VF Loss 0.0495233 +expl/num steps total 53000 +expl/num paths total 53 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0153718 +expl/Actions Std 0.786787 +expl/Actions Max 2.34893 +expl/Actions Min -2.60973 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 53000 +eval/num paths total 53 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00642972 +eval/Actions Std 0.686979 +eval/Actions Max 0.996979 +eval/Actions Min -0.99514 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.03099e-06 +time/evaluation sampling (s) 3.24778 +time/exploration sampling (s) 3.64463 +time/logging (s) 0.0108822 +time/saving (s) 0.0150861 +time/training (s) 14.8967 +time/epoch (s) 21.8151 +time/total (s) 1200.59 +Epoch -948 +------------------------------ ---------------- +2022-05-15 18:22:47.830765 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -947 finished +------------------------------ ---------------- +epoch -947 +replay_buffer/size 999047 +trainer/num train calls 54000 +trainer/QF1 Loss 1.14947 +trainer/QF2 Loss 1.37819 +trainer/Policy Loss 6.33452 +trainer/Q1 Predictions Mean -70.6009 +trainer/Q1 Predictions Std 17.7363 +trainer/Q1 Predictions Max -2.35009 +trainer/Q1 Predictions Min -88.7246 +trainer/Q2 Predictions Mean -70.7578 +trainer/Q2 Predictions Std 17.6666 +trainer/Q2 Predictions Max -3.23755 +trainer/Q2 Predictions Min -89.3191 +trainer/Q Targets Mean -70.2144 +trainer/Q Targets Std 17.648 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2788 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0127346 +trainer/policy/mean Std 0.666863 +trainer/policy/mean Max 0.998117 +trainer/policy/mean Min -0.99157 +trainer/policy/std Mean 0.466638 +trainer/policy/std Std 0.0252353 +trainer/policy/std Max 0.492675 +trainer/policy/std Min 0.423196 +trainer/Advantage Weights Mean 1.87033 +trainer/Advantage Weights Std 12.1747 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.47945e-12 +trainer/Advantage Score Mean -0.46778 +trainer/Advantage Score Std 0.431418 +trainer/Advantage Score Max 0.677742 +trainer/Advantage Score Min -2.57624 +trainer/V1 Predictions Mean -69.9967 +trainer/V1 Predictions Std 17.6601 +trainer/V1 Predictions Max -2.42368 +trainer/V1 Predictions Min -87.8479 +trainer/VF Loss 0.0458057 +expl/num steps total 54000 +expl/num paths total 54 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.026821 +expl/Actions Std 0.776349 +expl/Actions Max 2.72836 +expl/Actions Min -2.51297 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 54000 +eval/num paths total 54 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.070534 +eval/Actions Std 0.640588 +eval/Actions Max 0.996974 +eval/Actions Min -0.996396 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.04403e-06 +time/evaluation sampling (s) 2.98553 +time/exploration sampling (s) 3.7487 +time/logging (s) 0.0112478 +time/saving (s) 0.0145146 +time/training (s) 14.7511 +time/epoch (s) 21.5111 +time/total (s) 1222.1 +Epoch -947 +------------------------------ ---------------- +2022-05-15 18:23:09.183708 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -946 finished +------------------------------ ---------------- +epoch -946 +replay_buffer/size 999047 +trainer/num train calls 55000 +trainer/QF1 Loss 0.759317 +trainer/QF2 Loss 0.837911 +trainer/Policy Loss 31.6675 +trainer/Q1 Predictions Mean -68.908 +trainer/Q1 Predictions Std 18.3184 +trainer/Q1 Predictions Max -0.665323 +trainer/Q1 Predictions Min -87.591 +trainer/Q2 Predictions Mean -68.8619 +trainer/Q2 Predictions Std 18.3246 +trainer/Q2 Predictions Max -0.805124 +trainer/Q2 Predictions Min -87.5726 +trainer/Q Targets Mean -68.8899 +trainer/Q Targets Std 18.4811 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2283 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00752773 +trainer/policy/mean Std 0.670861 +trainer/policy/mean Max 0.995755 +trainer/policy/mean Min -0.993059 +trainer/policy/std Mean 0.465708 +trainer/policy/std Std 0.0255748 +trainer/policy/std Max 0.492755 +trainer/policy/std Min 0.42389 +trainer/Advantage Weights Mean 5.90826 +trainer/Advantage Weights Std 18.8175 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.27615e-15 +trainer/Advantage Score Mean -0.234825 +trainer/Advantage Score Std 0.529354 +trainer/Advantage Score Max 1.11465 +trainer/Advantage Score Min -3.42949 +trainer/V1 Predictions Mean -68.6437 +trainer/V1 Predictions Std 18.5972 +trainer/V1 Predictions Max -0.54722 +trainer/V1 Predictions Min -87.7696 +trainer/VF Loss 0.0553202 +expl/num steps total 55000 +expl/num paths total 55 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.027299 +expl/Actions Std 0.770348 +expl/Actions Max 2.59947 +expl/Actions Min -2.47467 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 55000 +eval/num paths total 55 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.239365 +eval/Actions Std 0.738104 +eval/Actions Max 0.99272 +eval/Actions Min -0.996942 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.24938e-06 +time/evaluation sampling (s) 2.98104 +time/exploration sampling (s) 3.90022 +time/logging (s) 0.00799558 +time/saving (s) 0.010609 +time/training (s) 14.4417 +time/epoch (s) 21.3416 +time/total (s) 1243.45 +Epoch -946 +------------------------------ ---------------- +2022-05-15 18:23:30.664627 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -945 finished +------------------------------ ---------------- +epoch -945 +replay_buffer/size 999047 +trainer/num train calls 56000 +trainer/QF1 Loss 0.773469 +trainer/QF2 Loss 0.845436 +trainer/Policy Loss 13.1 +trainer/Q1 Predictions Mean -70.0404 +trainer/Q1 Predictions Std 17.5973 +trainer/Q1 Predictions Max -2.71311 +trainer/Q1 Predictions Min -86.576 +trainer/Q2 Predictions Mean -70.0309 +trainer/Q2 Predictions Std 17.6561 +trainer/Q2 Predictions Max -2.64959 +trainer/Q2 Predictions Min -86.846 +trainer/Q Targets Mean -70.1126 +trainer/Q Targets Std 17.6103 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.556 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0123423 +trainer/policy/mean Std 0.672038 +trainer/policy/mean Max 0.995181 +trainer/policy/mean Min -0.986551 +trainer/policy/std Mean 0.464005 +trainer/policy/std Std 0.0256097 +trainer/policy/std Max 0.492365 +trainer/policy/std Min 0.425227 +trainer/Advantage Weights Mean 2.81371 +trainer/Advantage Weights Std 13.2259 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.7041e-11 +trainer/Advantage Score Mean -0.374547 +trainer/Advantage Score Std 0.443946 +trainer/Advantage Score Max 0.62105 +trainer/Advantage Score Min -2.30559 +trainer/V1 Predictions Mean -69.8894 +trainer/V1 Predictions Std 17.5924 +trainer/V1 Predictions Max -3.50378 +trainer/V1 Predictions Min -86.7623 +trainer/VF Loss 0.0415427 +expl/num steps total 56000 +expl/num paths total 56 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0242653 +expl/Actions Std 0.788395 +expl/Actions Max 2.44854 +expl/Actions Min -2.41325 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 56000 +eval/num paths total 56 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0428745 +eval/Actions Std 0.67132 +eval/Actions Max 0.997058 +eval/Actions Min -0.994867 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.03797e-06 +time/evaluation sampling (s) 2.86882 +time/exploration sampling (s) 3.76862 +time/logging (s) 0.00818492 +time/saving (s) 0.0110986 +time/training (s) 14.8191 +time/epoch (s) 21.4758 +time/total (s) 1264.93 +Epoch -945 +------------------------------ ---------------- +2022-05-15 18:23:52.229551 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -944 finished +------------------------------ ---------------- +epoch -944 +replay_buffer/size 999047 +trainer/num train calls 57000 +trainer/QF1 Loss 0.990784 +trainer/QF2 Loss 1.05272 +trainer/Policy Loss 30.2955 +trainer/Q1 Predictions Mean -69.8055 +trainer/Q1 Predictions Std 16.6948 +trainer/Q1 Predictions Max -1.95298 +trainer/Q1 Predictions Min -86.34 +trainer/Q2 Predictions Mean -69.7931 +trainer/Q2 Predictions Std 16.7441 +trainer/Q2 Predictions Max -1.45284 +trainer/Q2 Predictions Min -86.3321 +trainer/Q Targets Mean -70.281 +trainer/Q Targets Std 16.3838 +trainer/Q Targets Max -4.72072 +trainer/Q Targets Min -86.0336 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00999158 +trainer/policy/mean Std 0.667204 +trainer/policy/mean Max 0.996321 +trainer/policy/mean Min -0.992494 +trainer/policy/std Mean 0.464594 +trainer/policy/std Std 0.0250239 +trainer/policy/std Max 0.49325 +trainer/policy/std Min 0.425778 +trainer/Advantage Weights Mean 5.22307 +trainer/Advantage Weights Std 19.7459 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.04615e-09 +trainer/Advantage Score Mean -0.294544 +trainer/Advantage Score Std 0.428998 +trainer/Advantage Score Max 1.62721 +trainer/Advantage Score Min -2.06781 +trainer/V1 Predictions Mean -70.0202 +trainer/V1 Predictions Std 16.6324 +trainer/V1 Predictions Max -0.199039 +trainer/V1 Predictions Min -85.6393 +trainer/VF Loss 0.0516608 +expl/num steps total 57000 +expl/num paths total 57 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0419999 +expl/Actions Std 0.806935 +expl/Actions Max 2.54469 +expl/Actions Min -2.49141 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 57000 +eval/num paths total 57 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0610144 +eval/Actions Std 0.674277 +eval/Actions Max 0.998769 +eval/Actions Min -0.996134 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.63005e-06 +time/evaluation sampling (s) 3.02659 +time/exploration sampling (s) 3.72535 +time/logging (s) 0.0101975 +time/saving (s) 0.0141257 +time/training (s) 14.785 +time/epoch (s) 21.5613 +time/total (s) 1286.5 +Epoch -944 +------------------------------ ---------------- +2022-05-15 18:24:13.934728 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -943 finished +------------------------------ ---------------- +epoch -943 +replay_buffer/size 999047 +trainer/num train calls 58000 +trainer/QF1 Loss 1.3165 +trainer/QF2 Loss 1.28186 +trainer/Policy Loss 0.456111 +trainer/Q1 Predictions Mean -68.6142 +trainer/Q1 Predictions Std 19.3789 +trainer/Q1 Predictions Max -1.62969 +trainer/Q1 Predictions Min -85.8693 +trainer/Q2 Predictions Mean -68.5502 +trainer/Q2 Predictions Std 19.3551 +trainer/Q2 Predictions Max -1.92613 +trainer/Q2 Predictions Min -85.5431 +trainer/Q Targets Mean -67.9059 +trainer/Q Targets Std 19.7234 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7726 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0240155 +trainer/policy/mean Std 0.68738 +trainer/policy/mean Max 0.998991 +trainer/policy/mean Min -0.994548 +trainer/policy/std Mean 0.462558 +trainer/policy/std Std 0.0254698 +trainer/policy/std Max 0.495235 +trainer/policy/std Min 0.424827 +trainer/Advantage Weights Mean 0.123322 +trainer/Advantage Weights Std 0.74272 +trainer/Advantage Weights Max 9.80476 +trainer/Advantage Weights Min 3.27362e-13 +trainer/Advantage Score Mean -0.760651 +trainer/Advantage Score Std 0.520568 +trainer/Advantage Score Max 0.228287 +trainer/Advantage Score Min -2.87477 +trainer/V1 Predictions Mean -67.6322 +trainer/V1 Predictions Std 19.7061 +trainer/V1 Predictions Max -0.916124 +trainer/V1 Predictions Min -85.0012 +trainer/VF Loss 0.0852688 +expl/num steps total 58000 +expl/num paths total 58 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0355668 +expl/Actions Std 0.756694 +expl/Actions Max 2.45761 +expl/Actions Min -2.33749 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 58000 +eval/num paths total 58 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.218821 +eval/Actions Std 0.641495 +eval/Actions Max 0.995233 +eval/Actions Min -0.995939 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.77162e-06 +time/evaluation sampling (s) 2.96426 +time/exploration sampling (s) 3.82478 +time/logging (s) 0.00703665 +time/saving (s) 0.00947641 +time/training (s) 14.8884 +time/epoch (s) 21.694 +time/total (s) 1308.2 +Epoch -943 +------------------------------ ---------------- +2022-05-15 18:24:36.001525 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -942 finished +------------------------------ ---------------- +epoch -942 +replay_buffer/size 999047 +trainer/num train calls 59000 +trainer/QF1 Loss 0.733427 +trainer/QF2 Loss 0.794711 +trainer/Policy Loss 29.3002 +trainer/Q1 Predictions Mean -68.0217 +trainer/Q1 Predictions Std 19.9893 +trainer/Q1 Predictions Max -1.91788 +trainer/Q1 Predictions Min -87.6357 +trainer/Q2 Predictions Mean -68.0338 +trainer/Q2 Predictions Std 20.0476 +trainer/Q2 Predictions Max -3.39712 +trainer/Q2 Predictions Min -87.9345 +trainer/Q Targets Mean -68.1087 +trainer/Q Targets Std 19.8467 +trainer/Q Targets Max -3.15237 +trainer/Q Targets Min -86.9172 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.025755 +trainer/policy/mean Std 0.684338 +trainer/policy/mean Max 0.997612 +trainer/policy/mean Min -0.997419 +trainer/policy/std Mean 0.461384 +trainer/policy/std Std 0.0248548 +trainer/policy/std Max 0.493992 +trainer/policy/std Min 0.422768 +trainer/Advantage Weights Mean 4.95711 +trainer/Advantage Weights Std 17.3287 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.89992e-11 +trainer/Advantage Score Mean -0.236724 +trainer/Advantage Score Std 0.452796 +trainer/Advantage Score Max 0.988041 +trainer/Advantage Score Min -2.46866 +trainer/V1 Predictions Mean -67.757 +trainer/V1 Predictions Std 20.1091 +trainer/V1 Predictions Max -1.94957 +trainer/V1 Predictions Min -86.1841 +trainer/VF Loss 0.0445435 +expl/num steps total 59000 +expl/num paths total 59 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0372876 +expl/Actions Std 0.773156 +expl/Actions Max 2.44807 +expl/Actions Min -2.56061 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 59000 +eval/num paths total 59 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0158803 +eval/Actions Std 0.698926 +eval/Actions Max 0.997451 +eval/Actions Min -0.998366 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.08618e-06 +time/evaluation sampling (s) 2.8043 +time/exploration sampling (s) 3.83894 +time/logging (s) 0.00889593 +time/saving (s) 0.0119645 +time/training (s) 15.3982 +time/epoch (s) 22.0623 +time/total (s) 1330.26 +Epoch -942 +------------------------------ ---------------- +2022-05-15 18:24:57.701259 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -941 finished +------------------------------ ---------------- +epoch -941 +replay_buffer/size 999047 +trainer/num train calls 60000 +trainer/QF1 Loss 0.776251 +trainer/QF2 Loss 0.742859 +trainer/Policy Loss 6.43656 +trainer/Q1 Predictions Mean -68.014 +trainer/Q1 Predictions Std 19.567 +trainer/Q1 Predictions Max -1.17267 +trainer/Q1 Predictions Min -87.129 +trainer/Q2 Predictions Mean -68.037 +trainer/Q2 Predictions Std 19.5694 +trainer/Q2 Predictions Max -1.91051 +trainer/Q2 Predictions Min -87.0971 +trainer/Q Targets Mean -67.7407 +trainer/Q Targets Std 19.6775 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1325 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00652764 +trainer/policy/mean Std 0.669872 +trainer/policy/mean Max 0.99828 +trainer/policy/mean Min -0.99498 +trainer/policy/std Mean 0.459923 +trainer/policy/std Std 0.0252813 +trainer/policy/std Max 0.491677 +trainer/policy/std Min 0.420463 +trainer/Advantage Weights Mean 0.831901 +trainer/Advantage Weights Std 8.34476 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.24341e-16 +trainer/Advantage Score Mean -0.771613 +trainer/Advantage Score Std 0.555859 +trainer/Advantage Score Max 1.13915 +trainer/Advantage Score Min -3.60334 +trainer/V1 Predictions Mean -67.4397 +trainer/V1 Predictions Std 19.6968 +trainer/V1 Predictions Max -0.903774 +trainer/V1 Predictions Min -86.9575 +trainer/VF Loss 0.0953723 +expl/num steps total 60000 +expl/num paths total 60 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0500342 +expl/Actions Std 0.797049 +expl/Actions Max 2.38892 +expl/Actions Min -2.23245 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 60000 +eval/num paths total 60 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0159068 +eval/Actions Std 0.636305 +eval/Actions Max 0.998839 +eval/Actions Min -0.996499 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78326e-06 +time/evaluation sampling (s) 2.97067 +time/exploration sampling (s) 3.97389 +time/logging (s) 0.00681486 +time/saving (s) 0.0150687 +time/training (s) 14.7263 +time/epoch (s) 21.6927 +time/total (s) 1351.96 +Epoch -941 +------------------------------ ---------------- +2022-05-15 18:25:19.162224 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -940 finished +------------------------------ ---------------- +epoch -940 +replay_buffer/size 999047 +trainer/num train calls 61000 +trainer/QF1 Loss 0.900043 +trainer/QF2 Loss 0.951603 +trainer/Policy Loss 4.29625 +trainer/Q1 Predictions Mean -68.5749 +trainer/Q1 Predictions Std 19.2805 +trainer/Q1 Predictions Max -2.64417 +trainer/Q1 Predictions Min -86.834 +trainer/Q2 Predictions Mean -68.6486 +trainer/Q2 Predictions Std 19.2327 +trainer/Q2 Predictions Max -3.67405 +trainer/Q2 Predictions Min -86.7693 +trainer/Q Targets Mean -68.3496 +trainer/Q Targets Std 19.6721 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6241 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.015872 +trainer/policy/mean Std 0.682668 +trainer/policy/mean Max 0.995619 +trainer/policy/mean Min -0.994211 +trainer/policy/std Mean 0.460673 +trainer/policy/std Std 0.0255099 +trainer/policy/std Max 0.491464 +trainer/policy/std Min 0.418072 +trainer/Advantage Weights Mean 0.94487 +trainer/Advantage Weights Std 7.8583 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.20423e-14 +trainer/Advantage Score Mean -0.660116 +trainer/Advantage Score Std 0.548139 +trainer/Advantage Score Max 0.829548 +trainer/Advantage Score Min -3.14458 +trainer/V1 Predictions Mean -68.0412 +trainer/V1 Predictions Std 19.827 +trainer/V1 Predictions Max -1.07492 +trainer/V1 Predictions Min -86.9603 +trainer/VF Loss 0.0771556 +expl/num steps total 61000 +expl/num paths total 61 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0641972 +expl/Actions Std 0.80423 +expl/Actions Max 2.4234 +expl/Actions Min -2.4498 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 61000 +eval/num paths total 61 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0283012 +eval/Actions Std 0.643482 +eval/Actions Max 0.99823 +eval/Actions Min -0.997423 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.24869e-06 +time/evaluation sampling (s) 2.81313 +time/exploration sampling (s) 3.84417 +time/logging (s) 0.0099351 +time/saving (s) 0.0134835 +time/training (s) 14.7783 +time/epoch (s) 21.4591 +time/total (s) 1373.42 +Epoch -940 +------------------------------ ---------------- +2022-05-15 18:25:40.740374 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -939 finished +------------------------------ ---------------- +epoch -939 +replay_buffer/size 999047 +trainer/num train calls 62000 +trainer/QF1 Loss 0.744465 +trainer/QF2 Loss 0.790261 +trainer/Policy Loss 39.42 +trainer/Q1 Predictions Mean -70.9735 +trainer/Q1 Predictions Std 18.05 +trainer/Q1 Predictions Max -1.31928 +trainer/Q1 Predictions Min -87.9326 +trainer/Q2 Predictions Mean -70.8981 +trainer/Q2 Predictions Std 18.1192 +trainer/Q2 Predictions Max -1.599 +trainer/Q2 Predictions Min -88.2265 +trainer/Q Targets Mean -71.2715 +trainer/Q Targets Std 17.8284 +trainer/Q Targets Max -2.50983 +trainer/Q Targets Min -87.8509 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0107511 +trainer/policy/mean Std 0.676407 +trainer/policy/mean Max 0.998967 +trainer/policy/mean Min -0.997111 +trainer/policy/std Mean 0.461505 +trainer/policy/std Std 0.0244164 +trainer/policy/std Max 0.492343 +trainer/policy/std Min 0.422229 +trainer/Advantage Weights Mean 6.99096 +trainer/Advantage Weights Std 22.652 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.20588e-14 +trainer/Advantage Score Mean -0.40433 +trainer/Advantage Score Std 0.545847 +trainer/Advantage Score Max 0.92939 +trainer/Advantage Score Min -3.01313 +trainer/V1 Predictions Mean -70.9381 +trainer/V1 Predictions Std 17.9961 +trainer/V1 Predictions Max -1.49075 +trainer/V1 Predictions Min -87.7782 +trainer/VF Loss 0.0690806 +expl/num steps total 62000 +expl/num paths total 62 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0129235 +expl/Actions Std 0.792408 +expl/Actions Max 2.43074 +expl/Actions Min -2.83249 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 62000 +eval/num paths total 62 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0373856 +eval/Actions Std 0.712397 +eval/Actions Max 0.997194 +eval/Actions Min -0.997027 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.94298e-06 +time/evaluation sampling (s) 2.9815 +time/exploration sampling (s) 4.11172 +time/logging (s) 0.00725594 +time/saving (s) 0.010127 +time/training (s) 14.4565 +time/epoch (s) 21.5671 +time/total (s) 1395 +Epoch -939 +------------------------------ ---------------- +2022-05-15 18:26:02.665751 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -938 finished +------------------------------ ---------------- +epoch -938 +replay_buffer/size 999047 +trainer/num train calls 63000 +trainer/QF1 Loss 1.89261 +trainer/QF2 Loss 1.99706 +trainer/Policy Loss 206.606 +trainer/Q1 Predictions Mean -70.8379 +trainer/Q1 Predictions Std 16.7123 +trainer/Q1 Predictions Max -1.14711 +trainer/Q1 Predictions Min -86.9347 +trainer/Q2 Predictions Mean -70.795 +trainer/Q2 Predictions Std 16.6353 +trainer/Q2 Predictions Max -1.49373 +trainer/Q2 Predictions Min -87.1816 +trainer/Q Targets Mean -71.9721 +trainer/Q Targets Std 16.6012 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6779 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0360992 +trainer/policy/mean Std 0.679055 +trainer/policy/mean Max 0.996386 +trainer/policy/mean Min -0.997583 +trainer/policy/std Mean 0.460744 +trainer/policy/std Std 0.0238453 +trainer/policy/std Max 0.489005 +trainer/policy/std Min 0.421367 +trainer/Advantage Weights Mean 36.7998 +trainer/Advantage Weights Std 42.5691 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.10408e-13 +trainer/Advantage Score Mean 0.215347 +trainer/Advantage Score Std 0.476258 +trainer/Advantage Score Max 1.76039 +trainer/Advantage Score Min -2.88009 +trainer/V1 Predictions Mean -71.7184 +trainer/V1 Predictions Std 16.6399 +trainer/V1 Predictions Max -2.81945 +trainer/V1 Predictions Min -87.9358 +trainer/VF Loss 0.190375 +expl/num steps total 63000 +expl/num paths total 63 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00654715 +expl/Actions Std 0.818742 +expl/Actions Max 2.27198 +expl/Actions Min -2.37054 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 63000 +eval/num paths total 63 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0273265 +eval/Actions Std 0.682709 +eval/Actions Max 0.997869 +eval/Actions Min -0.998512 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.49222e-06 +time/evaluation sampling (s) 2.92297 +time/exploration sampling (s) 4.27073 +time/logging (s) 0.0105704 +time/saving (s) 0.0172916 +time/training (s) 14.7015 +time/epoch (s) 21.9231 +time/total (s) 1416.93 +Epoch -938 +------------------------------ ---------------- +2022-05-15 18:26:23.639231 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -937 finished +------------------------------ ---------------- +epoch -937 +replay_buffer/size 999047 +trainer/num train calls 64000 +trainer/QF1 Loss 1.09964 +trainer/QF2 Loss 1.20452 +trainer/Policy Loss 7.90918 +trainer/Q1 Predictions Mean -69.4328 +trainer/Q1 Predictions Std 19.908 +trainer/Q1 Predictions Max -1.95214 +trainer/Q1 Predictions Min -88.1848 +trainer/Q2 Predictions Mean -69.4953 +trainer/Q2 Predictions Std 19.9674 +trainer/Q2 Predictions Max -2.4529 +trainer/Q2 Predictions Min -89.1112 +trainer/Q Targets Mean -68.951 +trainer/Q Targets Std 19.7055 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.8125 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0137694 +trainer/policy/mean Std 0.684842 +trainer/policy/mean Max 0.995422 +trainer/policy/mean Min -0.995042 +trainer/policy/std Mean 0.46224 +trainer/policy/std Std 0.0239134 +trainer/policy/std Max 0.48941 +trainer/policy/std Min 0.424911 +trainer/Advantage Weights Mean 1.35913 +trainer/Advantage Weights Std 9.87021 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.45258e-16 +trainer/Advantage Score Mean -0.655538 +trainer/Advantage Score Std 0.499031 +trainer/Advantage Score Max 0.810127 +trainer/Advantage Score Min -3.48328 +trainer/V1 Predictions Mean -68.5849 +trainer/V1 Predictions Std 19.9207 +trainer/V1 Predictions Max -1.47777 +trainer/V1 Predictions Min -86.7432 +trainer/VF Loss 0.0724441 +expl/num steps total 64000 +expl/num paths total 64 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.015923 +expl/Actions Std 0.806142 +expl/Actions Max 2.71405 +expl/Actions Min -2.35327 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 64000 +eval/num paths total 64 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0184677 +eval/Actions Std 0.706516 +eval/Actions Max 0.996421 +eval/Actions Min -0.995694 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78093e-06 +time/evaluation sampling (s) 3.00488 +time/exploration sampling (s) 3.39999 +time/logging (s) 0.00798454 +time/saving (s) 0.0121883 +time/training (s) 14.5366 +time/epoch (s) 20.9616 +time/total (s) 1437.89 +Epoch -937 +------------------------------ ---------------- +2022-05-15 18:26:44.639188 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -936 finished +------------------------------ ---------------- +epoch -936 +replay_buffer/size 999047 +trainer/num train calls 65000 +trainer/QF1 Loss 2.61969 +trainer/QF2 Loss 2.67948 +trainer/Policy Loss 5.47894 +trainer/Q1 Predictions Mean -71.4454 +trainer/Q1 Predictions Std 18.6563 +trainer/Q1 Predictions Max -3.12398 +trainer/Q1 Predictions Min -87.5046 +trainer/Q2 Predictions Mean -71.4936 +trainer/Q2 Predictions Std 18.6238 +trainer/Q2 Predictions Max -3.14025 +trainer/Q2 Predictions Min -87.4855 +trainer/Q Targets Mean -70.7831 +trainer/Q Targets Std 18.7587 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1145 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0198059 +trainer/policy/mean Std 0.681331 +trainer/policy/mean Max 0.998284 +trainer/policy/mean Min -0.995174 +trainer/policy/std Mean 0.462144 +trainer/policy/std Std 0.0240208 +trainer/policy/std Max 0.493052 +trainer/policy/std Min 0.42221 +trainer/Advantage Weights Mean 1.21125 +trainer/Advantage Weights Std 9.13048 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.50112e-09 +trainer/Advantage Score Mean -0.499356 +trainer/Advantage Score Std 0.388615 +trainer/Advantage Score Max 1.15367 +trainer/Advantage Score Min -1.98065 +trainer/V1 Predictions Mean -70.5611 +trainer/V1 Predictions Std 18.6593 +trainer/V1 Predictions Max -2.96769 +trainer/V1 Predictions Min -86.2084 +trainer/VF Loss 0.0470991 +expl/num steps total 65000 +expl/num paths total 65 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0322591 +expl/Actions Std 0.791868 +expl/Actions Max 2.36779 +expl/Actions Min -2.51744 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 65000 +eval/num paths total 65 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0241348 +eval/Actions Std 0.697764 +eval/Actions Max 0.997967 +eval/Actions Min -0.996064 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.42405e-06 +time/evaluation sampling (s) 2.63887 +time/exploration sampling (s) 3.41684 +time/logging (s) 0.00783281 +time/saving (s) 0.0118775 +time/training (s) 14.9175 +time/epoch (s) 20.993 +time/total (s) 1458.89 +Epoch -936 +------------------------------ ---------------- +2022-05-15 18:27:08.034530 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -935 finished +------------------------------ ---------------- +epoch -935 +replay_buffer/size 999047 +trainer/num train calls 66000 +trainer/QF1 Loss 1.13549 +trainer/QF2 Loss 1.23372 +trainer/Policy Loss 2.54626 +trainer/Q1 Predictions Mean -69.3275 +trainer/Q1 Predictions Std 19.1623 +trainer/Q1 Predictions Max -2.0207 +trainer/Q1 Predictions Min -87.512 +trainer/Q2 Predictions Mean -69.3897 +trainer/Q2 Predictions Std 19.1526 +trainer/Q2 Predictions Max -2.19547 +trainer/Q2 Predictions Min -87.5277 +trainer/Q Targets Mean -68.7025 +trainer/Q Targets Std 19.5231 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4716 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00390367 +trainer/policy/mean Std 0.675301 +trainer/policy/mean Max 0.998119 +trainer/policy/mean Min -0.997266 +trainer/policy/std Mean 0.460509 +trainer/policy/std Std 0.0238219 +trainer/policy/std Max 0.490468 +trainer/policy/std Min 0.419144 +trainer/Advantage Weights Mean 0.172556 +trainer/Advantage Weights Std 1.53653 +trainer/Advantage Weights Max 24.1739 +trainer/Advantage Weights Min 1.25233e-14 +trainer/Advantage Score Mean -0.746242 +trainer/Advantage Score Std 0.500086 +trainer/Advantage Score Max 0.318527 +trainer/Advantage Score Min -3.20112 +trainer/V1 Predictions Mean -68.4529 +trainer/V1 Predictions Std 19.4505 +trainer/V1 Predictions Max -0.465866 +trainer/V1 Predictions Min -87.2904 +trainer/VF Loss 0.0810992 +expl/num steps total 66000 +expl/num paths total 66 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0109894 +expl/Actions Std 0.782849 +expl/Actions Max 2.36562 +expl/Actions Min -2.69738 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 66000 +eval/num paths total 66 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.122746 +eval/Actions Std 0.682122 +eval/Actions Max 0.998198 +eval/Actions Min -0.997871 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.30316e-06 +time/evaluation sampling (s) 3.52301 +time/exploration sampling (s) 4.44976 +time/logging (s) 0.00721855 +time/saving (s) 0.0101124 +time/training (s) 15.3983 +time/epoch (s) 23.3884 +time/total (s) 1482.29 +Epoch -935 +------------------------------ ---------------- +2022-05-15 18:27:31.136989 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -934 finished +------------------------------ ---------------- +epoch -934 +replay_buffer/size 999047 +trainer/num train calls 67000 +trainer/QF1 Loss 1.32351 +trainer/QF2 Loss 1.41212 +trainer/Policy Loss 2.58728 +trainer/Q1 Predictions Mean -68.0321 +trainer/Q1 Predictions Std 20.2502 +trainer/Q1 Predictions Max -1.10734 +trainer/Q1 Predictions Min -88.0289 +trainer/Q2 Predictions Mean -68.127 +trainer/Q2 Predictions Std 20.238 +trainer/Q2 Predictions Max -2.02438 +trainer/Q2 Predictions Min -88.0124 +trainer/Q Targets Mean -68.1644 +trainer/Q Targets Std 20.7845 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5107 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00987952 +trainer/policy/mean Std 0.677388 +trainer/policy/mean Max 0.99783 +trainer/policy/mean Min -0.997078 +trainer/policy/std Mean 0.460704 +trainer/policy/std Std 0.0250888 +trainer/policy/std Max 0.492818 +trainer/policy/std Min 0.419095 +trainer/Advantage Weights Mean 0.464754 +trainer/Advantage Weights Std 4.06091 +trainer/Advantage Weights Max 61.2563 +trainer/Advantage Weights Min 1.56604e-14 +trainer/Advantage Score Mean -0.736589 +trainer/Advantage Score Std 0.629046 +trainer/Advantage Score Max 0.411507 +trainer/Advantage Score Min -3.17876 +trainer/V1 Predictions Mean -67.8885 +trainer/V1 Predictions Std 20.7253 +trainer/V1 Predictions Max 0.276729 +trainer/V1 Predictions Min -86.965 +trainer/VF Loss 0.0949962 +expl/num steps total 67000 +expl/num paths total 67 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00609983 +expl/Actions Std 0.828334 +expl/Actions Max 2.68866 +expl/Actions Min -2.43158 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 67000 +eval/num paths total 67 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0788586 +eval/Actions Std 0.70142 +eval/Actions Max 0.996164 +eval/Actions Min -0.995976 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.20611e-05 +time/evaluation sampling (s) 3.33479 +time/exploration sampling (s) 4.16985 +time/logging (s) 0.012231 +time/saving (s) 0.0184782 +time/training (s) 15.5663 +time/epoch (s) 23.1017 +time/total (s) 1505.39 +Epoch -934 +------------------------------ ---------------- +2022-05-15 18:27:53.881705 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -933 finished +------------------------------ ---------------- +epoch -933 +replay_buffer/size 999047 +trainer/num train calls 68000 +trainer/QF1 Loss 0.519937 +trainer/QF2 Loss 0.457209 +trainer/Policy Loss 39.1178 +trainer/Q1 Predictions Mean -69.7832 +trainer/Q1 Predictions Std 17.3214 +trainer/Q1 Predictions Max -2.84987 +trainer/Q1 Predictions Min -87.3689 +trainer/Q2 Predictions Mean -69.7136 +trainer/Q2 Predictions Std 17.3272 +trainer/Q2 Predictions Max -3.21371 +trainer/Q2 Predictions Min -87.191 +trainer/Q Targets Mean -69.6492 +trainer/Q Targets Std 17.4267 +trainer/Q Targets Max -2.44567 +trainer/Q Targets Min -87.4078 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0243394 +trainer/policy/mean Std 0.681197 +trainer/policy/mean Max 0.998979 +trainer/policy/mean Min -0.991653 +trainer/policy/std Mean 0.460678 +trainer/policy/std Std 0.0247724 +trainer/policy/std Max 0.491443 +trainer/policy/std Min 0.421193 +trainer/Advantage Weights Mean 5.57517 +trainer/Advantage Weights Std 20.8341 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.52599e-13 +trainer/Advantage Score Mean -0.374733 +trainer/Advantage Score Std 0.513306 +trainer/Advantage Score Max 0.863077 +trainer/Advantage Score Min -2.82241 +trainer/V1 Predictions Mean -69.3197 +trainer/V1 Predictions Std 17.6072 +trainer/V1 Predictions Max -2.30157 +trainer/V1 Predictions Min -87.1948 +trainer/VF Loss 0.0584622 +expl/num steps total 68000 +expl/num paths total 68 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0184225 +expl/Actions Std 0.820445 +expl/Actions Max 2.4872 +expl/Actions Min -2.37373 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 68000 +eval/num paths total 68 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.105176 +eval/Actions Std 0.674549 +eval/Actions Max 0.998432 +eval/Actions Min -0.988733 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.81073e-06 +time/evaluation sampling (s) 3.31843 +time/exploration sampling (s) 4.04148 +time/logging (s) 0.00800017 +time/saving (s) 0.00965385 +time/training (s) 15.3536 +time/epoch (s) 22.7312 +time/total (s) 1528.13 +Epoch -933 +------------------------------ ---------------- +2022-05-15 18:28:16.317232 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -932 finished +------------------------------ ---------------- +epoch -932 +replay_buffer/size 999047 +trainer/num train calls 69000 +trainer/QF1 Loss 0.8463 +trainer/QF2 Loss 0.807408 +trainer/Policy Loss 17.2743 +trainer/Q1 Predictions Mean -68.4891 +trainer/Q1 Predictions Std 19.8243 +trainer/Q1 Predictions Max -1.79755 +trainer/Q1 Predictions Min -89.4617 +trainer/Q2 Predictions Mean -68.46 +trainer/Q2 Predictions Std 19.8588 +trainer/Q2 Predictions Max -2.68455 +trainer/Q2 Predictions Min -89.314 +trainer/Q Targets Mean -68.0104 +trainer/Q Targets Std 19.6492 +trainer/Q Targets Max -4.69257 +trainer/Q Targets Min -88.1755 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0294145 +trainer/policy/mean Std 0.683066 +trainer/policy/mean Max 0.996489 +trainer/policy/mean Min -0.99454 +trainer/policy/std Mean 0.457817 +trainer/policy/std Std 0.0245628 +trainer/policy/std Max 0.488265 +trainer/policy/std Min 0.416505 +trainer/Advantage Weights Mean 2.29501 +trainer/Advantage Weights Std 12.2803 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.85817e-13 +trainer/Advantage Score Mean -0.475565 +trainer/Advantage Score Std 0.528467 +trainer/Advantage Score Max 1.30921 +trainer/Advantage Score Min -2.77523 +trainer/V1 Predictions Mean -67.663 +trainer/V1 Predictions Std 20.0174 +trainer/V1 Predictions Max -1.01952 +trainer/V1 Predictions Min -88.3568 +trainer/VF Loss 0.0614286 +expl/num steps total 69000 +expl/num paths total 69 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00713311 +expl/Actions Std 0.796275 +expl/Actions Max 2.90015 +expl/Actions Min -2.49537 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 69000 +eval/num paths total 69 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.243781 +eval/Actions Std 0.67542 +eval/Actions Max 0.995412 +eval/Actions Min -0.995204 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.88896e-06 +time/evaluation sampling (s) 3.23928 +time/exploration sampling (s) 4.293 +time/logging (s) 0.0127723 +time/saving (s) 0.0159388 +time/training (s) 14.8736 +time/epoch (s) 22.4346 +time/total (s) 1550.57 +Epoch -932 +------------------------------ ---------------- +2022-05-15 18:28:38.966013 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -931 finished +------------------------------ ---------------- +epoch -931 +replay_buffer/size 999047 +trainer/num train calls 70000 +trainer/QF1 Loss 0.967829 +trainer/QF2 Loss 0.936324 +trainer/Policy Loss 27.0972 +trainer/Q1 Predictions Mean -71.2361 +trainer/Q1 Predictions Std 18.1933 +trainer/Q1 Predictions Max -2.56861 +trainer/Q1 Predictions Min -86.586 +trainer/Q2 Predictions Mean -71.2859 +trainer/Q2 Predictions Std 18.175 +trainer/Q2 Predictions Max -2.82299 +trainer/Q2 Predictions Min -86.912 +trainer/Q Targets Mean -71.7064 +trainer/Q Targets Std 18.2352 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1478 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00524656 +trainer/policy/mean Std 0.677665 +trainer/policy/mean Max 0.999255 +trainer/policy/mean Min -0.993832 +trainer/policy/std Mean 0.457935 +trainer/policy/std Std 0.0244848 +trainer/policy/std Max 0.487425 +trainer/policy/std Min 0.416866 +trainer/Advantage Weights Mean 4.59915 +trainer/Advantage Weights Std 17.485 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.70973e-07 +trainer/Advantage Score Mean -0.249131 +trainer/Advantage Score Std 0.381073 +trainer/Advantage Score Max 1.16549 +trainer/Advantage Score Min -1.55818 +trainer/V1 Predictions Mean -71.5257 +trainer/V1 Predictions Std 18.0472 +trainer/V1 Predictions Max -3.19891 +trainer/V1 Predictions Min -87.1177 +trainer/VF Loss 0.0405237 +expl/num steps total 70000 +expl/num paths total 70 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0183908 +expl/Actions Std 0.786762 +expl/Actions Max 2.23673 +expl/Actions Min -2.53325 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 70000 +eval/num paths total 70 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0247819 +eval/Actions Std 0.687102 +eval/Actions Max 0.997854 +eval/Actions Min -0.997973 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.15696e-06 +time/evaluation sampling (s) 3.43189 +time/exploration sampling (s) 4.15784 +time/logging (s) 0.0116469 +time/saving (s) 0.0171836 +time/training (s) 15.0204 +time/epoch (s) 22.639 +time/total (s) 1573.21 +Epoch -931 +------------------------------ ---------------- +2022-05-15 18:29:01.771114 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -930 finished +------------------------------ ---------------- +epoch -930 +replay_buffer/size 999047 +trainer/num train calls 71000 +trainer/QF1 Loss 0.656957 +trainer/QF2 Loss 0.712054 +trainer/Policy Loss 4.26219 +trainer/Q1 Predictions Mean -69.6523 +trainer/Q1 Predictions Std 18.2887 +trainer/Q1 Predictions Max -3.35065 +trainer/Q1 Predictions Min -87.1469 +trainer/Q2 Predictions Mean -69.7043 +trainer/Q2 Predictions Std 18.2562 +trainer/Q2 Predictions Max -3.63876 +trainer/Q2 Predictions Min -87.0695 +trainer/Q Targets Mean -69.302 +trainer/Q Targets Std 18.4943 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1156 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00273647 +trainer/policy/mean Std 0.673651 +trainer/policy/mean Max 0.997262 +trainer/policy/mean Min -0.996663 +trainer/policy/std Mean 0.456215 +trainer/policy/std Std 0.0243331 +trainer/policy/std Max 0.485106 +trainer/policy/std Min 0.41528 +trainer/Advantage Weights Mean 0.846853 +trainer/Advantage Weights Std 6.92125 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.79487e-16 +trainer/Advantage Score Mean -0.608651 +trainer/Advantage Score Std 0.567916 +trainer/Advantage Score Max 0.512669 +trainer/Advantage Score Min -3.49252 +trainer/V1 Predictions Mean -68.9787 +trainer/V1 Predictions Std 18.7041 +trainer/V1 Predictions Max -2.58398 +trainer/V1 Predictions Min -86.9311 +trainer/VF Loss 0.0713225 +expl/num steps total 71000 +expl/num paths total 71 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.000173418 +expl/Actions Std 0.784927 +expl/Actions Max 2.71348 +expl/Actions Min -2.37381 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 71000 +eval/num paths total 71 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0405575 +eval/Actions Std 0.713198 +eval/Actions Max 0.996705 +eval/Actions Min -0.999355 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.91295e-06 +time/evaluation sampling (s) 3.20339 +time/exploration sampling (s) 4.15351 +time/logging (s) 0.0110115 +time/saving (s) 0.0143419 +time/training (s) 15.4135 +time/epoch (s) 22.7958 +time/total (s) 1596.02 +Epoch -930 +------------------------------ ---------------- +2022-05-15 18:29:24.506895 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -929 finished +------------------------------ ---------------- +epoch -929 +replay_buffer/size 999047 +trainer/num train calls 72000 +trainer/QF1 Loss 2.09145 +trainer/QF2 Loss 2.15728 +trainer/Policy Loss 9.59926 +trainer/Q1 Predictions Mean -69.7851 +trainer/Q1 Predictions Std 19.4047 +trainer/Q1 Predictions Max -1.06203 +trainer/Q1 Predictions Min -89.3497 +trainer/Q2 Predictions Mean -69.7977 +trainer/Q2 Predictions Std 19.4051 +trainer/Q2 Predictions Max -1.12772 +trainer/Q2 Predictions Min -89.5282 +trainer/Q Targets Mean -69.401 +trainer/Q Targets Std 19.3257 +trainer/Q Targets Max 0 +trainer/Q Targets Min -89.2278 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.027411 +trainer/policy/mean Std 0.673635 +trainer/policy/mean Max 0.999195 +trainer/policy/mean Min -0.994318 +trainer/policy/std Mean 0.455451 +trainer/policy/std Std 0.0253097 +trainer/policy/std Max 0.482944 +trainer/policy/std Min 0.410425 +trainer/Advantage Weights Mean 1.69608 +trainer/Advantage Weights Std 10.9926 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.04379e-14 +trainer/Advantage Score Mean -0.621271 +trainer/Advantage Score Std 0.482211 +trainer/Advantage Score Max 0.975174 +trainer/Advantage Score Min -3.01513 +trainer/V1 Predictions Mean -69.0703 +trainer/V1 Predictions Std 19.4389 +trainer/V1 Predictions Max -0.818696 +trainer/V1 Predictions Min -89.4967 +trainer/VF Loss 0.0684311 +expl/num steps total 72000 +expl/num paths total 72 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0521566 +expl/Actions Std 0.819345 +expl/Actions Max 2.63359 +expl/Actions Min -2.21274 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 72000 +eval/num paths total 72 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.199729 +eval/Actions Std 0.725029 +eval/Actions Max 0.99913 +eval/Actions Min -0.997794 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.80864e-06 +time/evaluation sampling (s) 3.32425 +time/exploration sampling (s) 3.82474 +time/logging (s) 0.00729014 +time/saving (s) 0.0103747 +time/training (s) 15.557 +time/epoch (s) 22.7236 +time/total (s) 1618.75 +Epoch -929 +------------------------------ ---------------- +2022-05-15 18:29:47.321503 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -928 finished +------------------------------ ---------------- +epoch -928 +replay_buffer/size 999047 +trainer/num train calls 73000 +trainer/QF1 Loss 1.05457 +trainer/QF2 Loss 0.884359 +trainer/Policy Loss 59.1123 +trainer/Q1 Predictions Mean -69.9713 +trainer/Q1 Predictions Std 18.9098 +trainer/Q1 Predictions Max -1.16303 +trainer/Q1 Predictions Min -86.9282 +trainer/Q2 Predictions Mean -70.0716 +trainer/Q2 Predictions Std 18.8427 +trainer/Q2 Predictions Max -1.36544 +trainer/Q2 Predictions Min -86.975 +trainer/Q Targets Mean -70.312 +trainer/Q Targets Std 18.5732 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.98 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0299219 +trainer/policy/mean Std 0.690146 +trainer/policy/mean Max 0.998937 +trainer/policy/mean Min -0.998215 +trainer/policy/std Mean 0.456843 +trainer/policy/std Std 0.0242735 +trainer/policy/std Max 0.486535 +trainer/policy/std Min 0.416266 +trainer/Advantage Weights Mean 8.60587 +trainer/Advantage Weights Std 23.7106 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.38497e-09 +trainer/Advantage Score Mean -0.148492 +trainer/Advantage Score Std 0.456851 +trainer/Advantage Score Max 1.96053 +trainer/Advantage Score Min -2.03976 +trainer/V1 Predictions Mean -70.02 +trainer/V1 Predictions Std 18.7874 +trainer/V1 Predictions Max 0.122047 +trainer/V1 Predictions Min -86.8853 +trainer/VF Loss 0.0721811 +expl/num steps total 73000 +expl/num paths total 73 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0250353 +expl/Actions Std 0.817355 +expl/Actions Max 2.22831 +expl/Actions Min -3.03328 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 73000 +eval/num paths total 73 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0115679 +eval/Actions Std 0.711679 +eval/Actions Max 0.997958 +eval/Actions Min -0.99843 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.73203e-06 +time/evaluation sampling (s) 3.16344 +time/exploration sampling (s) 4.19625 +time/logging (s) 0.0105281 +time/saving (s) 0.0151913 +time/training (s) 15.4268 +time/epoch (s) 22.8122 +time/total (s) 1641.56 +Epoch -928 +------------------------------ ---------------- +2022-05-15 18:30:09.823106 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -927 finished +------------------------------ ---------------- +epoch -927 +replay_buffer/size 999047 +trainer/num train calls 74000 +trainer/QF1 Loss 0.870063 +trainer/QF2 Loss 0.968235 +trainer/Policy Loss 6.09411 +trainer/Q1 Predictions Mean -71.0932 +trainer/Q1 Predictions Std 17.7434 +trainer/Q1 Predictions Max -1.03651 +trainer/Q1 Predictions Min -88.1477 +trainer/Q2 Predictions Mean -71.1896 +trainer/Q2 Predictions Std 17.7521 +trainer/Q2 Predictions Max -1.40454 +trainer/Q2 Predictions Min -88.2302 +trainer/Q Targets Mean -70.8013 +trainer/Q Targets Std 17.9278 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.294 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00167039 +trainer/policy/mean Std 0.677235 +trainer/policy/mean Max 0.997759 +trainer/policy/mean Min -0.996222 +trainer/policy/std Mean 0.456995 +trainer/policy/std Std 0.0237865 +trainer/policy/std Max 0.485199 +trainer/policy/std Min 0.419869 +trainer/Advantage Weights Mean 1.32064 +trainer/Advantage Weights Std 9.50412 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.56611e-16 +trainer/Advantage Score Mean -0.602955 +trainer/Advantage Score Std 0.484091 +trainer/Advantage Score Max 0.787056 +trainer/Advantage Score Min -3.5899 +trainer/V1 Predictions Mean -70.6153 +trainer/V1 Predictions Std 17.7913 +trainer/V1 Predictions Max -0.711487 +trainer/V1 Predictions Min -87.9883 +trainer/VF Loss 0.063977 +expl/num steps total 74000 +expl/num paths total 74 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.012982 +expl/Actions Std 0.805314 +expl/Actions Max 2.36934 +expl/Actions Min -2.77654 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 74000 +eval/num paths total 74 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0585848 +eval/Actions Std 0.655293 +eval/Actions Max 0.999195 +eval/Actions Min -0.998398 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.90805e-06 +time/evaluation sampling (s) 3.24657 +time/exploration sampling (s) 4.23127 +time/logging (s) 0.0118333 +time/saving (s) 0.0172726 +time/training (s) 14.988 +time/epoch (s) 22.495 +time/total (s) 1664.06 +Epoch -927 +------------------------------ ---------------- +2022-05-15 18:30:32.687683 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -926 finished +------------------------------ ---------------- +epoch -926 +replay_buffer/size 999047 +trainer/num train calls 75000 +trainer/QF1 Loss 1.05498 +trainer/QF2 Loss 1.06996 +trainer/Policy Loss 38.4857 +trainer/Q1 Predictions Mean -72.1613 +trainer/Q1 Predictions Std 15.791 +trainer/Q1 Predictions Max -5.73413 +trainer/Q1 Predictions Min -87.3664 +trainer/Q2 Predictions Mean -72.1411 +trainer/Q2 Predictions Std 15.7761 +trainer/Q2 Predictions Max -6.05017 +trainer/Q2 Predictions Min -87.1396 +trainer/Q Targets Mean -71.7424 +trainer/Q Targets Std 16.2745 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2566 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.023185 +trainer/policy/mean Std 0.673025 +trainer/policy/mean Max 0.9983 +trainer/policy/mean Min -0.997992 +trainer/policy/std Mean 0.455298 +trainer/policy/std Std 0.0233842 +trainer/policy/std Max 0.486652 +trainer/policy/std Min 0.418135 +trainer/Advantage Weights Mean 6.02223 +trainer/Advantage Weights Std 18.6199 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.7261e-13 +trainer/Advantage Score Mean -0.194155 +trainer/Advantage Score Std 0.44892 +trainer/Advantage Score Max 0.866898 +trainer/Advantage Score Min -2.81886 +trainer/V1 Predictions Mean -71.4807 +trainer/V1 Predictions Std 16.3157 +trainer/V1 Predictions Max -4.12751 +trainer/V1 Predictions Min -87.116 +trainer/VF Loss 0.0428404 +expl/num steps total 75000 +expl/num paths total 75 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0199282 +expl/Actions Std 0.833746 +expl/Actions Max 2.75327 +expl/Actions Min -2.55312 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 75000 +eval/num paths total 75 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.136403 +eval/Actions Std 0.640577 +eval/Actions Max 0.998186 +eval/Actions Min -0.996708 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.71108e-06 +time/evaluation sampling (s) 3.41218 +time/exploration sampling (s) 3.83205 +time/logging (s) 0.0124425 +time/saving (s) 0.0160963 +time/training (s) 15.5833 +time/epoch (s) 22.8561 +time/total (s) 1686.93 +Epoch -926 +------------------------------ ---------------- +2022-05-15 18:30:55.624974 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -925 finished +------------------------------ ---------------- +epoch -925 +replay_buffer/size 999047 +trainer/num train calls 76000 +trainer/QF1 Loss 1.18685 +trainer/QF2 Loss 1.16484 +trainer/Policy Loss 6.94419 +trainer/Q1 Predictions Mean -68.5836 +trainer/Q1 Predictions Std 20.3303 +trainer/Q1 Predictions Max -0.397424 +trainer/Q1 Predictions Min -87.2515 +trainer/Q2 Predictions Mean -68.5731 +trainer/Q2 Predictions Std 20.3034 +trainer/Q2 Predictions Max -0.362453 +trainer/Q2 Predictions Min -87.2029 +trainer/Q Targets Mean -68.0293 +trainer/Q Targets Std 20.5961 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6299 +trainer/rewards Mean -0.976562 +trainer/rewards Std 0.151288 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0234375 +trainer/terminals Std 0.151288 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00320537 +trainer/policy/mean Std 0.676733 +trainer/policy/mean Max 0.997968 +trainer/policy/mean Min -0.995503 +trainer/policy/std Mean 0.456482 +trainer/policy/std Std 0.0232308 +trainer/policy/std Max 0.483357 +trainer/policy/std Min 0.416845 +trainer/Advantage Weights Mean 2.28989 +trainer/Advantage Weights Std 13.2556 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.43166e-16 +trainer/Advantage Score Mean -0.506519 +trainer/Advantage Score Std 0.514328 +trainer/Advantage Score Max 1.19393 +trainer/Advantage Score Min -3.56083 +trainer/V1 Predictions Mean -67.8659 +trainer/V1 Predictions Std 20.4383 +trainer/V1 Predictions Max 0.277572 +trainer/V1 Predictions Min -86.9207 +trainer/VF Loss 0.064116 +expl/num steps total 76000 +expl/num paths total 76 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0247225 +expl/Actions Std 0.776823 +expl/Actions Max 2.60368 +expl/Actions Min -2.50037 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 76000 +eval/num paths total 76 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0594552 +eval/Actions Std 0.6521 +eval/Actions Max 0.999171 +eval/Actions Min -0.996237 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.65382e-06 +time/evaluation sampling (s) 3.3459 +time/exploration sampling (s) 3.99018 +time/logging (s) 0.00845171 +time/saving (s) 0.0115027 +time/training (s) 15.5688 +time/epoch (s) 22.9248 +time/total (s) 1709.86 +Epoch -925 +------------------------------ ---------------- +2022-05-15 18:31:18.723308 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -924 finished +------------------------------ ---------------- +epoch -924 +replay_buffer/size 999047 +trainer/num train calls 77000 +trainer/QF1 Loss 0.84783 +trainer/QF2 Loss 0.903876 +trainer/Policy Loss 27.0792 +trainer/Q1 Predictions Mean -71.6915 +trainer/Q1 Predictions Std 15.8589 +trainer/Q1 Predictions Max -2.16885 +trainer/Q1 Predictions Min -87.4789 +trainer/Q2 Predictions Mean -71.6995 +trainer/Q2 Predictions Std 15.8492 +trainer/Q2 Predictions Max -2.2592 +trainer/Q2 Predictions Min -87.5145 +trainer/Q Targets Mean -71.8187 +trainer/Q Targets Std 15.935 +trainer/Q Targets Max -2.43635 +trainer/Q Targets Min -88.2124 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00969752 +trainer/policy/mean Std 0.691794 +trainer/policy/mean Max 0.996847 +trainer/policy/mean Min -0.994985 +trainer/policy/std Mean 0.455997 +trainer/policy/std Std 0.0248712 +trainer/policy/std Max 0.487341 +trainer/policy/std Min 0.41475 +trainer/Advantage Weights Mean 4.60054 +trainer/Advantage Weights Std 16.1044 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.18995e-10 +trainer/Advantage Score Mean -0.296347 +trainer/Advantage Score Std 0.461302 +trainer/Advantage Score Max 0.757355 +trainer/Advantage Score Min -2.18658 +trainer/V1 Predictions Mean -71.5193 +trainer/V1 Predictions Std 16.0475 +trainer/V1 Predictions Max -0.932288 +trainer/V1 Predictions Min -88.1766 +trainer/VF Loss 0.0438785 +expl/num steps total 77000 +expl/num paths total 77 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.320121 +expl/Actions Std 0.830443 +expl/Actions Max 2.38855 +expl/Actions Min -2.69936 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 77000 +eval/num paths total 77 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0011853 +eval/Actions Std 0.714882 +eval/Actions Max 0.996982 +eval/Actions Min -0.998473 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.08292e-06 +time/evaluation sampling (s) 3.41436 +time/exploration sampling (s) 3.90672 +time/logging (s) 0.00731529 +time/saving (s) 0.0102592 +time/training (s) 15.7525 +time/epoch (s) 23.0912 +time/total (s) 1732.95 +Epoch -924 +------------------------------ ---------------- +2022-05-15 18:31:41.588936 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -923 finished +------------------------------ ---------------- +epoch -923 +replay_buffer/size 999047 +trainer/num train calls 78000 +trainer/QF1 Loss 2.25687 +trainer/QF2 Loss 2.33546 +trainer/Policy Loss 5.02571 +trainer/Q1 Predictions Mean -69.3933 +trainer/Q1 Predictions Std 19.1649 +trainer/Q1 Predictions Max -3.98613 +trainer/Q1 Predictions Min -88.906 +trainer/Q2 Predictions Mean -69.5406 +trainer/Q2 Predictions Std 19.1197 +trainer/Q2 Predictions Max -3.80577 +trainer/Q2 Predictions Min -89.1223 +trainer/Q Targets Mean -68.8169 +trainer/Q Targets Std 19.2916 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5658 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0192043 +trainer/policy/mean Std 0.701896 +trainer/policy/mean Max 0.996163 +trainer/policy/mean Min -0.993033 +trainer/policy/std Mean 0.455111 +trainer/policy/std Std 0.0245339 +trainer/policy/std Max 0.485775 +trainer/policy/std Min 0.418121 +trainer/Advantage Weights Mean 1.64017 +trainer/Advantage Weights Std 11.4828 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.82325e-12 +trainer/Advantage Score Mean -0.740664 +trainer/Advantage Score Std 0.554612 +trainer/Advantage Score Max 1.06789 +trainer/Advantage Score Min -2.65931 +trainer/V1 Predictions Mean -68.466 +trainer/V1 Predictions Std 19.3758 +trainer/V1 Predictions Max -2.43088 +trainer/V1 Predictions Min -88.8736 +trainer/VF Loss 0.0943647 +expl/num steps total 78000 +expl/num paths total 78 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.122106 +expl/Actions Std 0.845591 +expl/Actions Max 2.90967 +expl/Actions Min -2.6095 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 77892 +eval/num paths total 78 +eval/path length Mean 892 +eval/path length Std 0 +eval/path length Max 892 +eval/path length Min 892 +eval/Rewards Mean 0.00112108 +eval/Rewards Std 0.0334637 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0150434 +eval/Actions Std 0.703476 +eval/Actions Max 0.997484 +eval/Actions Min -0.998063 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 8.14861e-06 +time/evaluation sampling (s) 3.39238 +time/exploration sampling (s) 3.88591 +time/logging (s) 0.00697272 +time/saving (s) 0.0110112 +time/training (s) 15.5604 +time/epoch (s) 22.8566 +time/total (s) 1755.82 +Epoch -923 +------------------------------ ---------------- +2022-05-15 18:32:03.617475 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -922 finished +------------------------------ ---------------- +epoch -922 +replay_buffer/size 999047 +trainer/num train calls 79000 +trainer/QF1 Loss 1.26401 +trainer/QF2 Loss 1.45364 +trainer/Policy Loss 42.8997 +trainer/Q1 Predictions Mean -68.2746 +trainer/Q1 Predictions Std 20.0846 +trainer/Q1 Predictions Max -0.760053 +trainer/Q1 Predictions Min -86.7169 +trainer/Q2 Predictions Mean -68.2239 +trainer/Q2 Predictions Std 20.1746 +trainer/Q2 Predictions Max -0.420662 +trainer/Q2 Predictions Min -86.1616 +trainer/Q Targets Mean -68.7458 +trainer/Q Targets Std 20.1687 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8292 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0146105 +trainer/policy/mean Std 0.692368 +trainer/policy/mean Max 0.997545 +trainer/policy/mean Min -0.998619 +trainer/policy/std Mean 0.454093 +trainer/policy/std Std 0.0233695 +trainer/policy/std Max 0.481243 +trainer/policy/std Min 0.417865 +trainer/Advantage Weights Mean 8.80187 +trainer/Advantage Weights Std 23.9039 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.02217e-11 +trainer/Advantage Score Mean -0.197309 +trainer/Advantage Score Std 0.482229 +trainer/Advantage Score Max 0.907982 +trainer/Advantage Score Min -2.39366 +trainer/V1 Predictions Mean -68.4456 +trainer/V1 Predictions Std 20.2315 +trainer/V1 Predictions Max -0.24974 +trainer/V1 Predictions Min -86.5151 +trainer/VF Loss 0.0545455 +expl/num steps total 79000 +expl/num paths total 79 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0141296 +expl/Actions Std 0.790585 +expl/Actions Max 2.43112 +expl/Actions Min -2.39518 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 78892 +eval/num paths total 79 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00174162 +eval/Actions Std 0.68477 +eval/Actions Max 0.996319 +eval/Actions Min -0.998128 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.51388e-06 +time/evaluation sampling (s) 3.07099 +time/exploration sampling (s) 3.77837 +time/logging (s) 0.00700435 +time/saving (s) 0.0103548 +time/training (s) 15.155 +time/epoch (s) 22.0218 +time/total (s) 1777.85 +Epoch -922 +------------------------------ ---------------- +2022-05-15 18:32:26.257498 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -921 finished +------------------------------ ---------------- +epoch -921 +replay_buffer/size 999047 +trainer/num train calls 80000 +trainer/QF1 Loss 0.534231 +trainer/QF2 Loss 0.552959 +trainer/Policy Loss 13.7066 +trainer/Q1 Predictions Mean -70.5922 +trainer/Q1 Predictions Std 18.2941 +trainer/Q1 Predictions Max -1.89394 +trainer/Q1 Predictions Min -87.1349 +trainer/Q2 Predictions Mean -70.6767 +trainer/Q2 Predictions Std 18.2778 +trainer/Q2 Predictions Max -2.93814 +trainer/Q2 Predictions Min -87.3127 +trainer/Q Targets Mean -70.5743 +trainer/Q Targets Std 18.3709 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9076 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00644793 +trainer/policy/mean Std 0.682861 +trainer/policy/mean Max 0.998613 +trainer/policy/mean Min -0.992263 +trainer/policy/std Mean 0.456151 +trainer/policy/std Std 0.0247218 +trainer/policy/std Max 0.485181 +trainer/policy/std Min 0.419925 +trainer/Advantage Weights Mean 3.0907 +trainer/Advantage Weights Std 14.814 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.02083e-10 +trainer/Advantage Score Mean -0.288437 +trainer/Advantage Score Std 0.373888 +trainer/Advantage Score Max 1.05746 +trainer/Advantage Score Min -2.19203 +trainer/V1 Predictions Mean -70.3912 +trainer/V1 Predictions Std 18.2827 +trainer/V1 Predictions Max -3.60461 +trainer/V1 Predictions Min -86.937 +trainer/VF Loss 0.0334311 +expl/num steps total 80000 +expl/num paths total 80 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.234375 +expl/Actions Std 0.798349 +expl/Actions Max 2.48461 +expl/Actions Min -2.59195 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 79892 +eval/num paths total 80 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0987271 +eval/Actions Std 0.706022 +eval/Actions Max 0.997138 +eval/Actions Min -0.995322 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.89688e-06 +time/evaluation sampling (s) 3.31205 +time/exploration sampling (s) 4.03381 +time/logging (s) 0.0120563 +time/saving (s) 0.0164741 +time/training (s) 15.2638 +time/epoch (s) 22.6382 +time/total (s) 1800.49 +Epoch -921 +------------------------------ ---------------- +2022-05-15 18:32:49.594756 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -920 finished +------------------------------ ---------------- +epoch -920 +replay_buffer/size 999047 +trainer/num train calls 81000 +trainer/QF1 Loss 0.559449 +trainer/QF2 Loss 0.635818 +trainer/Policy Loss 24.192 +trainer/Q1 Predictions Mean -70.9518 +trainer/Q1 Predictions Std 17.4796 +trainer/Q1 Predictions Max -2.67764 +trainer/Q1 Predictions Min -87.2541 +trainer/Q2 Predictions Mean -70.9153 +trainer/Q2 Predictions Std 17.4867 +trainer/Q2 Predictions Max -2.81812 +trainer/Q2 Predictions Min -87.2086 +trainer/Q Targets Mean -71.07 +trainer/Q Targets Std 17.3621 +trainer/Q Targets Max -2.31532 +trainer/Q Targets Min -87.1299 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0342586 +trainer/policy/mean Std 0.691292 +trainer/policy/mean Max 0.997972 +trainer/policy/mean Min -0.997567 +trainer/policy/std Mean 0.453756 +trainer/policy/std Std 0.0248033 +trainer/policy/std Max 0.48396 +trainer/policy/std Min 0.413758 +trainer/Advantage Weights Mean 4.66167 +trainer/Advantage Weights Std 17.437 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.36043e-12 +trainer/Advantage Score Mean -0.297498 +trainer/Advantage Score Std 0.484073 +trainer/Advantage Score Max 0.715748 +trainer/Advantage Score Min -2.61585 +trainer/V1 Predictions Mean -70.7586 +trainer/V1 Predictions Std 17.4856 +trainer/V1 Predictions Max -2.86383 +trainer/V1 Predictions Min -86.8812 +trainer/VF Loss 0.0447675 +expl/num steps total 81000 +expl/num paths total 81 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00755073 +expl/Actions Std 0.798701 +expl/Actions Max 2.38958 +expl/Actions Min -2.31579 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 80892 +eval/num paths total 81 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0248291 +eval/Actions Std 0.711499 +eval/Actions Max 0.998016 +eval/Actions Min -0.99891 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.13018e-06 +time/evaluation sampling (s) 3.4928 +time/exploration sampling (s) 4.15047 +time/logging (s) 0.0126171 +time/saving (s) 0.0183538 +time/training (s) 15.6541 +time/epoch (s) 23.3283 +time/total (s) 1823.82 +Epoch -920 +------------------------------ ---------------- +2022-05-15 18:33:11.804775 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -919 finished +------------------------------ ---------------- +epoch -919 +replay_buffer/size 999047 +trainer/num train calls 82000 +trainer/QF1 Loss 0.72223 +trainer/QF2 Loss 0.666416 +trainer/Policy Loss 22.9349 +trainer/Q1 Predictions Mean -70.8738 +trainer/Q1 Predictions Std 16.7707 +trainer/Q1 Predictions Max -1.0109 +trainer/Q1 Predictions Min -88.7022 +trainer/Q2 Predictions Mean -70.8336 +trainer/Q2 Predictions Std 16.8112 +trainer/Q2 Predictions Max -2.11646 +trainer/Q2 Predictions Min -88.9841 +trainer/Q Targets Mean -70.834 +trainer/Q Targets Std 16.9131 +trainer/Q Targets Max -3.33714 +trainer/Q Targets Min -89.2561 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00274131 +trainer/policy/mean Std 0.679974 +trainer/policy/mean Max 0.997467 +trainer/policy/mean Min -0.993205 +trainer/policy/std Mean 0.453798 +trainer/policy/std Std 0.0247348 +trainer/policy/std Max 0.483681 +trainer/policy/std Min 0.414181 +trainer/Advantage Weights Mean 4.47608 +trainer/Advantage Weights Std 16.3514 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.26879e-10 +trainer/Advantage Score Mean -0.227628 +trainer/Advantage Score Std 0.43596 +trainer/Advantage Score Max 0.686314 +trainer/Advantage Score Min -2.18414 +trainer/V1 Predictions Mean -70.5468 +trainer/V1 Predictions Std 17.0341 +trainer/V1 Predictions Max -0.636818 +trainer/V1 Predictions Min -88.6306 +trainer/VF Loss 0.0365551 +expl/num steps total 82000 +expl/num paths total 82 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.114781 +expl/Actions Std 0.85483 +expl/Actions Max 2.41819 +expl/Actions Min -2.41429 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 81892 +eval/num paths total 82 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00600069 +eval/Actions Std 0.700804 +eval/Actions Max 0.998774 +eval/Actions Min -0.998851 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.07826e-06 +time/evaluation sampling (s) 2.98161 +time/exploration sampling (s) 3.95267 +time/logging (s) 0.0122248 +time/saving (s) 0.018308 +time/training (s) 15.2351 +time/epoch (s) 22.1999 +time/total (s) 1846.03 +Epoch -919 +------------------------------ ---------------- +2022-05-15 18:33:34.659647 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -918 finished +------------------------------ ---------------- +epoch -918 +replay_buffer/size 999047 +trainer/num train calls 83000 +trainer/QF1 Loss 0.812507 +trainer/QF2 Loss 0.903583 +trainer/Policy Loss 18.9126 +trainer/Q1 Predictions Mean -69.9033 +trainer/Q1 Predictions Std 18.7267 +trainer/Q1 Predictions Max -1.65707 +trainer/Q1 Predictions Min -86.7725 +trainer/Q2 Predictions Mean -69.8957 +trainer/Q2 Predictions Std 18.7764 +trainer/Q2 Predictions Max -2.09194 +trainer/Q2 Predictions Min -86.5786 +trainer/Q Targets Mean -69.8709 +trainer/Q Targets Std 18.6851 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8101 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00546224 +trainer/policy/mean Std 0.684978 +trainer/policy/mean Max 0.999292 +trainer/policy/mean Min -0.997116 +trainer/policy/std Mean 0.452833 +trainer/policy/std Std 0.0252148 +trainer/policy/std Max 0.485421 +trainer/policy/std Min 0.411871 +trainer/Advantage Weights Mean 3.49224 +trainer/Advantage Weights Std 14.9986 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.2362e-13 +trainer/Advantage Score Mean -0.352479 +trainer/Advantage Score Std 0.473776 +trainer/Advantage Score Max 1.71525 +trainer/Advantage Score Min -2.77105 +trainer/V1 Predictions Mean -69.6195 +trainer/V1 Predictions Std 18.7675 +trainer/V1 Predictions Max -3.87469 +trainer/V1 Predictions Min -86.5044 +trainer/VF Loss 0.0570548 +expl/num steps total 83000 +expl/num paths total 83 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0546302 +expl/Actions Std 0.874647 +expl/Actions Max 2.68899 +expl/Actions Min -2.26285 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 82892 +eval/num paths total 83 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00575386 +eval/Actions Std 0.697745 +eval/Actions Max 0.998051 +eval/Actions Min -0.997843 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.25316e-06 +time/evaluation sampling (s) 2.95272 +time/exploration sampling (s) 4.18338 +time/logging (s) 0.0135283 +time/saving (s) 0.0197057 +time/training (s) 15.6766 +time/epoch (s) 22.8459 +time/total (s) 1868.89 +Epoch -918 +------------------------------ ---------------- +2022-05-15 18:33:57.514227 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -917 finished +------------------------------ ---------------- +epoch -917 +replay_buffer/size 999047 +trainer/num train calls 84000 +trainer/QF1 Loss 0.835187 +trainer/QF2 Loss 0.80091 +trainer/Policy Loss 24.2426 +trainer/Q1 Predictions Mean -69.5413 +trainer/Q1 Predictions Std 20.2383 +trainer/Q1 Predictions Max -1.65829 +trainer/Q1 Predictions Min -87.5022 +trainer/Q2 Predictions Mean -69.4936 +trainer/Q2 Predictions Std 20.2453 +trainer/Q2 Predictions Max -1.61778 +trainer/Q2 Predictions Min -87.3512 +trainer/Q Targets Mean -69.5075 +trainer/Q Targets Std 20.2062 +trainer/Q Targets Max -2.8674 +trainer/Q Targets Min -87.5694 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0178475 +trainer/policy/mean Std 0.696517 +trainer/policy/mean Max 0.995672 +trainer/policy/mean Min -0.993848 +trainer/policy/std Mean 0.452905 +trainer/policy/std Std 0.0244367 +trainer/policy/std Max 0.484339 +trainer/policy/std Min 0.413474 +trainer/Advantage Weights Mean 5.41846 +trainer/Advantage Weights Std 19.3466 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.28731e-13 +trainer/Advantage Score Mean -0.368552 +trainer/Advantage Score Std 0.573409 +trainer/Advantage Score Max 0.751175 +trainer/Advantage Score Min -2.96811 +trainer/V1 Predictions Mean -69.2002 +trainer/V1 Predictions Std 20.4462 +trainer/V1 Predictions Max -0.549334 +trainer/V1 Predictions Min -87.487 +trainer/VF Loss 0.0634031 +expl/num steps total 84000 +expl/num paths total 84 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0140524 +expl/Actions Std 0.794929 +expl/Actions Max 2.49364 +expl/Actions Min -2.38851 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 83892 +eval/num paths total 84 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0208915 +eval/Actions Std 0.6912 +eval/Actions Max 0.998756 +eval/Actions Min -0.998832 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.60398e-06 +time/evaluation sampling (s) 3.06553 +time/exploration sampling (s) 4.33113 +time/logging (s) 0.0111112 +time/saving (s) 0.0164338 +time/training (s) 15.4179 +time/epoch (s) 22.8421 +time/total (s) 1891.74 +Epoch -917 +------------------------------ ---------------- +2022-05-15 18:34:20.232248 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -916 finished +------------------------------ ---------------- +epoch -916 +replay_buffer/size 999047 +trainer/num train calls 85000 +trainer/QF1 Loss 0.516006 +trainer/QF2 Loss 0.552986 +trainer/Policy Loss 18.4859 +trainer/Q1 Predictions Mean -71.5514 +trainer/Q1 Predictions Std 17.2058 +trainer/Q1 Predictions Max -5.66971 +trainer/Q1 Predictions Min -87.1783 +trainer/Q2 Predictions Mean -71.5201 +trainer/Q2 Predictions Std 17.2183 +trainer/Q2 Predictions Max -5.95795 +trainer/Q2 Predictions Min -87.1648 +trainer/Q Targets Mean -71.563 +trainer/Q Targets Std 17.1735 +trainer/Q Targets Max -6.02466 +trainer/Q Targets Min -87.1809 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000823095 +trainer/policy/mean Std 0.679552 +trainer/policy/mean Max 0.996845 +trainer/policy/mean Min -0.994248 +trainer/policy/std Mean 0.452765 +trainer/policy/std Std 0.0234058 +trainer/policy/std Max 0.483168 +trainer/policy/std Min 0.415643 +trainer/Advantage Weights Mean 4.40202 +trainer/Advantage Weights Std 15.1763 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.13739e-16 +trainer/Advantage Score Mean -0.276634 +trainer/Advantage Score Std 0.513678 +trainer/Advantage Score Max 1.07981 +trainer/Advantage Score Min -3.47449 +trainer/V1 Predictions Mean -71.252 +trainer/V1 Predictions Std 17.4422 +trainer/V1 Predictions Max -4.45876 +trainer/V1 Predictions Min -87.0493 +trainer/VF Loss 0.0519092 +expl/num steps total 85000 +expl/num paths total 85 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0582098 +expl/Actions Std 0.776551 +expl/Actions Max 2.38115 +expl/Actions Min -2.38361 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 84892 +eval/num paths total 85 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0703719 +eval/Actions Std 0.717448 +eval/Actions Max 0.997707 +eval/Actions Min -0.996751 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.03568e-05 +time/evaluation sampling (s) 2.85053 +time/exploration sampling (s) 4.32341 +time/logging (s) 0.00825363 +time/saving (s) 0.0111988 +time/training (s) 15.509 +time/epoch (s) 22.7024 +time/total (s) 1914.45 +Epoch -916 +------------------------------ ---------------- +2022-05-15 18:34:42.691314 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -915 finished +------------------------------ ---------------- +epoch -915 +replay_buffer/size 999047 +trainer/num train calls 86000 +trainer/QF1 Loss 1.3633 +trainer/QF2 Loss 1.32772 +trainer/Policy Loss 0.0672765 +trainer/Q1 Predictions Mean -68.8475 +trainer/Q1 Predictions Std 20.0511 +trainer/Q1 Predictions Max -1.43455 +trainer/Q1 Predictions Min -86.8013 +trainer/Q2 Predictions Mean -68.8362 +trainer/Q2 Predictions Std 20.0994 +trainer/Q2 Predictions Max -1.98458 +trainer/Q2 Predictions Min -86.5962 +trainer/Q Targets Mean -68.0772 +trainer/Q Targets Std 20.2091 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8297 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0200341 +trainer/policy/mean Std 0.689918 +trainer/policy/mean Max 0.997274 +trainer/policy/mean Min -0.996036 +trainer/policy/std Mean 0.453634 +trainer/policy/std Std 0.0215545 +trainer/policy/std Max 0.484191 +trainer/policy/std Min 0.421318 +trainer/Advantage Weights Mean 0.0104174 +trainer/Advantage Weights Std 0.0665263 +trainer/Advantage Weights Max 0.92888 +trainer/Advantage Weights Min 2.25168e-19 +trainer/Advantage Score Mean -1.19126 +trainer/Advantage Score Std 0.692372 +trainer/Advantage Score Max -0.00737762 +trainer/Advantage Score Min -4.29374 +trainer/V1 Predictions Mean -67.7793 +trainer/V1 Predictions Std 20.4381 +trainer/V1 Predictions Max 0.588176 +trainer/V1 Predictions Min -86.5449 +trainer/VF Loss 0.189849 +expl/num steps total 86000 +expl/num paths total 86 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.062742 +expl/Actions Std 0.835199 +expl/Actions Max 2.37156 +expl/Actions Min -2.45042 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 85892 +eval/num paths total 86 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00888288 +eval/Actions Std 0.684143 +eval/Actions Max 0.999363 +eval/Actions Min -0.997987 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.92202e-06 +time/evaluation sampling (s) 3.08717 +time/exploration sampling (s) 4.47316 +time/logging (s) 0.0100833 +time/saving (s) 0.0156121 +time/training (s) 14.8681 +time/epoch (s) 22.4541 +time/total (s) 1936.91 +Epoch -915 +------------------------------ ---------------- +2022-05-15 18:35:04.495239 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -914 finished +------------------------------ ---------------- +epoch -914 +replay_buffer/size 999047 +trainer/num train calls 87000 +trainer/QF1 Loss 1.23661 +trainer/QF2 Loss 1.73588 +trainer/Policy Loss 42.8081 +trainer/Q1 Predictions Mean -70.525 +trainer/Q1 Predictions Std 17.8084 +trainer/Q1 Predictions Max -2.24016 +trainer/Q1 Predictions Min -88.1554 +trainer/Q2 Predictions Mean -70.3013 +trainer/Q2 Predictions Std 17.8601 +trainer/Q2 Predictions Max -2.35023 +trainer/Q2 Predictions Min -88.4118 +trainer/Q Targets Mean -71.0615 +trainer/Q Targets Std 17.7158 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.7373 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00728136 +trainer/policy/mean Std 0.687394 +trainer/policy/mean Max 0.998333 +trainer/policy/mean Min -0.994501 +trainer/policy/std Mean 0.452496 +trainer/policy/std Std 0.0221048 +trainer/policy/std Max 0.482223 +trainer/policy/std Min 0.419155 +trainer/Advantage Weights Mean 8.09089 +trainer/Advantage Weights Std 22.7865 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.33754e-08 +trainer/Advantage Score Mean -0.131962 +trainer/Advantage Score Std 0.375559 +trainer/Advantage Score Max 1.40008 +trainer/Advantage Score Min -1.61866 +trainer/V1 Predictions Mean -70.8737 +trainer/V1 Predictions Std 17.6179 +trainer/V1 Predictions Max -3.36276 +trainer/V1 Predictions Min -87.9646 +trainer/VF Loss 0.0474114 +expl/num steps total 87000 +expl/num paths total 87 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00556692 +expl/Actions Std 0.788331 +expl/Actions Max 2.54339 +expl/Actions Min -2.33716 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 86892 +eval/num paths total 87 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00925714 +eval/Actions Std 0.702977 +eval/Actions Max 0.998453 +eval/Actions Min -0.998506 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.54439e-06 +time/evaluation sampling (s) 3.33606 +time/exploration sampling (s) 3.98029 +time/logging (s) 0.00816188 +time/saving (s) 0.0132285 +time/training (s) 14.4551 +time/epoch (s) 21.7928 +time/total (s) 1958.71 +Epoch -914 +------------------------------ ---------------- +2022-05-15 18:35:26.382922 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -913 finished +------------------------------ ---------------- +epoch -913 +replay_buffer/size 999047 +trainer/num train calls 88000 +trainer/QF1 Loss 0.673889 +trainer/QF2 Loss 0.617059 +trainer/Policy Loss 5.53653 +trainer/Q1 Predictions Mean -72.0539 +trainer/Q1 Predictions Std 16.4587 +trainer/Q1 Predictions Max -1.19263 +trainer/Q1 Predictions Min -87.5365 +trainer/Q2 Predictions Mean -72.0724 +trainer/Q2 Predictions Std 16.4631 +trainer/Q2 Predictions Max -2.03769 +trainer/Q2 Predictions Min -87.9256 +trainer/Q Targets Mean -71.8526 +trainer/Q Targets Std 16.324 +trainer/Q Targets Max -3.19297 +trainer/Q Targets Min -86.9027 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0166771 +trainer/policy/mean Std 0.687613 +trainer/policy/mean Max 0.997086 +trainer/policy/mean Min -0.995159 +trainer/policy/std Mean 0.45221 +trainer/policy/std Std 0.0238263 +trainer/policy/std Max 0.482759 +trainer/policy/std Min 0.412247 +trainer/Advantage Weights Mean 1.54698 +trainer/Advantage Weights Std 9.24401 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.61976e-12 +trainer/Advantage Score Mean -0.431888 +trainer/Advantage Score Std 0.433332 +trainer/Advantage Score Max 0.683586 +trainer/Advantage Score Min -2.59047 +trainer/V1 Predictions Mean -71.6093 +trainer/V1 Predictions Std 16.4562 +trainer/V1 Predictions Max -0.430149 +trainer/V1 Predictions Min -86.8159 +trainer/VF Loss 0.0418602 +expl/num steps total 88000 +expl/num paths total 88 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0348704 +expl/Actions Std 0.864259 +expl/Actions Max 2.2996 +expl/Actions Min -2.22604 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 87892 +eval/num paths total 88 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.027606 +eval/Actions Std 0.690617 +eval/Actions Max 0.999203 +eval/Actions Min -0.998084 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.54298e-06 +time/evaluation sampling (s) 3.39359 +time/exploration sampling (s) 4.18391 +time/logging (s) 0.00821032 +time/saving (s) 0.0159447 +time/training (s) 14.2792 +time/epoch (s) 21.8809 +time/total (s) 1980.59 +Epoch -913 +------------------------------ ---------------- +2022-05-15 18:35:48.257601 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -912 finished +------------------------------ ---------------- +epoch -912 +replay_buffer/size 999047 +trainer/num train calls 89000 +trainer/QF1 Loss 0.646898 +trainer/QF2 Loss 0.70513 +trainer/Policy Loss 10.2969 +trainer/Q1 Predictions Mean -70.7975 +trainer/Q1 Predictions Std 17.3776 +trainer/Q1 Predictions Max -2.24353 +trainer/Q1 Predictions Min -87.7538 +trainer/Q2 Predictions Mean -70.758 +trainer/Q2 Predictions Std 17.4128 +trainer/Q2 Predictions Max -2.01923 +trainer/Q2 Predictions Min -87.9484 +trainer/Q Targets Mean -70.6706 +trainer/Q Targets Std 17.2176 +trainer/Q Targets Max -2.92977 +trainer/Q Targets Min -88.0035 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00429509 +trainer/policy/mean Std 0.67758 +trainer/policy/mean Max 0.997629 +trainer/policy/mean Min -0.999559 +trainer/policy/std Mean 0.452177 +trainer/policy/std Std 0.0251192 +trainer/policy/std Max 0.481326 +trainer/policy/std Min 0.408561 +trainer/Advantage Weights Mean 3.32147 +trainer/Advantage Weights Std 15.4003 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.47393e-12 +trainer/Advantage Score Mean -0.370564 +trainer/Advantage Score Std 0.478565 +trainer/Advantage Score Max 0.66996 +trainer/Advantage Score Min -2.72431 +trainer/V1 Predictions Mean -70.3752 +trainer/V1 Predictions Std 17.4898 +trainer/V1 Predictions Max -1.97479 +trainer/V1 Predictions Min -87.9686 +trainer/VF Loss 0.0459301 +expl/num steps total 89000 +expl/num paths total 89 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.040244 +expl/Actions Std 0.788606 +expl/Actions Max 2.43336 +expl/Actions Min -2.33921 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 88892 +eval/num paths total 89 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.020208 +eval/Actions Std 0.692929 +eval/Actions Max 0.998374 +eval/Actions Min -0.995109 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.54996e-06 +time/evaluation sampling (s) 3.263 +time/exploration sampling (s) 3.82439 +time/logging (s) 0.00961327 +time/saving (s) 0.0137194 +time/training (s) 14.7596 +time/epoch (s) 21.8703 +time/total (s) 2002.47 +Epoch -912 +------------------------------ ---------------- +2022-05-15 18:36:10.051010 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -911 finished +------------------------------ ---------------- +epoch -911 +replay_buffer/size 999047 +trainer/num train calls 90000 +trainer/QF1 Loss 11.2075 +trainer/QF2 Loss 11.285 +trainer/Policy Loss 11.7885 +trainer/Q1 Predictions Mean -67.4929 +trainer/Q1 Predictions Std 21.6629 +trainer/Q1 Predictions Max -1.0948 +trainer/Q1 Predictions Min -87.549 +trainer/Q2 Predictions Mean -67.4209 +trainer/Q2 Predictions Std 21.6782 +trainer/Q2 Predictions Max -0.925963 +trainer/Q2 Predictions Min -87.3292 +trainer/Q Targets Mean -67.162 +trainer/Q Targets Std 21.7499 +trainer/Q Targets Max 0.266189 +trainer/Q Targets Min -86.8306 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00687124 +trainer/policy/mean Std 0.68887 +trainer/policy/mean Max 0.993731 +trainer/policy/mean Min -0.997351 +trainer/policy/std Mean 0.451084 +trainer/policy/std Std 0.0225509 +trainer/policy/std Max 0.476694 +trainer/policy/std Min 0.410686 +trainer/Advantage Weights Mean 2.24701 +trainer/Advantage Weights Std 13.4529 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.63859e-18 +trainer/Advantage Score Mean -0.538022 +trainer/Advantage Score Std 0.539423 +trainer/Advantage Score Max 1.43677 +trainer/Advantage Score Min -3.94133 +trainer/V1 Predictions Mean -67.0849 +trainer/V1 Predictions Std 21.7056 +trainer/V1 Predictions Max 0.0779163 +trainer/V1 Predictions Min -86.6974 +trainer/VF Loss 0.0755719 +expl/num steps total 90000 +expl/num paths total 90 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0258229 +expl/Actions Std 0.79603 +expl/Actions Max 2.2748 +expl/Actions Min -2.52827 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 89892 +eval/num paths total 90 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.022376 +eval/Actions Std 0.648635 +eval/Actions Max 0.999666 +eval/Actions Min -0.997991 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.67196e-06 +time/evaluation sampling (s) 3.2588 +time/exploration sampling (s) 3.9326 +time/logging (s) 0.0125292 +time/saving (s) 0.0183431 +time/training (s) 14.5648 +time/epoch (s) 21.7871 +time/total (s) 2024.26 +Epoch -911 +------------------------------ ---------------- +2022-05-15 18:36:30.562967 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -910 finished +------------------------------ ---------------- +epoch -910 +replay_buffer/size 999047 +trainer/num train calls 91000 +trainer/QF1 Loss 0.685505 +trainer/QF2 Loss 0.652155 +trainer/Policy Loss 5.86482 +trainer/Q1 Predictions Mean -67.5751 +trainer/Q1 Predictions Std 20.9921 +trainer/Q1 Predictions Max -1.24917 +trainer/Q1 Predictions Min -86.8946 +trainer/Q2 Predictions Mean -67.5594 +trainer/Q2 Predictions Std 20.9672 +trainer/Q2 Predictions Max -1.49838 +trainer/Q2 Predictions Min -86.8361 +trainer/Q Targets Mean -67.3925 +trainer/Q Targets Std 21.1596 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5694 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0256487 +trainer/policy/mean Std 0.700039 +trainer/policy/mean Max 0.997753 +trainer/policy/mean Min -0.994341 +trainer/policy/std Mean 0.449182 +trainer/policy/std Std 0.0231418 +trainer/policy/std Max 0.475827 +trainer/policy/std Min 0.409006 +trainer/Advantage Weights Mean 1.2465 +trainer/Advantage Weights Std 7.28473 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.76247e-13 +trainer/Advantage Score Mean -0.53581 +trainer/Advantage Score Std 0.533897 +trainer/Advantage Score Max 0.562012 +trainer/Advantage Score Min -2.81822 +trainer/V1 Predictions Mean -67.153 +trainer/V1 Predictions Std 21.2168 +trainer/V1 Predictions Max -0.923603 +trainer/V1 Predictions Min -86.4936 +trainer/VF Loss 0.0608086 +expl/num steps total 91000 +expl/num paths total 91 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0366584 +expl/Actions Std 0.826166 +expl/Actions Max 2.52006 +expl/Actions Min -2.46887 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 90607 +eval/num paths total 91 +eval/path length Mean 715 +eval/path length Std 0 +eval/path length Max 715 +eval/path length Min 715 +eval/Rewards Mean 0.0013986 +eval/Rewards Std 0.0373717 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0312125 +eval/Actions Std 0.707897 +eval/Actions Max 0.99851 +eval/Actions Min -0.99893 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.55182e-06 +time/evaluation sampling (s) 3.06789 +time/exploration sampling (s) 3.38575 +time/logging (s) 0.00654966 +time/saving (s) 0.0113721 +time/training (s) 14.0251 +time/epoch (s) 20.4967 +time/total (s) 2044.76 +Epoch -910 +------------------------------ ---------------- +2022-05-15 18:36:51.548342 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -909 finished +------------------------------ ---------------- +epoch -909 +replay_buffer/size 999047 +trainer/num train calls 92000 +trainer/QF1 Loss 1.08323 +trainer/QF2 Loss 1.10011 +trainer/Policy Loss 7.46345 +trainer/Q1 Predictions Mean -69.8837 +trainer/Q1 Predictions Std 17.8196 +trainer/Q1 Predictions Max -1.85613 +trainer/Q1 Predictions Min -86.1246 +trainer/Q2 Predictions Mean -69.9349 +trainer/Q2 Predictions Std 17.8277 +trainer/Q2 Predictions Max -1.65905 +trainer/Q2 Predictions Min -86.2826 +trainer/Q Targets Mean -69.3345 +trainer/Q Targets Std 18.1677 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2696 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0141395 +trainer/policy/mean Std 0.695725 +trainer/policy/mean Max 0.998439 +trainer/policy/mean Min -0.997457 +trainer/policy/std Mean 0.450263 +trainer/policy/std Std 0.0231844 +trainer/policy/std Max 0.476003 +trainer/policy/std Min 0.414456 +trainer/Advantage Weights Mean 1.91525 +trainer/Advantage Weights Std 12.6242 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.40779e-13 +trainer/Advantage Score Mean -0.638442 +trainer/Advantage Score Std 0.488369 +trainer/Advantage Score Max 0.834198 +trainer/Advantage Score Min -2.87075 +trainer/V1 Predictions Mean -69.1088 +trainer/V1 Predictions Std 18.104 +trainer/V1 Predictions Max -1.23769 +trainer/V1 Predictions Min -86.1479 +trainer/VF Loss 0.0706879 +expl/num steps total 92000 +expl/num paths total 92 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.159696 +expl/Actions Std 0.833323 +expl/Actions Max 2.4199 +expl/Actions Min -2.28771 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 91607 +eval/num paths total 92 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.026565 +eval/Actions Std 0.693538 +eval/Actions Max 0.999717 +eval/Actions Min -0.99915 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.68314e-06 +time/evaluation sampling (s) 3.10151 +time/exploration sampling (s) 3.43595 +time/logging (s) 0.0123036 +time/saving (s) 0.0179412 +time/training (s) 14.4165 +time/epoch (s) 20.9842 +time/total (s) 2065.75 +Epoch -909 +------------------------------ ---------------- +2022-05-15 18:37:13.082190 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -908 finished +------------------------------ --------------- +epoch -908 +replay_buffer/size 999047 +trainer/num train calls 93000 +trainer/QF1 Loss 0.649363 +trainer/QF2 Loss 0.69572 +trainer/Policy Loss 13.7893 +trainer/Q1 Predictions Mean -70.0872 +trainer/Q1 Predictions Std 18.8417 +trainer/Q1 Predictions Max -0.993804 +trainer/Q1 Predictions Min -87.6507 +trainer/Q2 Predictions Mean -70.0699 +trainer/Q2 Predictions Std 18.8569 +trainer/Q2 Predictions Max -1.02215 +trainer/Q2 Predictions Min -87.2904 +trainer/Q Targets Mean -70.1226 +trainer/Q Targets Std 18.9801 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7485 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0195792 +trainer/policy/mean Std 0.689121 +trainer/policy/mean Max 0.997613 +trainer/policy/mean Min -0.993957 +trainer/policy/std Mean 0.451386 +trainer/policy/std Std 0.0230258 +trainer/policy/std Max 0.477366 +trainer/policy/std Min 0.413849 +trainer/Advantage Weights Mean 1.95856 +trainer/Advantage Weights Std 12.5236 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.3311e-15 +trainer/Advantage Score Mean -0.461984 +trainer/Advantage Score Std 0.475165 +trainer/Advantage Score Max 0.835966 +trainer/Advantage Score Min -3.28652 +trainer/V1 Predictions Mean -69.9068 +trainer/V1 Predictions Std 18.9975 +trainer/V1 Predictions Max -0.499483 +trainer/V1 Predictions Min -87.4785 +trainer/VF Loss 0.0507225 +expl/num steps total 93000 +expl/num paths total 93 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0711598 +expl/Actions Std 0.819134 +expl/Actions Max 2.54711 +expl/Actions Min -2.52311 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 92226 +eval/num paths total 93 +eval/path length Mean 619 +eval/path length Std 0 +eval/path length Max 619 +eval/path length Min 619 +eval/Rewards Mean 0.00161551 +eval/Rewards Std 0.0401609 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00232819 +eval/Actions Std 0.723501 +eval/Actions Max 0.999487 +eval/Actions Min -0.99836 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.5928e-06 +time/evaluation sampling (s) 3.00407 +time/exploration sampling (s) 3.75899 +time/logging (s) 0.00866937 +time/saving (s) 0.0138805 +time/training (s) 14.7351 +time/epoch (s) 21.5207 +time/total (s) 2087.28 +Epoch -908 +------------------------------ --------------- +2022-05-15 18:37:34.123279 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -907 finished +------------------------------ ---------------- +epoch -907 +replay_buffer/size 999047 +trainer/num train calls 94000 +trainer/QF1 Loss 0.821114 +trainer/QF2 Loss 1.02027 +trainer/Policy Loss 28.4349 +trainer/Q1 Predictions Mean -67.9096 +trainer/Q1 Predictions Std 19.8057 +trainer/Q1 Predictions Max -0.932907 +trainer/Q1 Predictions Min -86.2694 +trainer/Q2 Predictions Mean -67.8674 +trainer/Q2 Predictions Std 19.8449 +trainer/Q2 Predictions Max -0.888316 +trainer/Q2 Predictions Min -86.3258 +trainer/Q Targets Mean -68.3164 +trainer/Q Targets Std 20.0308 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0042 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000500236 +trainer/policy/mean Std 0.678253 +trainer/policy/mean Max 0.997048 +trainer/policy/mean Min -0.99602 +trainer/policy/std Mean 0.452046 +trainer/policy/std Std 0.0244645 +trainer/policy/std Max 0.483863 +trainer/policy/std Min 0.413728 +trainer/Advantage Weights Mean 5.84953 +trainer/Advantage Weights Std 20.7662 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.41019e-16 +trainer/Advantage Score Mean -0.38144 +trainer/Advantage Score Std 0.610244 +trainer/Advantage Score Max 2.09848 +trainer/Advantage Score Min -3.53574 +trainer/V1 Predictions Mean -68.1033 +trainer/V1 Predictions Std 20.0379 +trainer/V1 Predictions Max -0.835463 +trainer/V1 Predictions Min -86.7589 +trainer/VF Loss 0.103099 +expl/num steps total 94000 +expl/num paths total 94 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0346608 +expl/Actions Std 0.807737 +expl/Actions Max 2.40085 +expl/Actions Min -2.75925 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 93226 +eval/num paths total 94 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0176236 +eval/Actions Std 0.69004 +eval/Actions Max 0.996366 +eval/Actions Min -0.998863 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.13297e-06 +time/evaluation sampling (s) 2.76957 +time/exploration sampling (s) 3.68364 +time/logging (s) 0.00745594 +time/saving (s) 0.010803 +time/training (s) 14.5603 +time/epoch (s) 21.0317 +time/total (s) 2108.32 +Epoch -907 +------------------------------ ---------------- +2022-05-15 18:37:55.625259 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -906 finished +------------------------------ ---------------- +epoch -906 +replay_buffer/size 999047 +trainer/num train calls 95000 +trainer/QF1 Loss 0.696365 +trainer/QF2 Loss 0.632938 +trainer/Policy Loss 23.7836 +trainer/Q1 Predictions Mean -70.4453 +trainer/Q1 Predictions Std 17.516 +trainer/Q1 Predictions Max -1.75633 +trainer/Q1 Predictions Min -86.2954 +trainer/Q2 Predictions Mean -70.5173 +trainer/Q2 Predictions Std 17.5671 +trainer/Q2 Predictions Max -1.92274 +trainer/Q2 Predictions Min -86.5019 +trainer/Q Targets Mean -70.7248 +trainer/Q Targets Std 17.3317 +trainer/Q Targets Max -3.56956 +trainer/Q Targets Min -86.7147 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0106026 +trainer/policy/mean Std 0.700382 +trainer/policy/mean Max 0.998392 +trainer/policy/mean Min -0.997834 +trainer/policy/std Mean 0.450667 +trainer/policy/std Std 0.0239667 +trainer/policy/std Max 0.480542 +trainer/policy/std Min 0.409453 +trainer/Advantage Weights Mean 4.34631 +trainer/Advantage Weights Std 16.6392 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.42618e-08 +trainer/Advantage Score Mean -0.268163 +trainer/Advantage Score Std 0.427972 +trainer/Advantage Score Max 0.883712 +trainer/Advantage Score Min -1.80657 +trainer/V1 Predictions Mean -70.4367 +trainer/V1 Predictions Std 17.5315 +trainer/V1 Predictions Max -2.13951 +trainer/V1 Predictions Min -86.581 +trainer/VF Loss 0.0407122 +expl/num steps total 95000 +expl/num paths total 95 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0279053 +expl/Actions Std 0.830174 +expl/Actions Max 2.61068 +expl/Actions Min -2.32707 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 94226 +eval/num paths total 95 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0608658 +eval/Actions Std 0.713156 +eval/Actions Max 0.999654 +eval/Actions Min -0.997639 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 6.63521e-06 +time/evaluation sampling (s) 3.13905 +time/exploration sampling (s) 4.0448 +time/logging (s) 0.0108621 +time/saving (s) 0.0153452 +time/training (s) 14.2891 +time/epoch (s) 21.4992 +time/total (s) 2129.82 +Epoch -906 +------------------------------ ---------------- +2022-05-15 18:38:17.624469 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -905 finished +------------------------------ ---------------- +epoch -905 +replay_buffer/size 999047 +trainer/num train calls 96000 +trainer/QF1 Loss 0.817202 +trainer/QF2 Loss 0.740715 +trainer/Policy Loss 33.0775 +trainer/Q1 Predictions Mean -69.4815 +trainer/Q1 Predictions Std 19.7404 +trainer/Q1 Predictions Max -1.82136 +trainer/Q1 Predictions Min -87.1785 +trainer/Q2 Predictions Mean -69.4539 +trainer/Q2 Predictions Std 19.7581 +trainer/Q2 Predictions Max -1.50575 +trainer/Q2 Predictions Min -86.691 +trainer/Q Targets Mean -69.6467 +trainer/Q Targets Std 20.0204 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1203 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00225343 +trainer/policy/mean Std 0.696193 +trainer/policy/mean Max 0.995599 +trainer/policy/mean Min -0.99428 +trainer/policy/std Mean 0.449441 +trainer/policy/std Std 0.0226356 +trainer/policy/std Max 0.477173 +trainer/policy/std Min 0.411611 +trainer/Advantage Weights Mean 4.48318 +trainer/Advantage Weights Std 18.6714 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.23389e-13 +trainer/Advantage Score Mean -0.348772 +trainer/Advantage Score Std 0.499097 +trainer/Advantage Score Max 2.09375 +trainer/Advantage Score Min -2.84905 +trainer/V1 Predictions Mean -69.436 +trainer/V1 Predictions Std 19.8609 +trainer/V1 Predictions Max -2.22965 +trainer/V1 Predictions Min -86.6783 +trainer/VF Loss 0.067255 +expl/num steps total 96000 +expl/num paths total 96 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0517865 +expl/Actions Std 0.86207 +expl/Actions Max 2.23164 +expl/Actions Min -2.5461 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 95226 +eval/num paths total 96 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0158724 +eval/Actions Std 0.80078 +eval/Actions Max 0.99374 +eval/Actions Min -0.99815 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.74321e-06 +time/evaluation sampling (s) 3.27425 +time/exploration sampling (s) 4.15961 +time/logging (s) 0.00902902 +time/saving (s) 0.0134704 +time/training (s) 14.5317 +time/epoch (s) 21.9881 +time/total (s) 2151.82 +Epoch -905 +------------------------------ ---------------- +2022-05-15 18:38:39.169073 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -904 finished +------------------------------ ---------------- +epoch -904 +replay_buffer/size 999047 +trainer/num train calls 97000 +trainer/QF1 Loss 0.681637 +trainer/QF2 Loss 0.671029 +trainer/Policy Loss 13.4989 +trainer/Q1 Predictions Mean -72.0926 +trainer/Q1 Predictions Std 16.2578 +trainer/Q1 Predictions Max -1.60659 +trainer/Q1 Predictions Min -86.8619 +trainer/Q2 Predictions Mean -72.1531 +trainer/Q2 Predictions Std 16.29 +trainer/Q2 Predictions Max -2.54684 +trainer/Q2 Predictions Min -87.6228 +trainer/Q Targets Mean -72.0407 +trainer/Q Targets Std 16.0391 +trainer/Q Targets Max -2.99852 +trainer/Q Targets Min -87.6303 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000528468 +trainer/policy/mean Std 0.678391 +trainer/policy/mean Max 0.998692 +trainer/policy/mean Min -0.992921 +trainer/policy/std Mean 0.449413 +trainer/policy/std Std 0.022962 +trainer/policy/std Max 0.478774 +trainer/policy/std Min 0.410913 +trainer/Advantage Weights Mean 3.89007 +trainer/Advantage Weights Std 16.0592 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.15471e-10 +trainer/Advantage Score Mean -0.301656 +trainer/Advantage Score Std 0.472713 +trainer/Advantage Score Max 1.58179 +trainer/Advantage Score Min -2.2882 +trainer/V1 Predictions Mean -71.7566 +trainer/V1 Predictions Std 16.2878 +trainer/V1 Predictions Max -3.15625 +trainer/V1 Predictions Min -86.8668 +trainer/VF Loss 0.0549857 +expl/num steps total 97000 +expl/num paths total 97 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.119158 +expl/Actions Std 0.798786 +expl/Actions Max 2.27899 +expl/Actions Min -2.23771 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 96226 +eval/num paths total 97 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.039488 +eval/Actions Std 0.684523 +eval/Actions Max 0.997623 +eval/Actions Min -0.997459 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.86895e-06 +time/evaluation sampling (s) 3.23901 +time/exploration sampling (s) 4.10014 +time/logging (s) 0.00697955 +time/saving (s) 0.0131872 +time/training (s) 14.1775 +time/epoch (s) 21.5368 +time/total (s) 2173.36 +Epoch -904 +------------------------------ ---------------- +2022-05-15 18:39:00.391416 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -903 finished +------------------------------ ---------------- +epoch -903 +replay_buffer/size 999047 +trainer/num train calls 98000 +trainer/QF1 Loss 0.930122 +trainer/QF2 Loss 0.964849 +trainer/Policy Loss 22.7136 +trainer/Q1 Predictions Mean -69.2414 +trainer/Q1 Predictions Std 20.1782 +trainer/Q1 Predictions Max -1.34999 +trainer/Q1 Predictions Min -86.3148 +trainer/Q2 Predictions Mean -69.3391 +trainer/Q2 Predictions Std 20.068 +trainer/Q2 Predictions Max -1.5538 +trainer/Q2 Predictions Min -86.1417 +trainer/Q Targets Mean -69.3319 +trainer/Q Targets Std 20.2225 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3664 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0262726 +trainer/policy/mean Std 0.680097 +trainer/policy/mean Max 0.994088 +trainer/policy/mean Min -0.996739 +trainer/policy/std Mean 0.447585 +trainer/policy/std Std 0.0228448 +trainer/policy/std Max 0.476483 +trainer/policy/std Min 0.41102 +trainer/Advantage Weights Mean 3.79362 +trainer/Advantage Weights Std 16.3435 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.26155e-11 +trainer/Advantage Score Mean -0.314406 +trainer/Advantage Score Std 0.425863 +trainer/Advantage Score Max 0.972006 +trainer/Advantage Score Min -2.38788 +trainer/V1 Predictions Mean -69.1141 +trainer/V1 Predictions Std 20.2653 +trainer/V1 Predictions Max -1.5571 +trainer/V1 Predictions Min -86.1443 +trainer/VF Loss 0.042968 +expl/num steps total 98000 +expl/num paths total 98 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0354403 +expl/Actions Std 0.810714 +expl/Actions Max 2.40635 +expl/Actions Min -2.34759 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 97226 +eval/num paths total 98 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.107329 +eval/Actions Std 0.716193 +eval/Actions Max 0.991292 +eval/Actions Min -0.996177 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.65706e-06 +time/evaluation sampling (s) 3.31677 +time/exploration sampling (s) 3.77513 +time/logging (s) 0.00875205 +time/saving (s) 0.0123398 +time/training (s) 14.1037 +time/epoch (s) 21.2167 +time/total (s) 2194.58 +Epoch -903 +------------------------------ ---------------- +2022-05-15 18:39:21.375286 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -902 finished +------------------------------ ---------------- +epoch -902 +replay_buffer/size 999047 +trainer/num train calls 99000 +trainer/QF1 Loss 0.768961 +trainer/QF2 Loss 0.744576 +trainer/Policy Loss 1.32341 +trainer/Q1 Predictions Mean -70.7826 +trainer/Q1 Predictions Std 18.4729 +trainer/Q1 Predictions Max -1.66744 +trainer/Q1 Predictions Min -87.377 +trainer/Q2 Predictions Mean -70.7995 +trainer/Q2 Predictions Std 18.535 +trainer/Q2 Predictions Max -1.60289 +trainer/Q2 Predictions Min -87.3767 +trainer/Q Targets Mean -70.4272 +trainer/Q Targets Std 18.4357 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.858 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0157059 +trainer/policy/mean Std 0.690624 +trainer/policy/mean Max 0.998288 +trainer/policy/mean Min -0.997815 +trainer/policy/std Mean 0.447888 +trainer/policy/std Std 0.0221852 +trainer/policy/std Max 0.473177 +trainer/policy/std Min 0.412928 +trainer/Advantage Weights Mean 0.427657 +trainer/Advantage Weights Std 3.4995 +trainer/Advantage Weights Max 54.5449 +trainer/Advantage Weights Min 3.8313e-15 +trainer/Advantage Score Mean -0.60643 +trainer/Advantage Score Std 0.549751 +trainer/Advantage Score Max 0.399902 +trainer/Advantage Score Min -3.31956 +trainer/V1 Predictions Mean -70.132 +trainer/V1 Predictions Std 18.6506 +trainer/V1 Predictions Max 0.0625558 +trainer/V1 Predictions Min -86.7168 +trainer/VF Loss 0.0680969 +expl/num steps total 99000 +expl/num paths total 99 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.13303 +expl/Actions Std 0.813821 +expl/Actions Max 2.4348 +expl/Actions Min -2.58465 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 98035 +eval/num paths total 99 +eval/path length Mean 809 +eval/path length Std 0 +eval/path length Max 809 +eval/path length Min 809 +eval/Rewards Mean 0.00123609 +eval/Rewards Std 0.0351364 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0390279 +eval/Actions Std 0.69161 +eval/Actions Max 0.997579 +eval/Actions Min -0.995222 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.86009e-06 +time/evaluation sampling (s) 3.0457 +time/exploration sampling (s) 3.8456 +time/logging (s) 0.00874534 +time/saving (s) 0.011747 +time/training (s) 14.066 +time/epoch (s) 20.9778 +time/total (s) 2215.56 +Epoch -902 +------------------------------ ---------------- +2022-05-15 18:39:42.249360 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -901 finished +------------------------------ ---------------- +epoch -901 +replay_buffer/size 999047 +trainer/num train calls 100000 +trainer/QF1 Loss 1.04685 +trainer/QF2 Loss 1.17878 +trainer/Policy Loss 50.5126 +trainer/Q1 Predictions Mean -69.3714 +trainer/Q1 Predictions Std 19.2564 +trainer/Q1 Predictions Max -1.44249 +trainer/Q1 Predictions Min -85.5329 +trainer/Q2 Predictions Mean -69.328 +trainer/Q2 Predictions Std 19.2808 +trainer/Q2 Predictions Max -1.82815 +trainer/Q2 Predictions Min -85.6714 +trainer/Q Targets Mean -70.0217 +trainer/Q Targets Std 19.3683 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3956 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00878119 +trainer/policy/mean Std 0.693337 +trainer/policy/mean Max 0.999288 +trainer/policy/mean Min -0.999001 +trainer/policy/std Mean 0.447576 +trainer/policy/std Std 0.0232941 +trainer/policy/std Max 0.475523 +trainer/policy/std Min 0.408618 +trainer/Advantage Weights Mean 9.10244 +trainer/Advantage Weights Std 24.0634 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.0373e-10 +trainer/Advantage Score Mean -0.139517 +trainer/Advantage Score Std 0.454286 +trainer/Advantage Score Max 1.4866 +trainer/Advantage Score Min -2.29892 +trainer/V1 Predictions Mean -69.7913 +trainer/V1 Predictions Std 19.3208 +trainer/V1 Predictions Max -3.14143 +trainer/V1 Predictions Min -86.1698 +trainer/VF Loss 0.0635898 +expl/num steps total 100000 +expl/num paths total 100 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0422729 +expl/Actions Std 0.808377 +expl/Actions Max 2.57579 +expl/Actions Min -2.34976 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 99035 +eval/num paths total 100 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.200411 +eval/Actions Std 0.69693 +eval/Actions Max 0.999162 +eval/Actions Min -0.998375 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.07197e-06 +time/evaluation sampling (s) 3.12087 +time/exploration sampling (s) 3.39635 +time/logging (s) 0.0124376 +time/saving (s) 0.0179609 +time/training (s) 14.3204 +time/epoch (s) 20.868 +time/total (s) 2236.44 +Epoch -901 +------------------------------ ---------------- +2022-05-15 18:40:03.491664 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -900 finished +------------------------------ ---------------- +epoch -900 +replay_buffer/size 999047 +trainer/num train calls 101000 +trainer/QF1 Loss 1.03052 +trainer/QF2 Loss 0.941832 +trainer/Policy Loss 1.62201 +trainer/Q1 Predictions Mean -71.2637 +trainer/Q1 Predictions Std 18.6303 +trainer/Q1 Predictions Max -0.489052 +trainer/Q1 Predictions Min -86.5223 +trainer/Q2 Predictions Mean -71.2791 +trainer/Q2 Predictions Std 18.7021 +trainer/Q2 Predictions Max -0.380822 +trainer/Q2 Predictions Min -86.6208 +trainer/Q Targets Mean -70.7928 +trainer/Q Targets Std 18.9501 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7175 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00599635 +trainer/policy/mean Std 0.680667 +trainer/policy/mean Max 0.998361 +trainer/policy/mean Min -0.997074 +trainer/policy/std Mean 0.44544 +trainer/policy/std Std 0.0225966 +trainer/policy/std Max 0.475393 +trainer/policy/std Min 0.409884 +trainer/Advantage Weights Mean 0.262208 +trainer/Advantage Weights Std 1.79535 +trainer/Advantage Weights Max 25.9079 +trainer/Advantage Weights Min 2.18007e-18 +trainer/Advantage Score Mean -0.672387 +trainer/Advantage Score Std 0.57718 +trainer/Advantage Score Max 0.325455 +trainer/Advantage Score Min -4.06672 +trainer/V1 Predictions Mean -70.5616 +trainer/V1 Predictions Std 19.0139 +trainer/V1 Predictions Max -0.213525 +trainer/V1 Predictions Min -86.525 +trainer/VF Loss 0.0792536 +expl/num steps total 101000 +expl/num paths total 102 +expl/path length Mean 500 +expl/path length Std 310 +expl/path length Max 810 +expl/path length Min 190 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0145975 +expl/Actions Std 0.791554 +expl/Actions Max 2.44389 +expl/Actions Min -2.61291 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 100035 +eval/num paths total 101 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0526555 +eval/Actions Std 0.801698 +eval/Actions Max 0.997667 +eval/Actions Min -0.998159 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.19816e-06 +time/evaluation sampling (s) 3.38845 +time/exploration sampling (s) 3.43002 +time/logging (s) 0.00989435 +time/saving (s) 0.0288553 +time/training (s) 14.3732 +time/epoch (s) 21.2304 +time/total (s) 2257.68 +Epoch -900 +------------------------------ ---------------- +2022-05-15 18:40:24.520078 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -899 finished +------------------------------ ---------------- +epoch -899 +replay_buffer/size 999047 +trainer/num train calls 102000 +trainer/QF1 Loss 1.23889 +trainer/QF2 Loss 1.32746 +trainer/Policy Loss 21.0987 +trainer/Q1 Predictions Mean -70.1565 +trainer/Q1 Predictions Std 18.6442 +trainer/Q1 Predictions Max -1.4686 +trainer/Q1 Predictions Min -86.875 +trainer/Q2 Predictions Mean -70.1271 +trainer/Q2 Predictions Std 18.6113 +trainer/Q2 Predictions Max -1.82118 +trainer/Q2 Predictions Min -86.8951 +trainer/Q Targets Mean -70.5981 +trainer/Q Targets Std 18.549 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9472 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0310655 +trainer/policy/mean Std 0.69106 +trainer/policy/mean Max 0.997206 +trainer/policy/mean Min -0.998327 +trainer/policy/std Mean 0.447137 +trainer/policy/std Std 0.0223801 +trainer/policy/std Max 0.474949 +trainer/policy/std Min 0.413964 +trainer/Advantage Weights Mean 4.18499 +trainer/Advantage Weights Std 16.2184 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.36619e-15 +trainer/Advantage Score Mean -0.347038 +trainer/Advantage Score Std 0.528513 +trainer/Advantage Score Max 0.786463 +trainer/Advantage Score Min -3.28587 +trainer/V1 Predictions Mean -70.2691 +trainer/V1 Predictions Std 18.7089 +trainer/V1 Predictions Max -1.57746 +trainer/V1 Predictions Min -88.352 +trainer/VF Loss 0.0536055 +expl/num steps total 102000 +expl/num paths total 103 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0948978 +expl/Actions Std 0.861513 +expl/Actions Max 2.49141 +expl/Actions Min -2.54006 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 101035 +eval/num paths total 102 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0169493 +eval/Actions Std 0.689758 +eval/Actions Max 0.999703 +eval/Actions Min -0.996112 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.23961e-06 +time/evaluation sampling (s) 2.78427 +time/exploration sampling (s) 3.58946 +time/logging (s) 0.0112369 +time/saving (s) 0.0155993 +time/training (s) 14.6194 +time/epoch (s) 21.0199 +time/total (s) 2278.7 +Epoch -899 +------------------------------ ---------------- +2022-05-15 18:40:46.394329 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -898 finished +------------------------------ ---------------- +epoch -898 +replay_buffer/size 999047 +trainer/num train calls 103000 +trainer/QF1 Loss 1.42004 +trainer/QF2 Loss 1.10877 +trainer/Policy Loss 8.18103 +trainer/Q1 Predictions Mean -70.5902 +trainer/Q1 Predictions Std 18.6507 +trainer/Q1 Predictions Max -2.50036 +trainer/Q1 Predictions Min -87.537 +trainer/Q2 Predictions Mean -70.3737 +trainer/Q2 Predictions Std 18.6905 +trainer/Q2 Predictions Max -2.3607 +trainer/Q2 Predictions Min -86.9687 +trainer/Q Targets Mean -69.9581 +trainer/Q Targets Std 18.9266 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6842 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0132235 +trainer/policy/mean Std 0.695235 +trainer/policy/mean Max 0.997204 +trainer/policy/mean Min -0.997157 +trainer/policy/std Mean 0.448085 +trainer/policy/std Std 0.0230687 +trainer/policy/std Max 0.474822 +trainer/policy/std Min 0.414936 +trainer/Advantage Weights Mean 1.6537 +trainer/Advantage Weights Std 9.88374 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.07685e-11 +trainer/Advantage Score Mean -0.415355 +trainer/Advantage Score Std 0.4202 +trainer/Advantage Score Max 0.555817 +trainer/Advantage Score Min -2.52544 +trainer/V1 Predictions Mean -69.6959 +trainer/V1 Predictions Std 19.0009 +trainer/V1 Predictions Max -1.28491 +trainer/V1 Predictions Min -86.6043 +trainer/VF Loss 0.038975 +expl/num steps total 103000 +expl/num paths total 104 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00373702 +expl/Actions Std 0.772235 +expl/Actions Max 2.39652 +expl/Actions Min -2.35917 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 102035 +eval/num paths total 103 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0773261 +eval/Actions Std 0.621685 +eval/Actions Max 0.998087 +eval/Actions Min -0.996467 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78279e-06 +time/evaluation sampling (s) 3.03575 +time/exploration sampling (s) 3.93902 +time/logging (s) 0.00822344 +time/saving (s) 0.0101431 +time/training (s) 14.869 +time/epoch (s) 21.8622 +time/total (s) 2300.57 +Epoch -898 +------------------------------ ---------------- +2022-05-15 18:41:08.026107 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -897 finished +------------------------------ ---------------- +epoch -897 +replay_buffer/size 999047 +trainer/num train calls 104000 +trainer/QF1 Loss 0.696807 +trainer/QF2 Loss 0.64671 +trainer/Policy Loss 32.9299 +trainer/Q1 Predictions Mean -71.9415 +trainer/Q1 Predictions Std 16.005 +trainer/Q1 Predictions Max -1.37765 +trainer/Q1 Predictions Min -86.3569 +trainer/Q2 Predictions Mean -71.946 +trainer/Q2 Predictions Std 16.0189 +trainer/Q2 Predictions Max -1.37119 +trainer/Q2 Predictions Min -86.3097 +trainer/Q Targets Mean -72.4517 +trainer/Q Targets Std 15.7476 +trainer/Q Targets Max -3.95155 +trainer/Q Targets Min -86.3537 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.015223 +trainer/policy/mean Std 0.698107 +trainer/policy/mean Max 0.998243 +trainer/policy/mean Min -0.996912 +trainer/policy/std Mean 0.447557 +trainer/policy/std Std 0.0230034 +trainer/policy/std Max 0.475843 +trainer/policy/std Min 0.414696 +trainer/Advantage Weights Mean 7.26193 +trainer/Advantage Weights Std 21.1333 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.35359e-09 +trainer/Advantage Score Mean -0.146333 +trainer/Advantage Score Std 0.42288 +trainer/Advantage Score Max 1.25191 +trainer/Advantage Score Min -1.98673 +trainer/V1 Predictions Mean -72.145 +trainer/V1 Predictions Std 16.0313 +trainer/V1 Predictions Max -1.07535 +trainer/V1 Predictions Min -86.2103 +trainer/VF Loss 0.0512967 +expl/num steps total 104000 +expl/num paths total 105 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0229285 +expl/Actions Std 0.831735 +expl/Actions Max 2.31478 +expl/Actions Min -2.48183 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 102623 +eval/num paths total 104 +eval/path length Mean 588 +eval/path length Std 0 +eval/path length Max 588 +eval/path length Min 588 +eval/Rewards Mean 0.00170068 +eval/Rewards Std 0.0412042 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00591096 +eval/Actions Std 0.726133 +eval/Actions Max 0.99928 +eval/Actions Min -0.997537 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.05591e-06 +time/evaluation sampling (s) 3.20394 +time/exploration sampling (s) 3.80124 +time/logging (s) 0.0063617 +time/saving (s) 0.0127657 +time/training (s) 14.5991 +time/epoch (s) 21.6234 +time/total (s) 2322.2 +Epoch -897 +------------------------------ ---------------- +2022-05-15 18:41:30.483598 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -896 finished +------------------------------ ---------------- +epoch -896 +replay_buffer/size 999047 +trainer/num train calls 105000 +trainer/QF1 Loss 0.75409 +trainer/QF2 Loss 0.87044 +trainer/Policy Loss 7.73475 +trainer/Q1 Predictions Mean -72.1269 +trainer/Q1 Predictions Std 15.8155 +trainer/Q1 Predictions Max -0.660747 +trainer/Q1 Predictions Min -87.015 +trainer/Q2 Predictions Mean -72.1647 +trainer/Q2 Predictions Std 15.7466 +trainer/Q2 Predictions Max -0.6324 +trainer/Q2 Predictions Min -86.6665 +trainer/Q Targets Mean -71.6332 +trainer/Q Targets Std 15.9792 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6851 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00841269 +trainer/policy/mean Std 0.685706 +trainer/policy/mean Max 0.997767 +trainer/policy/mean Min -0.996951 +trainer/policy/std Mean 0.446142 +trainer/policy/std Std 0.0222734 +trainer/policy/std Max 0.473442 +trainer/policy/std Min 0.410579 +trainer/Advantage Weights Mean 2.00841 +trainer/Advantage Weights Std 11.8682 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.28337e-12 +trainer/Advantage Score Mean -0.456487 +trainer/Advantage Score Std 0.454601 +trainer/Advantage Score Max 0.811783 +trainer/Advantage Score Min -2.73815 +trainer/V1 Predictions Mean -71.4787 +trainer/V1 Predictions Std 15.9157 +trainer/V1 Predictions Max 0.341454 +trainer/V1 Predictions Min -86.5247 +trainer/VF Loss 0.0489312 +expl/num steps total 105000 +expl/num paths total 106 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.105622 +expl/Actions Std 0.8108 +expl/Actions Max 2.34436 +expl/Actions Min -2.51525 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 103623 +eval/num paths total 105 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.199651 +eval/Actions Std 0.661658 +eval/Actions Max 0.998231 +eval/Actions Min -0.995755 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.44379e-06 +time/evaluation sampling (s) 3.32305 +time/exploration sampling (s) 3.94905 +time/logging (s) 0.00776041 +time/saving (s) 0.0127909 +time/training (s) 15.161 +time/epoch (s) 22.4536 +time/total (s) 2344.66 +Epoch -896 +------------------------------ ---------------- +2022-05-15 18:41:52.690185 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -895 finished +------------------------------ ---------------- +epoch -895 +replay_buffer/size 999047 +trainer/num train calls 106000 +trainer/QF1 Loss 1.15968 +trainer/QF2 Loss 1.19189 +trainer/Policy Loss 2.111 +trainer/Q1 Predictions Mean -71.1278 +trainer/Q1 Predictions Std 18.7374 +trainer/Q1 Predictions Max -0.956154 +trainer/Q1 Predictions Min -87.0864 +trainer/Q2 Predictions Mean -71.1409 +trainer/Q2 Predictions Std 18.6887 +trainer/Q2 Predictions Max -1.6156 +trainer/Q2 Predictions Min -87.2003 +trainer/Q Targets Mean -70.5468 +trainer/Q Targets Std 18.8505 +trainer/Q Targets Max -0.811381 +trainer/Q Targets Min -86.4189 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0210754 +trainer/policy/mean Std 0.691273 +trainer/policy/mean Max 0.999247 +trainer/policy/mean Min -0.995861 +trainer/policy/std Mean 0.447104 +trainer/policy/std Std 0.0231021 +trainer/policy/std Max 0.476585 +trainer/policy/std Min 0.410607 +trainer/Advantage Weights Mean 0.378272 +trainer/Advantage Weights Std 2.25286 +trainer/Advantage Weights Max 26.1482 +trainer/Advantage Weights Min 1.05415e-18 +trainer/Advantage Score Mean -0.65041 +trainer/Advantage Score Std 0.643656 +trainer/Advantage Score Max 0.326378 +trainer/Advantage Score Min -4.13938 +trainer/V1 Predictions Mean -70.1207 +trainer/V1 Predictions Std 19.2425 +trainer/V1 Predictions Max 0.931614 +trainer/V1 Predictions Min -86.3212 +trainer/VF Loss 0.0848551 +expl/num steps total 106000 +expl/num paths total 107 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0201908 +expl/Actions Std 0.801363 +expl/Actions Max 2.51586 +expl/Actions Min -2.34198 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 104623 +eval/num paths total 106 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.322287 +eval/Actions Std 0.632439 +eval/Actions Max 0.997488 +eval/Actions Min -0.997431 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.89828e-06 +time/evaluation sampling (s) 3.37643 +time/exploration sampling (s) 4.12248 +time/logging (s) 0.00897047 +time/saving (s) 0.0107734 +time/training (s) 14.6828 +time/epoch (s) 22.2014 +time/total (s) 2366.86 +Epoch -895 +------------------------------ ---------------- +2022-05-15 18:42:14.401231 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -894 finished +------------------------------ ---------------- +epoch -894 +replay_buffer/size 999047 +trainer/num train calls 107000 +trainer/QF1 Loss 1.02946 +trainer/QF2 Loss 1.0927 +trainer/Policy Loss 6.83343 +trainer/Q1 Predictions Mean -70.6543 +trainer/Q1 Predictions Std 19.2 +trainer/Q1 Predictions Max -0.736088 +trainer/Q1 Predictions Min -86.6014 +trainer/Q2 Predictions Mean -70.6709 +trainer/Q2 Predictions Std 19.1844 +trainer/Q2 Predictions Max -0.942742 +trainer/Q2 Predictions Min -86.6074 +trainer/Q Targets Mean -70.2835 +trainer/Q Targets Std 19.7682 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9342 +trainer/rewards Mean -0.976562 +trainer/rewards Std 0.151288 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0234375 +trainer/terminals Std 0.151288 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0191243 +trainer/policy/mean Std 0.688498 +trainer/policy/mean Max 0.996916 +trainer/policy/mean Min -0.996073 +trainer/policy/std Mean 0.447702 +trainer/policy/std Std 0.0231069 +trainer/policy/std Max 0.474364 +trainer/policy/std Min 0.409842 +trainer/Advantage Weights Mean 0.943459 +trainer/Advantage Weights Std 6.74707 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.57545e-15 +trainer/Advantage Score Mean -0.53126 +trainer/Advantage Score Std 0.554577 +trainer/Advantage Score Max 0.857288 +trainer/Advantage Score Min -3.32647 +trainer/V1 Predictions Mean -70.1698 +trainer/V1 Predictions Std 19.4115 +trainer/V1 Predictions Max -0.139391 +trainer/V1 Predictions Min -87.0915 +trainer/VF Loss 0.0629259 +expl/num steps total 107000 +expl/num paths total 108 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0937827 +expl/Actions Std 0.84375 +expl/Actions Max 2.34021 +expl/Actions Min -2.81312 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 105623 +eval/num paths total 107 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0390691 +eval/Actions Std 0.773868 +eval/Actions Max 0.998862 +eval/Actions Min -0.998347 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.77581e-06 +time/evaluation sampling (s) 3.33644 +time/exploration sampling (s) 3.74642 +time/logging (s) 0.00792773 +time/saving (s) 0.0117258 +time/training (s) 14.6016 +time/epoch (s) 21.7041 +time/total (s) 2388.57 +Epoch -894 +------------------------------ ---------------- +2022-05-15 18:42:35.248043 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -893 finished +------------------------------ ---------------- +epoch -893 +replay_buffer/size 999047 +trainer/num train calls 108000 +trainer/QF1 Loss 1.04367 +trainer/QF2 Loss 1.07043 +trainer/Policy Loss 4.95888 +trainer/Q1 Predictions Mean -69.7258 +trainer/Q1 Predictions Std 19.3334 +trainer/Q1 Predictions Max -1.53566 +trainer/Q1 Predictions Min -87.3119 +trainer/Q2 Predictions Mean -69.737 +trainer/Q2 Predictions Std 19.3516 +trainer/Q2 Predictions Max -1.59074 +trainer/Q2 Predictions Min -87.2746 +trainer/Q Targets Mean -69.2823 +trainer/Q Targets Std 19.5896 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9931 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0294175 +trainer/policy/mean Std 0.710055 +trainer/policy/mean Max 0.998503 +trainer/policy/mean Min -0.998519 +trainer/policy/std Mean 0.447657 +trainer/policy/std Std 0.0236913 +trainer/policy/std Max 0.47577 +trainer/policy/std Min 0.409807 +trainer/Advantage Weights Mean 1.46961 +trainer/Advantage Weights Std 10.0946 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.37874e-14 +trainer/Advantage Score Mean -0.624451 +trainer/Advantage Score Std 0.517367 +trainer/Advantage Score Max 0.729013 +trainer/Advantage Score Min -3.07594 +trainer/V1 Predictions Mean -69.035 +trainer/V1 Predictions Std 19.562 +trainer/V1 Predictions Max -2.13595 +trainer/V1 Predictions Min -86.821 +trainer/VF Loss 0.0698144 +expl/num steps total 108000 +expl/num paths total 109 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0616704 +expl/Actions Std 0.815201 +expl/Actions Max 2.4431 +expl/Actions Min -2.20125 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 106314 +eval/num paths total 108 +eval/path length Mean 691 +eval/path length Std 0 +eval/path length Max 691 +eval/path length Min 691 +eval/Rewards Mean 0.00144718 +eval/Rewards Std 0.0380143 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0293442 +eval/Actions Std 0.716982 +eval/Actions Max 0.999092 +eval/Actions Min -0.999459 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.88896e-06 +time/evaluation sampling (s) 3.07102 +time/exploration sampling (s) 3.33208 +time/logging (s) 0.00642798 +time/saving (s) 0.0110281 +time/training (s) 14.4175 +time/epoch (s) 20.8381 +time/total (s) 2409.42 +Epoch -893 +------------------------------ ---------------- +2022-05-15 18:42:56.075870 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -892 finished +------------------------------ ---------------- +epoch -892 +replay_buffer/size 999047 +trainer/num train calls 109000 +trainer/QF1 Loss 1.3329 +trainer/QF2 Loss 1.67605 +trainer/Policy Loss 10.8095 +trainer/Q1 Predictions Mean -70.7943 +trainer/Q1 Predictions Std 17.837 +trainer/Q1 Predictions Max -0.489034 +trainer/Q1 Predictions Min -86.9628 +trainer/Q2 Predictions Mean -70.8761 +trainer/Q2 Predictions Std 17.9329 +trainer/Q2 Predictions Max -0.353038 +trainer/Q2 Predictions Min -86.8778 +trainer/Q Targets Mean -70.422 +trainer/Q Targets Std 18.357 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5692 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.018904 +trainer/policy/mean Std 0.702303 +trainer/policy/mean Max 0.996179 +trainer/policy/mean Min -0.996247 +trainer/policy/std Mean 0.447974 +trainer/policy/std Std 0.0231947 +trainer/policy/std Max 0.475226 +trainer/policy/std Min 0.408642 +trainer/Advantage Weights Mean 1.70214 +trainer/Advantage Weights Std 11.3173 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.43018e-19 +trainer/Advantage Score Mean -0.748399 +trainer/Advantage Score Std 0.759985 +trainer/Advantage Score Max 0.666321 +trainer/Advantage Score Min -4.16173 +trainer/V1 Predictions Mean -70.0962 +trainer/V1 Predictions Std 18.5483 +trainer/V1 Predictions Max 2.48551 +trainer/V1 Predictions Min -86.9202 +trainer/VF Loss 0.118485 +expl/num steps total 109000 +expl/num paths total 110 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0112314 +expl/Actions Std 0.80418 +expl/Actions Max 2.37798 +expl/Actions Min -2.20454 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 107314 +eval/num paths total 109 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0466031 +eval/Actions Std 0.706238 +eval/Actions Max 0.9982 +eval/Actions Min -0.998683 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.1558e-06 +time/evaluation sampling (s) 2.86246 +time/exploration sampling (s) 3.59124 +time/logging (s) 0.00717631 +time/saving (s) 0.0118531 +time/training (s) 14.3497 +time/epoch (s) 20.8224 +time/total (s) 2430.24 +Epoch -892 +------------------------------ ---------------- +2022-05-15 18:43:16.937153 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -891 finished +------------------------------ ---------------- +epoch -891 +replay_buffer/size 999047 +trainer/num train calls 110000 +trainer/QF1 Loss 0.7709 +trainer/QF2 Loss 0.84019 +trainer/Policy Loss 21.9169 +trainer/Q1 Predictions Mean -70.5634 +trainer/Q1 Predictions Std 18.0864 +trainer/Q1 Predictions Max -1.14355 +trainer/Q1 Predictions Min -87.8987 +trainer/Q2 Predictions Mean -70.6008 +trainer/Q2 Predictions Std 18.1473 +trainer/Q2 Predictions Max -1.2475 +trainer/Q2 Predictions Min -87.9653 +trainer/Q Targets Mean -70.6398 +trainer/Q Targets Std 17.9645 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0805 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0261203 +trainer/policy/mean Std 0.694021 +trainer/policy/mean Max 0.99715 +trainer/policy/mean Min -0.995917 +trainer/policy/std Mean 0.446493 +trainer/policy/std Std 0.0228653 +trainer/policy/std Max 0.472651 +trainer/policy/std Min 0.406622 +trainer/Advantage Weights Mean 4.26133 +trainer/Advantage Weights Std 17.4473 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.24675e-11 +trainer/Advantage Score Mean -0.334083 +trainer/Advantage Score Std 0.482299 +trainer/Advantage Score Max 1.51764 +trainer/Advantage Score Min -2.51079 +trainer/V1 Predictions Mean -70.2719 +trainer/V1 Predictions Std 18.0806 +trainer/V1 Predictions Max -0.903702 +trainer/V1 Predictions Min -87.4855 +trainer/VF Loss 0.0540998 +expl/num steps total 110000 +expl/num paths total 112 +expl/path length Mean 500 +expl/path length Std 470 +expl/path length Max 970 +expl/path length Min 30 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0177322 +expl/Actions Std 0.82558 +expl/Actions Max 2.50999 +expl/Actions Min -2.38532 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 108242 +eval/num paths total 110 +eval/path length Mean 928 +eval/path length Std 0 +eval/path length Max 928 +eval/path length Min 928 +eval/Rewards Mean 0.00107759 +eval/Rewards Std 0.0328089 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00264161 +eval/Actions Std 0.712264 +eval/Actions Max 0.999719 +eval/Actions Min -0.998131 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.8708e-06 +time/evaluation sampling (s) 2.80999 +time/exploration sampling (s) 3.86289 +time/logging (s) 0.0105049 +time/saving (s) 0.0151448 +time/training (s) 14.1589 +time/epoch (s) 20.8574 +time/total (s) 2451.11 +Epoch -891 +------------------------------ ---------------- +2022-05-15 18:43:37.712109 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -890 finished +------------------------------ ---------------- +epoch -890 +replay_buffer/size 999047 +trainer/num train calls 111000 +trainer/QF1 Loss 0.906402 +trainer/QF2 Loss 1.06993 +trainer/Policy Loss 17.9272 +trainer/Q1 Predictions Mean -71.8042 +trainer/Q1 Predictions Std 16.4567 +trainer/Q1 Predictions Max -0.476033 +trainer/Q1 Predictions Min -86.6431 +trainer/Q2 Predictions Mean -71.7615 +trainer/Q2 Predictions Std 16.5222 +trainer/Q2 Predictions Max -0.381282 +trainer/Q2 Predictions Min -86.3753 +trainer/Q Targets Mean -72.0375 +trainer/Q Targets Std 15.9496 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9781 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00562686 +trainer/policy/mean Std 0.695477 +trainer/policy/mean Max 0.999122 +trainer/policy/mean Min -0.9994 +trainer/policy/std Mean 0.446792 +trainer/policy/std Std 0.0212534 +trainer/policy/std Max 0.47427 +trainer/policy/std Min 0.413187 +trainer/Advantage Weights Mean 5.41464 +trainer/Advantage Weights Std 18.7923 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.38769e-10 +trainer/Advantage Score Mean -0.336478 +trainer/Advantage Score Std 0.472066 +trainer/Advantage Score Max 2.06198 +trainer/Advantage Score Min -2.21555 +trainer/V1 Predictions Mean -71.7712 +trainer/V1 Predictions Std 16.1578 +trainer/V1 Predictions Max -1.57358 +trainer/V1 Predictions Min -85.7102 +trainer/VF Loss 0.0626681 +expl/num steps total 111000 +expl/num paths total 113 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0351962 +expl/Actions Std 0.845013 +expl/Actions Max 2.29244 +expl/Actions Min -2.32723 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 109242 +eval/num paths total 111 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0146612 +eval/Actions Std 0.681453 +eval/Actions Max 0.999198 +eval/Actions Min -0.997445 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.87406e-06 +time/evaluation sampling (s) 3.38634 +time/exploration sampling (s) 3.62708 +time/logging (s) 0.00678578 +time/saving (s) 0.0093575 +time/training (s) 13.7316 +time/epoch (s) 20.7612 +time/total (s) 2471.87 +Epoch -890 +------------------------------ ---------------- +2022-05-15 18:43:59.098026 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -889 finished +------------------------------ ---------------- +epoch -889 +replay_buffer/size 999047 +trainer/num train calls 112000 +trainer/QF1 Loss 1.17282 +trainer/QF2 Loss 1.31554 +trainer/Policy Loss 68.8454 +trainer/Q1 Predictions Mean -70.4222 +trainer/Q1 Predictions Std 18.6992 +trainer/Q1 Predictions Max -0.394915 +trainer/Q1 Predictions Min -88.16 +trainer/Q2 Predictions Mean -70.3409 +trainer/Q2 Predictions Std 18.7428 +trainer/Q2 Predictions Max -0.505098 +trainer/Q2 Predictions Min -88.2538 +trainer/Q Targets Mean -71.0738 +trainer/Q Targets Std 18.7211 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8264 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00257751 +trainer/policy/mean Std 0.700386 +trainer/policy/mean Max 0.999615 +trainer/policy/mean Min -0.995951 +trainer/policy/std Mean 0.448616 +trainer/policy/std Std 0.0214258 +trainer/policy/std Max 0.476462 +trainer/policy/std Min 0.417682 +trainer/Advantage Weights Mean 14.4453 +trainer/Advantage Weights Std 30.2008 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.80131e-08 +trainer/Advantage Score Mean -0.0318075 +trainer/Advantage Score Std 0.458038 +trainer/Advantage Score Max 1.66773 +trainer/Advantage Score Min -1.66626 +trainer/V1 Predictions Mean -70.7719 +trainer/V1 Predictions Std 18.8374 +trainer/V1 Predictions Max 0.772624 +trainer/V1 Predictions Min -88.2094 +trainer/VF Loss 0.084614 +expl/num steps total 112000 +expl/num paths total 114 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.043283 +expl/Actions Std 0.795095 +expl/Actions Max 2.61772 +expl/Actions Min -2.46083 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 110242 +eval/num paths total 112 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.198466 +eval/Actions Std 0.674628 +eval/Actions Max 0.997424 +eval/Actions Min -0.993573 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.80468e-06 +time/evaluation sampling (s) 3.2072 +time/exploration sampling (s) 3.91954 +time/logging (s) 0.0082027 +time/saving (s) 0.0113332 +time/training (s) 14.2353 +time/epoch (s) 21.3816 +time/total (s) 2493.26 +Epoch -889 +------------------------------ ---------------- +2022-05-15 18:44:20.762512 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -888 finished +------------------------------ ---------------- +epoch -888 +replay_buffer/size 999047 +trainer/num train calls 113000 +trainer/QF1 Loss 0.78899 +trainer/QF2 Loss 0.832194 +trainer/Policy Loss 10.4416 +trainer/Q1 Predictions Mean -71.7532 +trainer/Q1 Predictions Std 16.3726 +trainer/Q1 Predictions Max -3.26171 +trainer/Q1 Predictions Min -87.3994 +trainer/Q2 Predictions Mean -71.7828 +trainer/Q2 Predictions Std 16.3619 +trainer/Q2 Predictions Max -3.55055 +trainer/Q2 Predictions Min -87.5322 +trainer/Q Targets Mean -71.3202 +trainer/Q Targets Std 16.4454 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0533 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00633238 +trainer/policy/mean Std 0.678577 +trainer/policy/mean Max 0.996426 +trainer/policy/mean Min -0.992579 +trainer/policy/std Mean 0.44667 +trainer/policy/std Std 0.022093 +trainer/policy/std Max 0.473003 +trainer/policy/std Min 0.412747 +trainer/Advantage Weights Mean 1.82168 +trainer/Advantage Weights Std 12.4231 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.88921e-17 +trainer/Advantage Score Mean -0.577736 +trainer/Advantage Score Std 0.512208 +trainer/Advantage Score Max 0.692078 +trainer/Advantage Score Min -3.8083 +trainer/V1 Predictions Mean -71.0389 +trainer/V1 Predictions Std 16.5789 +trainer/V1 Predictions Max -2.92709 +trainer/V1 Predictions Min -86.377 +trainer/VF Loss 0.0648954 +expl/num steps total 113000 +expl/num paths total 115 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.071196 +expl/Actions Std 0.804946 +expl/Actions Max 2.27751 +expl/Actions Min -2.35587 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 111242 +eval/num paths total 113 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0235704 +eval/Actions Std 0.782411 +eval/Actions Max 0.994746 +eval/Actions Min -0.995753 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.99491e-06 +time/evaluation sampling (s) 3.34677 +time/exploration sampling (s) 3.88124 +time/logging (s) 0.00851584 +time/saving (s) 0.0113152 +time/training (s) 14.4106 +time/epoch (s) 21.6585 +time/total (s) 2514.92 +Epoch -888 +------------------------------ ---------------- +2022-05-15 18:44:41.755742 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -887 finished +------------------------------ ---------------- +epoch -887 +replay_buffer/size 999047 +trainer/num train calls 114000 +trainer/QF1 Loss 0.639698 +trainer/QF2 Loss 0.683384 +trainer/Policy Loss 27.8002 +trainer/Q1 Predictions Mean -72.5661 +trainer/Q1 Predictions Std 16.6189 +trainer/Q1 Predictions Max -0.81421 +trainer/Q1 Predictions Min -86.4178 +trainer/Q2 Predictions Mean -72.519 +trainer/Q2 Predictions Std 16.6563 +trainer/Q2 Predictions Max -1.01664 +trainer/Q2 Predictions Min -86.2241 +trainer/Q Targets Mean -72.7368 +trainer/Q Targets Std 16.4182 +trainer/Q Targets Max 0.909593 +trainer/Q Targets Min -86.0495 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00623338 +trainer/policy/mean Std 0.701919 +trainer/policy/mean Max 0.99851 +trainer/policy/mean Min -0.998001 +trainer/policy/std Mean 0.444371 +trainer/policy/std Std 0.0217708 +trainer/policy/std Max 0.468374 +trainer/policy/std Min 0.408578 +trainer/Advantage Weights Mean 5.88382 +trainer/Advantage Weights Std 19.4471 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.80289e-10 +trainer/Advantage Score Mean -0.283292 +trainer/Advantage Score Std 0.456931 +trainer/Advantage Score Max 1.12806 +trainer/Advantage Score Min -2.19952 +trainer/V1 Predictions Mean -72.4506 +trainer/V1 Predictions Std 16.5862 +trainer/V1 Predictions Max -0.663038 +trainer/V1 Predictions Min -85.9088 +trainer/VF Loss 0.0502217 +expl/num steps total 114000 +expl/num paths total 116 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0607155 +expl/Actions Std 0.804412 +expl/Actions Max 2.51561 +expl/Actions Min -2.4273 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 112242 +eval/num paths total 114 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00374485 +eval/Actions Std 0.706951 +eval/Actions Max 0.999176 +eval/Actions Min -0.997714 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.73297e-06 +time/evaluation sampling (s) 2.85915 +time/exploration sampling (s) 3.59098 +time/logging (s) 0.0108502 +time/saving (s) 0.0149698 +time/training (s) 14.5131 +time/epoch (s) 20.9891 +time/total (s) 2535.92 +Epoch -887 +------------------------------ ---------------- +2022-05-15 18:45:03.697045 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -886 finished +------------------------------ ---------------- +epoch -886 +replay_buffer/size 999047 +trainer/num train calls 115000 +trainer/QF1 Loss 1.43847 +trainer/QF2 Loss 1.43769 +trainer/Policy Loss 19.958 +trainer/Q1 Predictions Mean -68.5954 +trainer/Q1 Predictions Std 20.4206 +trainer/Q1 Predictions Max -0.516037 +trainer/Q1 Predictions Min -86.5245 +trainer/Q2 Predictions Mean -68.6296 +trainer/Q2 Predictions Std 20.3546 +trainer/Q2 Predictions Max -0.758199 +trainer/Q2 Predictions Min -86.3799 +trainer/Q Targets Mean -68.9238 +trainer/Q Targets Std 20.2193 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0366 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0129476 +trainer/policy/mean Std 0.691238 +trainer/policy/mean Max 0.999672 +trainer/policy/mean Min -0.994247 +trainer/policy/std Mean 0.446166 +trainer/policy/std Std 0.0220901 +trainer/policy/std Max 0.472794 +trainer/policy/std Min 0.411121 +trainer/Advantage Weights Mean 4.41176 +trainer/Advantage Weights Std 18.3967 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.82733e-12 +trainer/Advantage Score Mean -0.380146 +trainer/Advantage Score Std 0.471183 +trainer/Advantage Score Max 0.998129 +trainer/Advantage Score Min -2.62889 +trainer/V1 Predictions Mean -68.6818 +trainer/V1 Predictions Std 20.3772 +trainer/V1 Predictions Max -0.612919 +trainer/V1 Predictions Min -86.2996 +trainer/VF Loss 0.0505004 +expl/num steps total 115000 +expl/num paths total 117 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0402788 +expl/Actions Std 0.807275 +expl/Actions Max 2.30682 +expl/Actions Min -2.25316 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 113242 +eval/num paths total 115 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0513847 +eval/Actions Std 0.727804 +eval/Actions Max 0.999087 +eval/Actions Min -0.997847 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.27105e-06 +time/evaluation sampling (s) 3.00955 +time/exploration sampling (s) 3.81102 +time/logging (s) 0.00916674 +time/saving (s) 0.0130665 +time/training (s) 15.0893 +time/epoch (s) 21.9321 +time/total (s) 2557.85 +Epoch -886 +------------------------------ ---------------- +2022-05-15 18:45:24.844420 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -885 finished +------------------------------ ---------------- +epoch -885 +replay_buffer/size 999047 +trainer/num train calls 116000 +trainer/QF1 Loss 1.0626 +trainer/QF2 Loss 1.12552 +trainer/Policy Loss 124.409 +trainer/Q1 Predictions Mean -69.5686 +trainer/Q1 Predictions Std 19.7701 +trainer/Q1 Predictions Max -1.32042 +trainer/Q1 Predictions Min -86.9311 +trainer/Q2 Predictions Mean -69.5132 +trainer/Q2 Predictions Std 19.8482 +trainer/Q2 Predictions Max -1.37866 +trainer/Q2 Predictions Min -86.6932 +trainer/Q Targets Mean -70.1773 +trainer/Q Targets Std 19.8879 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2441 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0132609 +trainer/policy/mean Std 0.690499 +trainer/policy/mean Max 0.99806 +trainer/policy/mean Min -0.998367 +trainer/policy/std Mean 0.444266 +trainer/policy/std Std 0.0230133 +trainer/policy/std Max 0.474137 +trainer/policy/std Min 0.409208 +trainer/Advantage Weights Mean 24.9616 +trainer/Advantage Weights Std 36.8062 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.06891e-12 +trainer/Advantage Score Mean 0.068201 +trainer/Advantage Score Std 0.533267 +trainer/Advantage Score Max 1.48225 +trainer/Advantage Score Min -2.62276 +trainer/V1 Predictions Mean -69.9424 +trainer/V1 Predictions Std 19.9859 +trainer/V1 Predictions Max -0.986743 +trainer/V1 Predictions Min -87.1549 +trainer/VF Loss 0.133166 +expl/num steps total 116000 +expl/num paths total 118 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0607191 +expl/Actions Std 0.805052 +expl/Actions Max 2.35444 +expl/Actions Min -2.49095 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 114242 +eval/num paths total 116 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0440864 +eval/Actions Std 0.726996 +eval/Actions Max 0.998932 +eval/Actions Min -0.998985 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.7861e-06 +time/evaluation sampling (s) 3.21582 +time/exploration sampling (s) 3.38646 +time/logging (s) 0.00954407 +time/saving (s) 0.0138559 +time/training (s) 14.5161 +time/epoch (s) 21.1418 +time/total (s) 2579 +Epoch -885 +------------------------------ ---------------- +2022-05-15 18:45:45.128270 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -884 finished +------------------------------ ---------------- +epoch -884 +replay_buffer/size 999047 +trainer/num train calls 117000 +trainer/QF1 Loss 1.08472 +trainer/QF2 Loss 1.20708 +trainer/Policy Loss 33.1757 +trainer/Q1 Predictions Mean -69.2462 +trainer/Q1 Predictions Std 20.3863 +trainer/Q1 Predictions Max -1.10787 +trainer/Q1 Predictions Min -86.8453 +trainer/Q2 Predictions Mean -69.2279 +trainer/Q2 Predictions Std 20.4386 +trainer/Q2 Predictions Max -0.564255 +trainer/Q2 Predictions Min -87.6151 +trainer/Q Targets Mean -69.2108 +trainer/Q Targets Std 20.5432 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4638 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0134666 +trainer/policy/mean Std 0.679485 +trainer/policy/mean Max 0.995226 +trainer/policy/mean Min -0.990371 +trainer/policy/std Mean 0.446384 +trainer/policy/std Std 0.0241864 +trainer/policy/std Max 0.477083 +trainer/policy/std Min 0.405031 +trainer/Advantage Weights Mean 5.28681 +trainer/Advantage Weights Std 18.0249 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.0683e-14 +trainer/Advantage Score Mean -0.258083 +trainer/Advantage Score Std 0.439275 +trainer/Advantage Score Max 0.587463 +trainer/Advantage Score Min -3.02806 +trainer/V1 Predictions Mean -68.9685 +trainer/V1 Predictions Std 20.501 +trainer/V1 Predictions Max -0.422383 +trainer/V1 Predictions Min -86.5599 +trainer/VF Loss 0.039204 +expl/num steps total 117000 +expl/num paths total 119 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0438949 +expl/Actions Std 0.820756 +expl/Actions Max 2.37518 +expl/Actions Min -2.4852 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 115242 +eval/num paths total 117 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.140079 +eval/Actions Std 0.630751 +eval/Actions Max 0.998408 +eval/Actions Min -0.999476 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.62914e-06 +time/evaluation sampling (s) 3.11375 +time/exploration sampling (s) 3.25482 +time/logging (s) 0.0071965 +time/saving (s) 0.0101968 +time/training (s) 13.8893 +time/epoch (s) 20.2753 +time/total (s) 2599.28 +Epoch -884 +------------------------------ ---------------- +2022-05-15 18:46:05.372703 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -883 finished +------------------------------ ---------------- +epoch -883 +replay_buffer/size 999047 +trainer/num train calls 118000 +trainer/QF1 Loss 10.9008 +trainer/QF2 Loss 11.0464 +trainer/Policy Loss 8.27517 +trainer/Q1 Predictions Mean -71.3928 +trainer/Q1 Predictions Std 17.3235 +trainer/Q1 Predictions Max -1.0456 +trainer/Q1 Predictions Min -86.6615 +trainer/Q2 Predictions Mean -71.4368 +trainer/Q2 Predictions Std 17.2894 +trainer/Q2 Predictions Max -0.967698 +trainer/Q2 Predictions Min -86.8223 +trainer/Q Targets Mean -71.4147 +trainer/Q Targets Std 17.4146 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3448 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0164981 +trainer/policy/mean Std 0.683456 +trainer/policy/mean Max 0.998464 +trainer/policy/mean Min -0.994376 +trainer/policy/std Mean 0.445413 +trainer/policy/std Std 0.0226116 +trainer/policy/std Max 0.471622 +trainer/policy/std Min 0.406623 +trainer/Advantage Weights Mean 2.21116 +trainer/Advantage Weights Std 10.6388 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.83836e-13 +trainer/Advantage Score Mean -0.370017 +trainer/Advantage Score Std 0.465128 +trainer/Advantage Score Max 0.802351 +trainer/Advantage Score Min -2.88904 +trainer/V1 Predictions Mean -71.3451 +trainer/V1 Predictions Std 17.3688 +trainer/V1 Predictions Max -1.27037 +trainer/V1 Predictions Min -86.2871 +trainer/VF Loss 0.0422144 +expl/num steps total 118000 +expl/num paths total 120 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0190668 +expl/Actions Std 0.799137 +expl/Actions Max 2.71723 +expl/Actions Min -2.47517 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 116242 +eval/num paths total 118 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.305102 +eval/Actions Std 0.701243 +eval/Actions Max 0.995269 +eval/Actions Min -0.998185 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.87825e-06 +time/evaluation sampling (s) 2.95122 +time/exploration sampling (s) 3.27709 +time/logging (s) 0.00741608 +time/saving (s) 0.0125031 +time/training (s) 13.9907 +time/epoch (s) 20.2389 +time/total (s) 2619.52 +Epoch -883 +------------------------------ ---------------- +2022-05-15 18:46:26.473448 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -882 finished +------------------------------ ---------------- +epoch -882 +replay_buffer/size 999047 +trainer/num train calls 119000 +trainer/QF1 Loss 1.43961 +trainer/QF2 Loss 1.41988 +trainer/Policy Loss 38.4688 +trainer/Q1 Predictions Mean -71.7501 +trainer/Q1 Predictions Std 17.1087 +trainer/Q1 Predictions Max -0.260289 +trainer/Q1 Predictions Min -86.8409 +trainer/Q2 Predictions Mean -71.7703 +trainer/Q2 Predictions Std 17.1073 +trainer/Q2 Predictions Max -0.263926 +trainer/Q2 Predictions Min -87.0096 +trainer/Q Targets Mean -72.4412 +trainer/Q Targets Std 16.7721 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.5445 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0131214 +trainer/policy/mean Std 0.687266 +trainer/policy/mean Max 0.999563 +trainer/policy/mean Min -0.998673 +trainer/policy/std Mean 0.4462 +trainer/policy/std Std 0.0226816 +trainer/policy/std Max 0.469668 +trainer/policy/std Min 0.406149 +trainer/Advantage Weights Mean 6.50441 +trainer/Advantage Weights Std 20.6001 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.08576e-08 +trainer/Advantage Score Mean -0.181254 +trainer/Advantage Score Std 0.432632 +trainer/Advantage Score Max 1.53084 +trainer/Advantage Score Min -1.76855 +trainer/V1 Predictions Mean -72.0616 +trainer/V1 Predictions Std 17.0888 +trainer/V1 Predictions Max 0.201525 +trainer/V1 Predictions Min -87.8471 +trainer/VF Loss 0.0506145 +expl/num steps total 119000 +expl/num paths total 121 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0330887 +expl/Actions Std 0.837649 +expl/Actions Max 2.56402 +expl/Actions Min -2.37485 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 117242 +eval/num paths total 119 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0403531 +eval/Actions Std 0.726957 +eval/Actions Max 0.999252 +eval/Actions Min -0.997578 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.84705e-06 +time/evaluation sampling (s) 3.28409 +time/exploration sampling (s) 3.3075 +time/logging (s) 0.00812549 +time/saving (s) 0.0107148 +time/training (s) 14.4842 +time/epoch (s) 21.0946 +time/total (s) 2640.62 +Epoch -882 +------------------------------ ---------------- +2022-05-15 18:46:47.569528 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -881 finished +------------------------------ ---------------- +epoch -881 +replay_buffer/size 999047 +trainer/num train calls 120000 +trainer/QF1 Loss 0.544903 +trainer/QF2 Loss 0.618363 +trainer/Policy Loss 28.3216 +trainer/Q1 Predictions Mean -70.6499 +trainer/Q1 Predictions Std 17.7913 +trainer/Q1 Predictions Max -0.900426 +trainer/Q1 Predictions Min -88.3499 +trainer/Q2 Predictions Mean -70.5468 +trainer/Q2 Predictions Std 17.8295 +trainer/Q2 Predictions Max -0.351145 +trainer/Q2 Predictions Min -88.0536 +trainer/Q Targets Mean -70.818 +trainer/Q Targets Std 18.0884 +trainer/Q Targets Max -1.46662 +trainer/Q Targets Min -87.8274 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00244166 +trainer/policy/mean Std 0.693437 +trainer/policy/mean Max 0.996195 +trainer/policy/mean Min -0.998975 +trainer/policy/std Mean 0.444263 +trainer/policy/std Std 0.022292 +trainer/policy/std Max 0.471865 +trainer/policy/std Min 0.407108 +trainer/Advantage Weights Mean 6.00228 +trainer/Advantage Weights Std 18.399 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.74853e-11 +trainer/Advantage Score Mean -0.186785 +trainer/Advantage Score Std 0.44144 +trainer/Advantage Score Max 0.963596 +trainer/Advantage Score Min -2.40071 +trainer/V1 Predictions Mean -70.5136 +trainer/V1 Predictions Std 18.2528 +trainer/V1 Predictions Max 0.0303064 +trainer/V1 Predictions Min -87.6728 +trainer/VF Loss 0.0422442 +expl/num steps total 120000 +expl/num paths total 122 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0117938 +expl/Actions Std 0.862919 +expl/Actions Max 2.87956 +expl/Actions Min -2.42972 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 118242 +eval/num paths total 120 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.081336 +eval/Actions Std 0.640068 +eval/Actions Max 0.999494 +eval/Actions Min -0.997043 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.7311e-06 +time/evaluation sampling (s) 3.10362 +time/exploration sampling (s) 3.18072 +time/logging (s) 0.00877136 +time/saving (s) 0.0122733 +time/training (s) 14.7857 +time/epoch (s) 21.091 +time/total (s) 2661.72 +Epoch -881 +------------------------------ ---------------- +2022-05-15 18:47:07.975090 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -880 finished +------------------------------ ---------------- +epoch -880 +replay_buffer/size 999047 +trainer/num train calls 121000 +trainer/QF1 Loss 0.905507 +trainer/QF2 Loss 0.968855 +trainer/Policy Loss 9.24423 +trainer/Q1 Predictions Mean -69.5638 +trainer/Q1 Predictions Std 19.4636 +trainer/Q1 Predictions Max -0.499942 +trainer/Q1 Predictions Min -86.5261 +trainer/Q2 Predictions Mean -69.6038 +trainer/Q2 Predictions Std 19.4957 +trainer/Q2 Predictions Max -0.390068 +trainer/Q2 Predictions Min -86.543 +trainer/Q Targets Mean -69.6044 +trainer/Q Targets Std 19.7448 +trainer/Q Targets Max 0.283167 +trainer/Q Targets Min -86.7594 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000651401 +trainer/policy/mean Std 0.688026 +trainer/policy/mean Max 0.99872 +trainer/policy/mean Min -0.996894 +trainer/policy/std Mean 0.444449 +trainer/policy/std Std 0.0232104 +trainer/policy/std Max 0.473742 +trainer/policy/std Min 0.41071 +trainer/Advantage Weights Mean 1.77661 +trainer/Advantage Weights Std 9.84208 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.92348e-14 +trainer/Advantage Score Mean -0.467607 +trainer/Advantage Score Std 0.536539 +trainer/Advantage Score Max 0.610611 +trainer/Advantage Score Min -3.06422 +trainer/V1 Predictions Mean -69.3256 +trainer/V1 Predictions Std 19.8472 +trainer/V1 Predictions Max 0.836771 +trainer/V1 Predictions Min -86.4298 +trainer/VF Loss 0.0557867 +expl/num steps total 121000 +expl/num paths total 123 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.037982 +expl/Actions Std 0.818311 +expl/Actions Max 2.64374 +expl/Actions Min -2.28451 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 119242 +eval/num paths total 121 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0217287 +eval/Actions Std 0.707201 +eval/Actions Max 0.999341 +eval/Actions Min -0.99698 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.71482e-06 +time/evaluation sampling (s) 2.83494 +time/exploration sampling (s) 3.14704 +time/logging (s) 0.0076724 +time/saving (s) 0.011459 +time/training (s) 14.3971 +time/epoch (s) 20.3982 +time/total (s) 2682.12 +Epoch -880 +------------------------------ ---------------- +2022-05-15 18:47:28.585108 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -879 finished +------------------------------ ---------------- +epoch -879 +replay_buffer/size 999047 +trainer/num train calls 122000 +trainer/QF1 Loss 0.551876 +trainer/QF2 Loss 0.662537 +trainer/Policy Loss 19.9022 +trainer/Q1 Predictions Mean -70.9917 +trainer/Q1 Predictions Std 18.6733 +trainer/Q1 Predictions Max -1.63427 +trainer/Q1 Predictions Min -86.5031 +trainer/Q2 Predictions Mean -70.9494 +trainer/Q2 Predictions Std 18.6674 +trainer/Q2 Predictions Max -1.77368 +trainer/Q2 Predictions Min -85.9183 +trainer/Q Targets Mean -70.8909 +trainer/Q Targets Std 18.8835 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5537 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0256566 +trainer/policy/mean Std 0.695205 +trainer/policy/mean Max 0.999061 +trainer/policy/mean Min -0.997103 +trainer/policy/std Mean 0.44391 +trainer/policy/std Std 0.0217357 +trainer/policy/std Max 0.470486 +trainer/policy/std Min 0.411072 +trainer/Advantage Weights Mean 4.92554 +trainer/Advantage Weights Std 18.5029 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.90917e-11 +trainer/Advantage Score Mean -0.290101 +trainer/Advantage Score Std 0.46328 +trainer/Advantage Score Max 1.88342 +trainer/Advantage Score Min -2.33956 +trainer/V1 Predictions Mean -70.6587 +trainer/V1 Predictions Std 18.9646 +trainer/V1 Predictions Max -1.14566 +trainer/V1 Predictions Min -86.4826 +trainer/VF Loss 0.0572029 +expl/num steps total 122000 +expl/num paths total 124 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0463694 +expl/Actions Std 0.833384 +expl/Actions Max 2.54144 +expl/Actions Min -2.38134 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 120242 +eval/num paths total 122 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0177787 +eval/Actions Std 0.689294 +eval/Actions Max 0.998483 +eval/Actions Min -0.996707 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.01097e-06 +time/evaluation sampling (s) 2.72999 +time/exploration sampling (s) 3.2181 +time/logging (s) 0.0103719 +time/saving (s) 0.0127413 +time/training (s) 14.6351 +time/epoch (s) 20.6064 +time/total (s) 2702.73 +Epoch -879 +------------------------------ ---------------- +2022-05-15 18:47:49.494950 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -878 finished +------------------------------ ---------------- +epoch -878 +replay_buffer/size 999047 +trainer/num train calls 123000 +trainer/QF1 Loss 0.448825 +trainer/QF2 Loss 0.479544 +trainer/Policy Loss 17.8199 +trainer/Q1 Predictions Mean -71.3846 +trainer/Q1 Predictions Std 18.9115 +trainer/Q1 Predictions Max -0.563478 +trainer/Q1 Predictions Min -87.2118 +trainer/Q2 Predictions Mean -71.3605 +trainer/Q2 Predictions Std 18.9027 +trainer/Q2 Predictions Max -0.819683 +trainer/Q2 Predictions Min -86.8618 +trainer/Q Targets Mean -71.4189 +trainer/Q Targets Std 18.8633 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.227 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0106718 +trainer/policy/mean Std 0.695511 +trainer/policy/mean Max 0.999376 +trainer/policy/mean Min -0.996751 +trainer/policy/std Mean 0.443351 +trainer/policy/std Std 0.0210961 +trainer/policy/std Max 0.468785 +trainer/policy/std Min 0.410868 +trainer/Advantage Weights Mean 4.62685 +trainer/Advantage Weights Std 17.8244 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.07515e-15 +trainer/Advantage Score Mean -0.341349 +trainer/Advantage Score Std 0.480696 +trainer/Advantage Score Max 0.944157 +trainer/Advantage Score Min -3.25822 +trainer/V1 Predictions Mean -71.1841 +trainer/V1 Predictions Std 18.9309 +trainer/V1 Predictions Max -1.12034 +trainer/V1 Predictions Min -87.2147 +trainer/VF Loss 0.0498676 +expl/num steps total 123000 +expl/num paths total 125 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0312795 +expl/Actions Std 0.833364 +expl/Actions Max 2.45404 +expl/Actions Min -2.3515 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 121242 +eval/num paths total 123 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.129586 +eval/Actions Std 0.662002 +eval/Actions Max 0.999337 +eval/Actions Min -0.997662 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.71108e-06 +time/evaluation sampling (s) 2.80581 +time/exploration sampling (s) 3.68608 +time/logging (s) 0.0111566 +time/saving (s) 0.0151115 +time/training (s) 14.3842 +time/epoch (s) 20.9024 +time/total (s) 2723.64 +Epoch -878 +------------------------------ ---------------- +2022-05-15 18:48:10.991775 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -877 finished +------------------------------ ---------------- +epoch -877 +replay_buffer/size 999047 +trainer/num train calls 124000 +trainer/QF1 Loss 0.630567 +trainer/QF2 Loss 0.769727 +trainer/Policy Loss 7.14597 +trainer/Q1 Predictions Mean -71.5412 +trainer/Q1 Predictions Std 18.7962 +trainer/Q1 Predictions Max -0.656638 +trainer/Q1 Predictions Min -87.4385 +trainer/Q2 Predictions Mean -71.5307 +trainer/Q2 Predictions Std 18.8277 +trainer/Q2 Predictions Max -0.458787 +trainer/Q2 Predictions Min -87.1482 +trainer/Q Targets Mean -71.3598 +trainer/Q Targets Std 18.8076 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7732 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0193451 +trainer/policy/mean Std 0.701139 +trainer/policy/mean Max 0.998567 +trainer/policy/mean Min -0.99368 +trainer/policy/std Mean 0.444779 +trainer/policy/std Std 0.0223944 +trainer/policy/std Max 0.470945 +trainer/policy/std Min 0.406739 +trainer/Advantage Weights Mean 2.07294 +trainer/Advantage Weights Std 11.2922 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.55248e-12 +trainer/Advantage Score Mean -0.495455 +trainer/Advantage Score Std 0.447204 +trainer/Advantage Score Max 0.931625 +trainer/Advantage Score Min -2.54848 +trainer/V1 Predictions Mean -71.1178 +trainer/V1 Predictions Std 18.9232 +trainer/V1 Predictions Max 0.608956 +trainer/V1 Predictions Min -86.5719 +trainer/VF Loss 0.0513373 +expl/num steps total 124000 +expl/num paths total 126 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0278989 +expl/Actions Std 0.817727 +expl/Actions Max 2.59091 +expl/Actions Min -2.39173 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 122242 +eval/num paths total 124 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.127391 +eval/Actions Std 0.754831 +eval/Actions Max 0.998903 +eval/Actions Min -0.997797 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.11508e-06 +time/evaluation sampling (s) 3.06182 +time/exploration sampling (s) 3.62211 +time/logging (s) 0.00734368 +time/saving (s) 0.0148006 +time/training (s) 14.7797 +time/epoch (s) 21.4858 +time/total (s) 2745.13 +Epoch -877 +------------------------------ ---------------- +2022-05-15 18:48:31.879667 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -876 finished +------------------------------ ---------------- +epoch -876 +replay_buffer/size 999047 +trainer/num train calls 125000 +trainer/QF1 Loss 1.51469 +trainer/QF2 Loss 1.52313 +trainer/Policy Loss 21.5 +trainer/Q1 Predictions Mean -70.2174 +trainer/Q1 Predictions Std 20.6088 +trainer/Q1 Predictions Max -0.539696 +trainer/Q1 Predictions Min -87.7171 +trainer/Q2 Predictions Mean -70.1403 +trainer/Q2 Predictions Std 20.5794 +trainer/Q2 Predictions Max -0.341197 +trainer/Q2 Predictions Min -87.7283 +trainer/Q Targets Mean -69.9376 +trainer/Q Targets Std 20.661 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6377 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0108168 +trainer/policy/mean Std 0.704692 +trainer/policy/mean Max 0.998685 +trainer/policy/mean Min -0.998107 +trainer/policy/std Mean 0.441895 +trainer/policy/std Std 0.0223118 +trainer/policy/std Max 0.470413 +trainer/policy/std Min 0.405491 +trainer/Advantage Weights Mean 4.11733 +trainer/Advantage Weights Std 17.5418 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.34558e-13 +trainer/Advantage Score Mean -0.361429 +trainer/Advantage Score Std 0.51991 +trainer/Advantage Score Max 1.05177 +trainer/Advantage Score Min -2.8726 +trainer/V1 Predictions Mean -69.7312 +trainer/V1 Predictions Std 20.569 +trainer/V1 Predictions Max -0.054551 +trainer/V1 Predictions Min -87.6052 +trainer/VF Loss 0.0584745 +expl/num steps total 125000 +expl/num paths total 127 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0460934 +expl/Actions Std 0.799079 +expl/Actions Max 2.44076 +expl/Actions Min -2.3802 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 123242 +eval/num paths total 125 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.120495 +eval/Actions Std 0.719519 +eval/Actions Max 0.998642 +eval/Actions Min -0.996228 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.70782e-06 +time/evaluation sampling (s) 3.27034 +time/exploration sampling (s) 3.60897 +time/logging (s) 0.00811997 +time/saving (s) 0.013413 +time/training (s) 13.9824 +time/epoch (s) 20.8832 +time/total (s) 2766.02 +Epoch -876 +------------------------------ ---------------- +2022-05-15 18:48:52.551349 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -875 finished +------------------------------ ---------------- +epoch -875 +replay_buffer/size 999047 +trainer/num train calls 126000 +trainer/QF1 Loss 0.428947 +trainer/QF2 Loss 0.395139 +trainer/Policy Loss 9.06984 +trainer/Q1 Predictions Mean -72.3323 +trainer/Q1 Predictions Std 16.8428 +trainer/Q1 Predictions Max -1.3691 +trainer/Q1 Predictions Min -87.334 +trainer/Q2 Predictions Mean -72.3743 +trainer/Q2 Predictions Std 16.7995 +trainer/Q2 Predictions Max -1.11429 +trainer/Q2 Predictions Min -87.429 +trainer/Q Targets Mean -72.4696 +trainer/Q Targets Std 16.8683 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4354 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0106686 +trainer/policy/mean Std 0.692194 +trainer/policy/mean Max 0.996698 +trainer/policy/mean Min -0.996594 +trainer/policy/std Mean 0.442252 +trainer/policy/std Std 0.022472 +trainer/policy/std Max 0.468772 +trainer/policy/std Min 0.405828 +trainer/Advantage Weights Mean 1.323 +trainer/Advantage Weights Std 8.93074 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.59819e-14 +trainer/Advantage Score Mean -0.431111 +trainer/Advantage Score Std 0.526237 +trainer/Advantage Score Max 0.767084 +trainer/Advantage Score Min -3.12814 +trainer/V1 Predictions Mean -72.1785 +trainer/V1 Predictions Std 17.0519 +trainer/V1 Predictions Max -0.603712 +trainer/V1 Predictions Min -87.3617 +trainer/VF Loss 0.0502379 +expl/num steps total 126000 +expl/num paths total 128 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0249383 +expl/Actions Std 0.808316 +expl/Actions Max 2.29916 +expl/Actions Min -2.40942 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 124242 +eval/num paths total 126 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.155782 +eval/Actions Std 0.721912 +eval/Actions Max 0.997247 +eval/Actions Min -0.995846 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.06522e-06 +time/evaluation sampling (s) 3.01008 +time/exploration sampling (s) 3.50738 +time/logging (s) 0.00787451 +time/saving (s) 0.0141778 +time/training (s) 14.1236 +time/epoch (s) 20.6631 +time/total (s) 2786.69 +Epoch -875 +------------------------------ ---------------- +2022-05-15 18:49:13.910244 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -874 finished +------------------------------ ---------------- +epoch -874 +replay_buffer/size 999047 +trainer/num train calls 127000 +trainer/QF1 Loss 0.576917 +trainer/QF2 Loss 0.551816 +trainer/Policy Loss 28.0216 +trainer/Q1 Predictions Mean -70.3568 +trainer/Q1 Predictions Std 19.3357 +trainer/Q1 Predictions Max -0.944228 +trainer/Q1 Predictions Min -86.372 +trainer/Q2 Predictions Mean -70.3993 +trainer/Q2 Predictions Std 19.403 +trainer/Q2 Predictions Max -0.675496 +trainer/Q2 Predictions Min -86.489 +trainer/Q Targets Mean -70.5219 +trainer/Q Targets Std 19.5604 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4325 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00398724 +trainer/policy/mean Std 0.708399 +trainer/policy/mean Max 0.998615 +trainer/policy/mean Min -0.99967 +trainer/policy/std Mean 0.440996 +trainer/policy/std Std 0.022388 +trainer/policy/std Max 0.465106 +trainer/policy/std Min 0.405361 +trainer/Advantage Weights Mean 4.52663 +trainer/Advantage Weights Std 19.3932 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.90713e-13 +trainer/Advantage Score Mean -0.303717 +trainer/Advantage Score Std 0.422082 +trainer/Advantage Score Max 1.65635 +trainer/Advantage Score Min -2.9288 +trainer/V1 Predictions Mean -70.3584 +trainer/V1 Predictions Std 19.44 +trainer/V1 Predictions Max -0.808281 +trainer/V1 Predictions Min -86.3326 +trainer/VF Loss 0.0500113 +expl/num steps total 127000 +expl/num paths total 129 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0469111 +expl/Actions Std 0.797078 +expl/Actions Max 2.15847 +expl/Actions Min -2.38129 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 125242 +eval/num paths total 127 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.293824 +eval/Actions Std 0.641288 +eval/Actions Max 0.999247 +eval/Actions Min -0.998353 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.9684e-06 +time/evaluation sampling (s) 3.13606 +time/exploration sampling (s) 3.61623 +time/logging (s) 0.0110237 +time/saving (s) 0.0155721 +time/training (s) 14.5752 +time/epoch (s) 21.3541 +time/total (s) 2808.05 +Epoch -874 +------------------------------ ---------------- +2022-05-15 18:49:34.260797 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -873 finished +------------------------------ ---------------- +epoch -873 +replay_buffer/size 999047 +trainer/num train calls 128000 +trainer/QF1 Loss 1.28757 +trainer/QF2 Loss 1.30462 +trainer/Policy Loss 32.1053 +trainer/Q1 Predictions Mean -70.7351 +trainer/Q1 Predictions Std 18.2073 +trainer/Q1 Predictions Max -2.78922 +trainer/Q1 Predictions Min -86.9409 +trainer/Q2 Predictions Mean -70.7085 +trainer/Q2 Predictions Std 18.2587 +trainer/Q2 Predictions Max -2.78724 +trainer/Q2 Predictions Min -86.797 +trainer/Q Targets Mean -71.2656 +trainer/Q Targets Std 18.2077 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9233 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00706633 +trainer/policy/mean Std 0.718094 +trainer/policy/mean Max 0.996818 +trainer/policy/mean Min -0.997323 +trainer/policy/std Mean 0.442982 +trainer/policy/std Std 0.0238042 +trainer/policy/std Max 0.471221 +trainer/policy/std Min 0.405619 +trainer/Advantage Weights Mean 5.95252 +trainer/Advantage Weights Std 19.8716 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.33684e-09 +trainer/Advantage Score Mean -0.236105 +trainer/Advantage Score Std 0.456203 +trainer/Advantage Score Max 1.95696 +trainer/Advantage Score Min -2.0433 +trainer/V1 Predictions Mean -71.0906 +trainer/V1 Predictions Std 18.3351 +trainer/V1 Predictions Max -4.05555 +trainer/V1 Predictions Min -87.2137 +trainer/VF Loss 0.0611882 +expl/num steps total 128000 +expl/num paths total 130 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0187176 +expl/Actions Std 0.815326 +expl/Actions Max 2.7678 +expl/Actions Min -2.41342 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 126242 +eval/num paths total 128 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.079375 +eval/Actions Std 0.719464 +eval/Actions Max 0.998698 +eval/Actions Min -0.997866 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.81399e-06 +time/evaluation sampling (s) 3.03465 +time/exploration sampling (s) 3.41231 +time/logging (s) 0.00784436 +time/saving (s) 0.00926163 +time/training (s) 13.8754 +time/epoch (s) 20.3395 +time/total (s) 2828.39 +Epoch -873 +------------------------------ ---------------- +2022-05-15 18:49:55.364323 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -872 finished +------------------------------ ---------------- +epoch -872 +replay_buffer/size 999047 +trainer/num train calls 129000 +trainer/QF1 Loss 0.479953 +trainer/QF2 Loss 0.513539 +trainer/Policy Loss 1.66117 +trainer/Q1 Predictions Mean -68.6583 +trainer/Q1 Predictions Std 21.3016 +trainer/Q1 Predictions Max -0.495635 +trainer/Q1 Predictions Min -85.9137 +trainer/Q2 Predictions Mean -68.6868 +trainer/Q2 Predictions Std 21.2347 +trainer/Q2 Predictions Max -0.842816 +trainer/Q2 Predictions Min -85.9914 +trainer/Q Targets Mean -68.692 +trainer/Q Targets Std 21.3797 +trainer/Q Targets Max 1.15418 +trainer/Q Targets Min -86.0676 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00800692 +trainer/policy/mean Std 0.696189 +trainer/policy/mean Max 0.997563 +trainer/policy/mean Min -0.999246 +trainer/policy/std Mean 0.441041 +trainer/policy/std Std 0.0219143 +trainer/policy/std Max 0.467202 +trainer/policy/std Min 0.405897 +trainer/Advantage Weights Mean 0.43319 +trainer/Advantage Weights Std 3.111 +trainer/Advantage Weights Max 41.7878 +trainer/Advantage Weights Min 1.91227e-13 +trainer/Advantage Score Mean -0.76523 +trainer/Advantage Score Std 0.54755 +trainer/Advantage Score Max 0.37326 +trainer/Advantage Score Min -2.92853 +trainer/V1 Predictions Mean -68.3528 +trainer/V1 Predictions Std 21.5335 +trainer/V1 Predictions Max 1.30578 +trainer/V1 Predictions Min -86.1564 +trainer/VF Loss 0.0898504 +expl/num steps total 129000 +expl/num paths total 131 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00973055 +expl/Actions Std 0.806408 +expl/Actions Max 2.42217 +expl/Actions Min -2.28 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 127101 +eval/num paths total 129 +eval/path length Mean 859 +eval/path length Std 0 +eval/path length Max 859 +eval/path length Min 859 +eval/Rewards Mean 0.00116414 +eval/Rewards Std 0.0340997 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0110298 +eval/Actions Std 0.705135 +eval/Actions Max 0.9992 +eval/Actions Min -0.999559 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.05288e-06 +time/evaluation sampling (s) 3.09775 +time/exploration sampling (s) 3.0646 +time/logging (s) 0.00662319 +time/saving (s) 0.0102743 +time/training (s) 14.9177 +time/epoch (s) 21.097 +time/total (s) 2849.49 +Epoch -872 +------------------------------ ---------------- +2022-05-15 18:50:15.375662 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -871 finished +------------------------------ ---------------- +epoch -871 +replay_buffer/size 999047 +trainer/num train calls 130000 +trainer/QF1 Loss 0.474575 +trainer/QF2 Loss 0.515411 +trainer/Policy Loss 24.6911 +trainer/Q1 Predictions Mean -70.709 +trainer/Q1 Predictions Std 17.3981 +trainer/Q1 Predictions Max -1.30911 +trainer/Q1 Predictions Min -87.4789 +trainer/Q2 Predictions Mean -70.6875 +trainer/Q2 Predictions Std 17.3739 +trainer/Q2 Predictions Max -1.25462 +trainer/Q2 Predictions Min -87.2903 +trainer/Q Targets Mean -70.6023 +trainer/Q Targets Std 17.3406 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0456 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00437777 +trainer/policy/mean Std 0.69803 +trainer/policy/mean Max 0.997681 +trainer/policy/mean Min -0.998877 +trainer/policy/std Mean 0.442013 +trainer/policy/std Std 0.0216494 +trainer/policy/std Max 0.466698 +trainer/policy/std Min 0.406581 +trainer/Advantage Weights Mean 3.41206 +trainer/Advantage Weights Std 15.5829 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.25089e-18 +trainer/Advantage Score Mean -0.389308 +trainer/Advantage Score Std 0.539991 +trainer/Advantage Score Max 0.957996 +trainer/Advantage Score Min -3.96138 +trainer/V1 Predictions Mean -70.2975 +trainer/V1 Predictions Std 17.4998 +trainer/V1 Predictions Max -0.298354 +trainer/V1 Predictions Min -86.8247 +trainer/VF Loss 0.0574213 +expl/num steps total 130000 +expl/num paths total 132 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0404413 +expl/Actions Std 0.819449 +expl/Actions Max 2.38512 +expl/Actions Min -2.45897 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 127874 +eval/num paths total 130 +eval/path length Mean 773 +eval/path length Std 0 +eval/path length Max 773 +eval/path length Min 773 +eval/Rewards Mean 0.00129366 +eval/Rewards Std 0.0359442 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00511026 +eval/Actions Std 0.718445 +eval/Actions Max 0.9992 +eval/Actions Min -0.998747 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.48523e-06 +time/evaluation sampling (s) 2.78481 +time/exploration sampling (s) 2.99421 +time/logging (s) 0.0120955 +time/saving (s) 0.019064 +time/training (s) 14.1998 +time/epoch (s) 20.0099 +time/total (s) 2869.51 +Epoch -871 +------------------------------ ---------------- +2022-05-15 18:50:35.739227 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -870 finished +------------------------------ ---------------- +epoch -870 +replay_buffer/size 999047 +trainer/num train calls 131000 +trainer/QF1 Loss 0.647849 +trainer/QF2 Loss 0.72454 +trainer/Policy Loss 4.94867 +trainer/Q1 Predictions Mean -70.9807 +trainer/Q1 Predictions Std 18.3016 +trainer/Q1 Predictions Max -1.7706 +trainer/Q1 Predictions Min -86.4242 +trainer/Q2 Predictions Mean -71.0136 +trainer/Q2 Predictions Std 18.3329 +trainer/Q2 Predictions Max -1.33995 +trainer/Q2 Predictions Min -86.5469 +trainer/Q Targets Mean -70.9476 +trainer/Q Targets Std 18.5987 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.566 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0255681 +trainer/policy/mean Std 0.700353 +trainer/policy/mean Max 0.998237 +trainer/policy/mean Min -0.997551 +trainer/policy/std Mean 0.442917 +trainer/policy/std Std 0.0221623 +trainer/policy/std Max 0.467037 +trainer/policy/std Min 0.408176 +trainer/Advantage Weights Mean 0.661118 +trainer/Advantage Weights Std 6.36787 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.90236e-14 +trainer/Advantage Score Mean -0.581412 +trainer/Advantage Score Std 0.471331 +trainer/Advantage Score Max 0.510315 +trainer/Advantage Score Min -3.15931 +trainer/V1 Predictions Mean -70.6815 +trainer/V1 Predictions Std 18.5746 +trainer/V1 Predictions Max -0.954731 +trainer/V1 Predictions Min -86.4312 +trainer/VF Loss 0.057668 +expl/num steps total 131000 +expl/num paths total 133 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.310362 +expl/Actions Std 0.84986 +expl/Actions Max 2.42916 +expl/Actions Min -2.49621 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 128874 +eval/num paths total 131 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.187852 +eval/Actions Std 0.649202 +eval/Actions Max 0.997571 +eval/Actions Min -0.996179 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.746e-06 +time/evaluation sampling (s) 2.63477 +time/exploration sampling (s) 3.54986 +time/logging (s) 0.00764102 +time/saving (s) 0.0100472 +time/training (s) 14.1474 +time/epoch (s) 20.3498 +time/total (s) 2889.86 +Epoch -870 +------------------------------ ---------------- +2022-05-15 18:50:56.516196 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -869 finished +------------------------------ ---------------- +epoch -869 +replay_buffer/size 999047 +trainer/num train calls 132000 +trainer/QF1 Loss 0.810182 +trainer/QF2 Loss 0.754224 +trainer/Policy Loss 6.9313 +trainer/Q1 Predictions Mean -72.2394 +trainer/Q1 Predictions Std 16.9303 +trainer/Q1 Predictions Max -0.471865 +trainer/Q1 Predictions Min -86.927 +trainer/Q2 Predictions Mean -72.2321 +trainer/Q2 Predictions Std 16.9351 +trainer/Q2 Predictions Max -0.489639 +trainer/Q2 Predictions Min -86.9905 +trainer/Q Targets Mean -71.8304 +trainer/Q Targets Std 17.0231 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7082 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0112832 +trainer/policy/mean Std 0.702024 +trainer/policy/mean Max 0.997876 +trainer/policy/mean Min -0.995959 +trainer/policy/std Mean 0.443451 +trainer/policy/std Std 0.0230617 +trainer/policy/std Max 0.469752 +trainer/policy/std Min 0.408331 +trainer/Advantage Weights Mean 1.34832 +trainer/Advantage Weights Std 8.64253 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.71062e-13 +trainer/Advantage Score Mean -0.517342 +trainer/Advantage Score Std 0.478129 +trainer/Advantage Score Max 0.803625 +trainer/Advantage Score Min -2.80299 +trainer/V1 Predictions Mean -71.5745 +trainer/V1 Predictions Std 17.109 +trainer/V1 Predictions Max 0.653872 +trainer/V1 Predictions Min -86.6271 +trainer/VF Loss 0.0542515 +expl/num steps total 132000 +expl/num paths total 134 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0327866 +expl/Actions Std 0.801078 +expl/Actions Max 2.34772 +expl/Actions Min -2.20923 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 129651 +eval/num paths total 132 +eval/path length Mean 777 +eval/path length Std 0 +eval/path length Max 777 +eval/path length Min 777 +eval/Rewards Mean 0.001287 +eval/Rewards Std 0.0358517 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0558947 +eval/Actions Std 0.729299 +eval/Actions Max 0.998483 +eval/Actions Min -0.998594 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.69618e-06 +time/evaluation sampling (s) 2.62717 +time/exploration sampling (s) 3.74256 +time/logging (s) 0.00644025 +time/saving (s) 0.0131226 +time/training (s) 14.381 +time/epoch (s) 20.7703 +time/total (s) 2910.64 +Epoch -869 +------------------------------ ---------------- +2022-05-15 18:51:17.782518 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -868 finished +------------------------------ ---------------- +epoch -868 +replay_buffer/size 999047 +trainer/num train calls 133000 +trainer/QF1 Loss 0.628369 +trainer/QF2 Loss 0.526557 +trainer/Policy Loss 51.246 +trainer/Q1 Predictions Mean -69.9831 +trainer/Q1 Predictions Std 18.6536 +trainer/Q1 Predictions Max -3.90539 +trainer/Q1 Predictions Min -87.6616 +trainer/Q2 Predictions Mean -70.0028 +trainer/Q2 Predictions Std 18.695 +trainer/Q2 Predictions Max -4.33087 +trainer/Q2 Predictions Min -87.8064 +trainer/Q Targets Mean -70.1171 +trainer/Q Targets Std 18.8904 +trainer/Q Targets Max -2.62097 +trainer/Q Targets Min -88.2967 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000371397 +trainer/policy/mean Std 0.694799 +trainer/policy/mean Max 0.996499 +trainer/policy/mean Min -0.996846 +trainer/policy/std Mean 0.442117 +trainer/policy/std Std 0.0236393 +trainer/policy/std Max 0.467552 +trainer/policy/std Min 0.405132 +trainer/Advantage Weights Mean 9.18005 +trainer/Advantage Weights Std 23.4189 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.56682e-13 +trainer/Advantage Score Mean -0.202779 +trainer/Advantage Score Std 0.568658 +trainer/Advantage Score Max 0.837135 +trainer/Advantage Score Min -2.94846 +trainer/V1 Predictions Mean -69.8936 +trainer/V1 Predictions Std 18.9677 +trainer/V1 Predictions Max -4.41322 +trainer/V1 Predictions Min -88.0028 +trainer/VF Loss 0.0642713 +expl/num steps total 133000 +expl/num paths total 135 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0319982 +expl/Actions Std 0.808499 +expl/Actions Max 2.77846 +expl/Actions Min -2.39079 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 130651 +eval/num paths total 133 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.155848 +eval/Actions Std 0.769051 +eval/Actions Max 0.999261 +eval/Actions Min -0.998093 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.83379e-06 +time/evaluation sampling (s) 2.91066 +time/exploration sampling (s) 3.77216 +time/logging (s) 0.00874381 +time/saving (s) 0.0110294 +time/training (s) 14.5591 +time/epoch (s) 21.2617 +time/total (s) 2931.9 +Epoch -868 +------------------------------ ---------------- +2022-05-15 18:51:38.633126 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -867 finished +------------------------------ ---------------- +epoch -867 +replay_buffer/size 999047 +trainer/num train calls 134000 +trainer/QF1 Loss 0.662081 +trainer/QF2 Loss 0.733792 +trainer/Policy Loss 42.7824 +trainer/Q1 Predictions Mean -70.4047 +trainer/Q1 Predictions Std 17.7888 +trainer/Q1 Predictions Max -0.545844 +trainer/Q1 Predictions Min -87.228 +trainer/Q2 Predictions Mean -70.3139 +trainer/Q2 Predictions Std 17.8298 +trainer/Q2 Predictions Max -0.579434 +trainer/Q2 Predictions Min -87.6211 +trainer/Q Targets Mean -70.6303 +trainer/Q Targets Std 17.7697 +trainer/Q Targets Max -1.79987 +trainer/Q Targets Min -87.6624 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00390065 +trainer/policy/mean Std 0.68742 +trainer/policy/mean Max 0.998925 +trainer/policy/mean Min -0.996983 +trainer/policy/std Mean 0.442032 +trainer/policy/std Std 0.0236004 +trainer/policy/std Max 0.468615 +trainer/policy/std Min 0.402916 +trainer/Advantage Weights Mean 7.53092 +trainer/Advantage Weights Std 23.0606 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.02287e-11 +trainer/Advantage Score Mean -0.282222 +trainer/Advantage Score Std 0.548271 +trainer/Advantage Score Max 1.33827 +trainer/Advantage Score Min -2.53058 +trainer/V1 Predictions Mean -70.304 +trainer/V1 Predictions Std 17.9929 +trainer/V1 Predictions Max 0.29289 +trainer/V1 Predictions Min -87.9847 +trainer/VF Loss 0.0718256 +expl/num steps total 134000 +expl/num paths total 136 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0507393 +expl/Actions Std 0.819389 +expl/Actions Max 2.32416 +expl/Actions Min -2.87347 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 131630 +eval/num paths total 134 +eval/path length Mean 979 +eval/path length Std 0 +eval/path length Max 979 +eval/path length Min 979 +eval/Rewards Mean 0.00102145 +eval/Rewards Std 0.0319438 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0304601 +eval/Actions Std 0.700169 +eval/Actions Max 0.999578 +eval/Actions Min -0.998795 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.78838e-06 +time/evaluation sampling (s) 2.7677 +time/exploration sampling (s) 3.57661 +time/logging (s) 0.0102922 +time/saving (s) 0.013967 +time/training (s) 14.4759 +time/epoch (s) 20.8445 +time/total (s) 2952.75 +Epoch -867 +------------------------------ ---------------- +2022-05-15 18:51:59.692116 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -866 finished +------------------------------ ---------------- +epoch -866 +replay_buffer/size 999047 +trainer/num train calls 135000 +trainer/QF1 Loss 0.548365 +trainer/QF2 Loss 0.631677 +trainer/Policy Loss 25.5643 +trainer/Q1 Predictions Mean -71.5705 +trainer/Q1 Predictions Std 17.7984 +trainer/Q1 Predictions Max -4.12777 +trainer/Q1 Predictions Min -86.7438 +trainer/Q2 Predictions Mean -71.5433 +trainer/Q2 Predictions Std 17.8714 +trainer/Q2 Predictions Max -3.90529 +trainer/Q2 Predictions Min -87.0292 +trainer/Q Targets Mean -71.8552 +trainer/Q Targets Std 17.6541 +trainer/Q Targets Max -4.48371 +trainer/Q Targets Min -87.3499 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0133543 +trainer/policy/mean Std 0.696591 +trainer/policy/mean Max 0.999441 +trainer/policy/mean Min -0.99799 +trainer/policy/std Mean 0.442929 +trainer/policy/std Std 0.0211399 +trainer/policy/std Max 0.468415 +trainer/policy/std Min 0.408944 +trainer/Advantage Weights Mean 5.63393 +trainer/Advantage Weights Std 18.5717 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.77194e-09 +trainer/Advantage Score Mean -0.150416 +trainer/Advantage Score Std 0.389053 +trainer/Advantage Score Max 1.27248 +trainer/Advantage Score Min -1.93957 +trainer/V1 Predictions Mean -71.5818 +trainer/V1 Predictions Std 17.8499 +trainer/V1 Predictions Max -2.85049 +trainer/V1 Predictions Min -86.7594 +trainer/VF Loss 0.0427893 +expl/num steps total 135000 +expl/num paths total 137 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0614456 +expl/Actions Std 0.83486 +expl/Actions Max 2.36592 +expl/Actions Min -2.44285 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 132630 +eval/num paths total 135 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.289879 +eval/Actions Std 0.642205 +eval/Actions Max 0.999394 +eval/Actions Min -0.996998 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.66498e-06 +time/evaluation sampling (s) 2.77181 +time/exploration sampling (s) 3.74143 +time/logging (s) 0.0104903 +time/saving (s) 0.0181642 +time/training (s) 14.5084 +time/epoch (s) 21.0503 +time/total (s) 2973.81 +Epoch -866 +------------------------------ ---------------- +2022-05-15 18:52:20.907095 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -865 finished +------------------------------ ---------------- +epoch -865 +replay_buffer/size 999047 +trainer/num train calls 136000 +trainer/QF1 Loss 0.699857 +trainer/QF2 Loss 0.762194 +trainer/Policy Loss 17.2532 +trainer/Q1 Predictions Mean -71.3658 +trainer/Q1 Predictions Std 18.0134 +trainer/Q1 Predictions Max -0.939552 +trainer/Q1 Predictions Min -86.5523 +trainer/Q2 Predictions Mean -71.3456 +trainer/Q2 Predictions Std 18.1355 +trainer/Q2 Predictions Max -0.83431 +trainer/Q2 Predictions Min -86.8218 +trainer/Q Targets Mean -71.3163 +trainer/Q Targets Std 17.9186 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.5784 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00635065 +trainer/policy/mean Std 0.696677 +trainer/policy/mean Max 0.999383 +trainer/policy/mean Min -0.997996 +trainer/policy/std Mean 0.441686 +trainer/policy/std Std 0.0205867 +trainer/policy/std Max 0.466119 +trainer/policy/std Min 0.407743 +trainer/Advantage Weights Mean 4.00546 +trainer/Advantage Weights Std 17.1737 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.41376e-11 +trainer/Advantage Score Mean -0.402166 +trainer/Advantage Score Std 0.489775 +trainer/Advantage Score Max 2.41502 +trainer/Advantage Score Min -2.44473 +trainer/V1 Predictions Mean -71.0813 +trainer/V1 Predictions Std 17.8187 +trainer/V1 Predictions Max -1.20666 +trainer/V1 Predictions Min -86.7903 +trainer/VF Loss 0.0825432 +expl/num steps total 136000 +expl/num paths total 138 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.105771 +expl/Actions Std 0.868578 +expl/Actions Max 2.33067 +expl/Actions Min -2.39658 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 133630 +eval/num paths total 136 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0326298 +eval/Actions Std 0.687384 +eval/Actions Max 0.999565 +eval/Actions Min -0.99903 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.74833e-06 +time/evaluation sampling (s) 2.93925 +time/exploration sampling (s) 3.81529 +time/logging (s) 0.00733649 +time/saving (s) 0.0113955 +time/training (s) 14.4276 +time/epoch (s) 21.2009 +time/total (s) 2995.02 +Epoch -865 +------------------------------ ---------------- +2022-05-15 18:52:42.018885 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -864 finished +------------------------------ ---------------- +epoch -864 +replay_buffer/size 999047 +trainer/num train calls 137000 +trainer/QF1 Loss 1.5758 +trainer/QF2 Loss 1.59685 +trainer/Policy Loss 39.6178 +trainer/Q1 Predictions Mean -72.2503 +trainer/Q1 Predictions Std 15.5296 +trainer/Q1 Predictions Max -1.77689 +trainer/Q1 Predictions Min -87.0362 +trainer/Q2 Predictions Mean -72.2749 +trainer/Q2 Predictions Std 15.4918 +trainer/Q2 Predictions Max -1.38892 +trainer/Q2 Predictions Min -87.0005 +trainer/Q Targets Mean -72.7662 +trainer/Q Targets Std 15.7308 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4125 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00744412 +trainer/policy/mean Std 0.692986 +trainer/policy/mean Max 0.998925 +trainer/policy/mean Min -0.99537 +trainer/policy/std Mean 0.440399 +trainer/policy/std Std 0.0220727 +trainer/policy/std Max 0.466477 +trainer/policy/std Min 0.406479 +trainer/Advantage Weights Mean 8.72754 +trainer/Advantage Weights Std 23.005 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.73717e-16 +trainer/Advantage Score Mean -0.135264 +trainer/Advantage Score Std 0.506388 +trainer/Advantage Score Max 1.66335 +trainer/Advantage Score Min -3.5523 +trainer/V1 Predictions Mean -72.6104 +trainer/V1 Predictions Std 15.7234 +trainer/V1 Predictions Max -2.32071 +trainer/V1 Predictions Min -87.3385 +trainer/VF Loss 0.0711754 +expl/num steps total 137000 +expl/num paths total 139 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0738051 +expl/Actions Std 0.886047 +expl/Actions Max 2.44135 +expl/Actions Min -2.47498 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 134630 +eval/num paths total 137 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.223032 +eval/Actions Std 0.71149 +eval/Actions Max 0.998539 +eval/Actions Min -0.99518 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.68221e-06 +time/evaluation sampling (s) 2.82051 +time/exploration sampling (s) 3.95679 +time/logging (s) 0.00688456 +time/saving (s) 0.00958379 +time/training (s) 14.3113 +time/epoch (s) 21.105 +time/total (s) 3016.12 +Epoch -864 +------------------------------ ---------------- +2022-05-15 18:53:02.497535 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -863 finished +------------------------------ ---------------- +epoch -863 +replay_buffer/size 999047 +trainer/num train calls 138000 +trainer/QF1 Loss 0.80336 +trainer/QF2 Loss 0.770918 +trainer/Policy Loss 8.37988 +trainer/Q1 Predictions Mean -71.072 +trainer/Q1 Predictions Std 18.6979 +trainer/Q1 Predictions Max -0.993421 +trainer/Q1 Predictions Min -86.949 +trainer/Q2 Predictions Mean -71.0034 +trainer/Q2 Predictions Std 18.8017 +trainer/Q2 Predictions Max -0.777187 +trainer/Q2 Predictions Min -86.4381 +trainer/Q Targets Mean -70.9311 +trainer/Q Targets Std 18.6913 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0483 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0208364 +trainer/policy/mean Std 0.704608 +trainer/policy/mean Max 0.999603 +trainer/policy/mean Min -0.995716 +trainer/policy/std Mean 0.441432 +trainer/policy/std Std 0.0219922 +trainer/policy/std Max 0.465455 +trainer/policy/std Min 0.405695 +trainer/Advantage Weights Mean 1.5834 +trainer/Advantage Weights Std 9.5286 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.45822e-14 +trainer/Advantage Score Mean -0.423135 +trainer/Advantage Score Std 0.432643 +trainer/Advantage Score Max 0.66399 +trainer/Advantage Score Min -3.13368 +trainer/V1 Predictions Mean -70.7185 +trainer/V1 Predictions Std 18.7703 +trainer/V1 Predictions Max -1.45608 +trainer/V1 Predictions Min -87.1153 +trainer/VF Loss 0.0411037 +expl/num steps total 138000 +expl/num paths total 140 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0474259 +expl/Actions Std 0.838881 +expl/Actions Max 2.59659 +expl/Actions Min -2.38824 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 135630 +eval/num paths total 138 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.140885 +eval/Actions Std 0.620008 +eval/Actions Max 0.999733 +eval/Actions Min -0.99961 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.61096e-06 +time/evaluation sampling (s) 2.90226 +time/exploration sampling (s) 3.51746 +time/logging (s) 0.012074 +time/saving (s) 0.0134954 +time/training (s) 14.033 +time/epoch (s) 20.4783 +time/total (s) 3036.61 +Epoch -863 +------------------------------ ---------------- +2022-05-15 18:53:23.870542 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -862 finished +------------------------------ ---------------- +epoch -862 +replay_buffer/size 999047 +trainer/num train calls 139000 +trainer/QF1 Loss 0.84345 +trainer/QF2 Loss 0.783906 +trainer/Policy Loss 12.6438 +trainer/Q1 Predictions Mean -71.53 +trainer/Q1 Predictions Std 18.4947 +trainer/Q1 Predictions Max -1.24242 +trainer/Q1 Predictions Min -87.6758 +trainer/Q2 Predictions Mean -71.5478 +trainer/Q2 Predictions Std 18.4956 +trainer/Q2 Predictions Max -1.22481 +trainer/Q2 Predictions Min -87.5096 +trainer/Q Targets Mean -71.3691 +trainer/Q Targets Std 18.462 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1237 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00717105 +trainer/policy/mean Std 0.694838 +trainer/policy/mean Max 0.996166 +trainer/policy/mean Min -0.997201 +trainer/policy/std Mean 0.438905 +trainer/policy/std Std 0.0213607 +trainer/policy/std Max 0.462948 +trainer/policy/std Min 0.401904 +trainer/Advantage Weights Mean 4.30841 +trainer/Advantage Weights Std 17.3916 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.3249e-15 +trainer/Advantage Score Mean -0.297029 +trainer/Advantage Score Std 0.550649 +trainer/Advantage Score Max 0.850525 +trainer/Advantage Score Min -3.28664 +trainer/V1 Predictions Mean -71.0504 +trainer/V1 Predictions Std 18.5536 +trainer/V1 Predictions Max -1.96325 +trainer/V1 Predictions Min -87.1724 +trainer/VF Loss 0.0522631 +expl/num steps total 139000 +expl/num paths total 141 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0378804 +expl/Actions Std 0.801377 +expl/Actions Max 2.29325 +expl/Actions Min -2.44381 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 136630 +eval/num paths total 139 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0608334 +eval/Actions Std 0.754564 +eval/Actions Max 0.998642 +eval/Actions Min -0.998149 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.71993e-06 +time/evaluation sampling (s) 3.37552 +time/exploration sampling (s) 3.65728 +time/logging (s) 0.00727839 +time/saving (s) 0.0124215 +time/training (s) 14.3055 +time/epoch (s) 21.358 +time/total (s) 3057.97 +Epoch -862 +------------------------------ ---------------- +2022-05-15 18:53:44.632459 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -861 finished +------------------------------ ---------------- +epoch -861 +replay_buffer/size 999047 +trainer/num train calls 140000 +trainer/QF1 Loss 0.752668 +trainer/QF2 Loss 0.738553 +trainer/Policy Loss 1.64869 +trainer/Q1 Predictions Mean -70.0897 +trainer/Q1 Predictions Std 18.4627 +trainer/Q1 Predictions Max -0.768446 +trainer/Q1 Predictions Min -86.9043 +trainer/Q2 Predictions Mean -70.0551 +trainer/Q2 Predictions Std 18.4462 +trainer/Q2 Predictions Max -1.15534 +trainer/Q2 Predictions Min -86.6671 +trainer/Q Targets Mean -69.6585 +trainer/Q Targets Std 18.6261 +trainer/Q Targets Max -1.58742 +trainer/Q Targets Min -86.6407 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00475182 +trainer/policy/mean Std 0.704883 +trainer/policy/mean Max 0.999829 +trainer/policy/mean Min -0.996637 +trainer/policy/std Mean 0.440275 +trainer/policy/std Std 0.0215411 +trainer/policy/std Max 0.463934 +trainer/policy/std Min 0.405104 +trainer/Advantage Weights Mean 0.534427 +trainer/Advantage Weights Std 6.28054 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.24176e-16 +trainer/Advantage Score Mean -0.645047 +trainer/Advantage Score Std 0.547994 +trainer/Advantage Score Max 0.56646 +trainer/Advantage Score Min -3.56652 +trainer/V1 Predictions Mean -69.3799 +trainer/V1 Predictions Std 18.8114 +trainer/V1 Predictions Max -0.846539 +trainer/V1 Predictions Min -86.4026 +trainer/VF Loss 0.072962 +expl/num steps total 140000 +expl/num paths total 142 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0368979 +expl/Actions Std 0.786983 +expl/Actions Max 2.26781 +expl/Actions Min -2.42428 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 137630 +eval/num paths total 140 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0423007 +eval/Actions Std 0.723158 +eval/Actions Max 0.99955 +eval/Actions Min -0.996191 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.77488e-06 +time/evaluation sampling (s) 3.06958 +time/exploration sampling (s) 3.27452 +time/logging (s) 0.00691287 +time/saving (s) 0.0104203 +time/training (s) 14.3928 +time/epoch (s) 20.7542 +time/total (s) 3078.73 +Epoch -861 +------------------------------ ---------------- +2022-05-15 18:54:04.974189 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -860 finished +------------------------------ ---------------- +epoch -860 +replay_buffer/size 999047 +trainer/num train calls 141000 +trainer/QF1 Loss 1.62166 +trainer/QF2 Loss 1.94684 +trainer/Policy Loss 19.3971 +trainer/Q1 Predictions Mean -71.5619 +trainer/Q1 Predictions Std 17.303 +trainer/Q1 Predictions Max -0.685442 +trainer/Q1 Predictions Min -86.3348 +trainer/Q2 Predictions Mean -71.503 +trainer/Q2 Predictions Std 17.3862 +trainer/Q2 Predictions Max -0.294176 +trainer/Q2 Predictions Min -86.6118 +trainer/Q Targets Mean -71.2912 +trainer/Q Targets Std 17.3858 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0913 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0340381 +trainer/policy/mean Std 0.710739 +trainer/policy/mean Max 0.998324 +trainer/policy/mean Min -0.999155 +trainer/policy/std Mean 0.440703 +trainer/policy/std Std 0.0212185 +trainer/policy/std Max 0.464665 +trainer/policy/std Min 0.407009 +trainer/Advantage Weights Mean 4.27444 +trainer/Advantage Weights Std 16.7098 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.593e-13 +trainer/Advantage Score Mean -0.381749 +trainer/Advantage Score Std 0.522918 +trainer/Advantage Score Max 1.14828 +trainer/Advantage Score Min -2.82121 +trainer/V1 Predictions Mean -71.0248 +trainer/V1 Predictions Std 17.5188 +trainer/V1 Predictions Max -0.367983 +trainer/V1 Predictions Min -86.9682 +trainer/VF Loss 0.0615548 +expl/num steps total 141000 +expl/num paths total 143 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0167873 +expl/Actions Std 0.793265 +expl/Actions Max 2.38052 +expl/Actions Min -2.33964 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 138630 +eval/num paths total 141 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.248003 +eval/Actions Std 0.816287 +eval/Actions Max 0.996989 +eval/Actions Min -0.998572 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.99094e-06 +time/evaluation sampling (s) 2.76608 +time/exploration sampling (s) 3.17457 +time/logging (s) 0.00986368 +time/saving (s) 0.0139007 +time/training (s) 14.3734 +time/epoch (s) 20.3378 +time/total (s) 3099.07 +Epoch -860 +------------------------------ ---------------- +2022-05-15 18:54:25.191505 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -859 finished +------------------------------ ---------------- +epoch -859 +replay_buffer/size 999047 +trainer/num train calls 142000 +trainer/QF1 Loss 0.72811 +trainer/QF2 Loss 0.697405 +trainer/Policy Loss 2.08716 +trainer/Q1 Predictions Mean -70.5161 +trainer/Q1 Predictions Std 18.1001 +trainer/Q1 Predictions Max -1.58834 +trainer/Q1 Predictions Min -86.7642 +trainer/Q2 Predictions Mean -70.5607 +trainer/Q2 Predictions Std 18.16 +trainer/Q2 Predictions Max -1.28472 +trainer/Q2 Predictions Min -86.5324 +trainer/Q Targets Mean -70.2484 +trainer/Q Targets Std 18.3339 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5409 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0150361 +trainer/policy/mean Std 0.694009 +trainer/policy/mean Max 0.998154 +trainer/policy/mean Min -0.99907 +trainer/policy/std Mean 0.439389 +trainer/policy/std Std 0.0212037 +trainer/policy/std Max 0.464839 +trainer/policy/std Min 0.404527 +trainer/Advantage Weights Mean 0.625951 +trainer/Advantage Weights Std 5.89402 +trainer/Advantage Weights Max 93.5716 +trainer/Advantage Weights Min 9.71096e-14 +trainer/Advantage Score Mean -0.592007 +trainer/Advantage Score Std 0.521141 +trainer/Advantage Score Max 0.453873 +trainer/Advantage Score Min -2.99629 +trainer/V1 Predictions Mean -69.9804 +trainer/V1 Predictions Std 18.3684 +trainer/V1 Predictions Max 0.173808 +trainer/V1 Predictions Min -86.4437 +trainer/VF Loss 0.063512 +expl/num steps total 142000 +expl/num paths total 144 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0867749 +expl/Actions Std 0.820692 +expl/Actions Max 2.50686 +expl/Actions Min -2.29304 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 139630 +eval/num paths total 142 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0265172 +eval/Actions Std 0.691031 +eval/Actions Max 0.999234 +eval/Actions Min -0.999521 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.79304e-06 +time/evaluation sampling (s) 2.46919 +time/exploration sampling (s) 3.39423 +time/logging (s) 0.00682272 +time/saving (s) 0.00936824 +time/training (s) 14.3248 +time/epoch (s) 20.2044 +time/total (s) 3119.29 +Epoch -859 +------------------------------ ---------------- +2022-05-15 18:54:44.692722 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -858 finished +------------------------------ ---------------- +epoch -858 +replay_buffer/size 999047 +trainer/num train calls 143000 +trainer/QF1 Loss 0.541404 +trainer/QF2 Loss 0.543905 +trainer/Policy Loss 30.7278 +trainer/Q1 Predictions Mean -71.0158 +trainer/Q1 Predictions Std 18.1765 +trainer/Q1 Predictions Max -1.19618 +trainer/Q1 Predictions Min -86.9634 +trainer/Q2 Predictions Mean -70.9557 +trainer/Q2 Predictions Std 18.1984 +trainer/Q2 Predictions Max -1.27295 +trainer/Q2 Predictions Min -87.1367 +trainer/Q Targets Mean -70.9501 +trainer/Q Targets Std 18.2757 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6221 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0124458 +trainer/policy/mean Std 0.691628 +trainer/policy/mean Max 0.9977 +trainer/policy/mean Min -0.997675 +trainer/policy/std Mean 0.4413 +trainer/policy/std Std 0.0216771 +trainer/policy/std Max 0.465801 +trainer/policy/std Min 0.405556 +trainer/Advantage Weights Mean 5.94031 +trainer/Advantage Weights Std 20.1536 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.56721e-13 +trainer/Advantage Score Mean -0.312072 +trainer/Advantage Score Std 0.520089 +trainer/Advantage Score Max 1.21684 +trainer/Advantage Score Min -2.94843 +trainer/V1 Predictions Mean -70.6993 +trainer/V1 Predictions Std 18.2897 +trainer/V1 Predictions Max -0.994871 +trainer/V1 Predictions Min -86.9845 +trainer/VF Loss 0.0601549 +expl/num steps total 143000 +expl/num paths total 145 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0374431 +expl/Actions Std 0.794131 +expl/Actions Max 2.47358 +expl/Actions Min -2.20643 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 140630 +eval/num paths total 143 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.154887 +eval/Actions Std 0.627571 +eval/Actions Max 0.998304 +eval/Actions Min -0.995139 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.15978e-06 +time/evaluation sampling (s) 2.74002 +time/exploration sampling (s) 3.2884 +time/logging (s) 0.00724582 +time/saving (s) 0.0116899 +time/training (s) 13.4485 +time/epoch (s) 19.4959 +time/total (s) 3138.78 +Epoch -858 +------------------------------ ---------------- +2022-05-15 18:55:05.358187 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -857 finished +------------------------------ ---------------- +epoch -857 +replay_buffer/size 999047 +trainer/num train calls 144000 +trainer/QF1 Loss 0.499722 +trainer/QF2 Loss 0.463733 +trainer/Policy Loss 5.9195 +trainer/Q1 Predictions Mean -71.2092 +trainer/Q1 Predictions Std 18.7784 +trainer/Q1 Predictions Max -0.533182 +trainer/Q1 Predictions Min -87.6438 +trainer/Q2 Predictions Mean -71.2171 +trainer/Q2 Predictions Std 18.7784 +trainer/Q2 Predictions Max -0.487055 +trainer/Q2 Predictions Min -86.8871 +trainer/Q Targets Mean -71.0299 +trainer/Q Targets Std 18.7066 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3644 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00787313 +trainer/policy/mean Std 0.709198 +trainer/policy/mean Max 0.998592 +trainer/policy/mean Min -0.995529 +trainer/policy/std Mean 0.440418 +trainer/policy/std Std 0.0226934 +trainer/policy/std Max 0.46861 +trainer/policy/std Min 0.403023 +trainer/Advantage Weights Mean 2.19498 +trainer/Advantage Weights Std 11.6755 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.16721e-15 +trainer/Advantage Score Mean -0.391526 +trainer/Advantage Score Std 0.483746 +trainer/Advantage Score Max 0.645119 +trainer/Advantage Score Min -3.33859 +trainer/V1 Predictions Mean -70.7815 +trainer/V1 Predictions Std 18.7879 +trainer/V1 Predictions Max -0.569405 +trainer/V1 Predictions Min -86.7716 +trainer/VF Loss 0.0446811 +expl/num steps total 144000 +expl/num paths total 146 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0206386 +expl/Actions Std 0.844471 +expl/Actions Max 2.55238 +expl/Actions Min -2.62225 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 141630 +eval/num paths total 144 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0499246 +eval/Actions Std 0.724429 +eval/Actions Max 0.999607 +eval/Actions Min -0.997547 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.7162e-06 +time/evaluation sampling (s) 2.989 +time/exploration sampling (s) 3.47322 +time/logging (s) 0.00808822 +time/saving (s) 0.0114355 +time/training (s) 14.1779 +time/epoch (s) 20.6596 +time/total (s) 3159.45 +Epoch -857 +------------------------------ ---------------- +2022-05-15 18:55:26.271947 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -856 finished +------------------------------ ---------------- +epoch -856 +replay_buffer/size 999047 +trainer/num train calls 145000 +trainer/QF1 Loss 1.00785 +trainer/QF2 Loss 0.930115 +trainer/Policy Loss 33.318 +trainer/Q1 Predictions Mean -70.2393 +trainer/Q1 Predictions Std 19.1702 +trainer/Q1 Predictions Max -0.296536 +trainer/Q1 Predictions Min -86.371 +trainer/Q2 Predictions Mean -70.2931 +trainer/Q2 Predictions Std 19.1479 +trainer/Q2 Predictions Max -0.514394 +trainer/Q2 Predictions Min -86.4026 +trainer/Q Targets Mean -70.0073 +trainer/Q Targets Std 19.1743 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.6159 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000298144 +trainer/policy/mean Std 0.69575 +trainer/policy/mean Max 0.998916 +trainer/policy/mean Min -0.999295 +trainer/policy/std Mean 0.439018 +trainer/policy/std Std 0.0217432 +trainer/policy/std Max 0.467567 +trainer/policy/std Min 0.403715 +trainer/Advantage Weights Mean 6.4793 +trainer/Advantage Weights Std 21.9696 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.18034e-13 +trainer/Advantage Score Mean -0.286824 +trainer/Advantage Score Std 0.467378 +trainer/Advantage Score Max 1.07715 +trainer/Advantage Score Min -2.82887 +trainer/V1 Predictions Mean -69.7747 +trainer/V1 Predictions Std 19.2907 +trainer/V1 Predictions Max -0.675071 +trainer/V1 Predictions Min -85.718 +trainer/VF Loss 0.0528117 +expl/num steps total 145000 +expl/num paths total 147 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0467279 +expl/Actions Std 0.819719 +expl/Actions Max 2.22098 +expl/Actions Min -2.32335 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 142630 +eval/num paths total 145 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.16589 +eval/Actions Std 0.762213 +eval/Actions Max 0.999399 +eval/Actions Min -0.997731 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.14787e-06 +time/evaluation sampling (s) 3.24381 +time/exploration sampling (s) 3.64253 +time/logging (s) 0.00826709 +time/saving (s) 0.0102053 +time/training (s) 14.0018 +time/epoch (s) 20.9066 +time/total (s) 3180.36 +Epoch -856 +------------------------------ ---------------- +2022-05-15 18:55:47.237868 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -855 finished +------------------------------ ---------------- +epoch -855 +replay_buffer/size 999047 +trainer/num train calls 146000 +trainer/QF1 Loss 0.906133 +trainer/QF2 Loss 1.00489 +trainer/Policy Loss 22.7922 +trainer/Q1 Predictions Mean -70.7508 +trainer/Q1 Predictions Std 17.8521 +trainer/Q1 Predictions Max -2.93199 +trainer/Q1 Predictions Min -86.6008 +trainer/Q2 Predictions Mean -70.7183 +trainer/Q2 Predictions Std 17.7835 +trainer/Q2 Predictions Max -3.19771 +trainer/Q2 Predictions Min -86.321 +trainer/Q Targets Mean -71.2695 +trainer/Q Targets Std 17.8674 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6311 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00245376 +trainer/policy/mean Std 0.700232 +trainer/policy/mean Max 0.999619 +trainer/policy/mean Min -0.997732 +trainer/policy/std Mean 0.441096 +trainer/policy/std Std 0.020577 +trainer/policy/std Max 0.466318 +trainer/policy/std Min 0.407904 +trainer/Advantage Weights Mean 4.59823 +trainer/Advantage Weights Std 17.2365 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.55595e-15 +trainer/Advantage Score Mean -0.26051 +trainer/Advantage Score Std 0.48081 +trainer/Advantage Score Max 0.871189 +trainer/Advantage Score Min -3.25164 +trainer/V1 Predictions Mean -70.9193 +trainer/V1 Predictions Std 18.0403 +trainer/V1 Predictions Max -2.07527 +trainer/V1 Predictions Min -86.2828 +trainer/VF Loss 0.0434339 +expl/num steps total 146000 +expl/num paths total 148 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0017704 +expl/Actions Std 0.83389 +expl/Actions Max 2.45727 +expl/Actions Min -2.35337 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 143630 +eval/num paths total 146 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.092914 +eval/Actions Std 0.665494 +eval/Actions Max 0.998781 +eval/Actions Min -0.997207 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.624e-06 +time/evaluation sampling (s) 3.09715 +time/exploration sampling (s) 3.43398 +time/logging (s) 0.0106635 +time/saving (s) 0.0158495 +time/training (s) 14.4048 +time/epoch (s) 20.9624 +time/total (s) 3201.33 +Epoch -855 +------------------------------ ---------------- +2022-05-15 18:56:08.530985 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -854 finished +------------------------------ ---------------- +epoch -854 +replay_buffer/size 999047 +trainer/num train calls 147000 +trainer/QF1 Loss 0.718909 +trainer/QF2 Loss 0.761346 +trainer/Policy Loss 24.2725 +trainer/Q1 Predictions Mean -70.8625 +trainer/Q1 Predictions Std 18.8792 +trainer/Q1 Predictions Max -1.176 +trainer/Q1 Predictions Min -86.3847 +trainer/Q2 Predictions Mean -70.8792 +trainer/Q2 Predictions Std 18.8518 +trainer/Q2 Predictions Max -0.833399 +trainer/Q2 Predictions Min -86.4293 +trainer/Q Targets Mean -71.1333 +trainer/Q Targets Std 18.962 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4689 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00117725 +trainer/policy/mean Std 0.692161 +trainer/policy/mean Max 0.995721 +trainer/policy/mean Min -0.997065 +trainer/policy/std Mean 0.439073 +trainer/policy/std Std 0.0206263 +trainer/policy/std Max 0.463442 +trainer/policy/std Min 0.405299 +trainer/Advantage Weights Mean 4.96538 +trainer/Advantage Weights Std 16.7976 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.14167e-09 +trainer/Advantage Score Mean -0.213037 +trainer/Advantage Score Std 0.446952 +trainer/Advantage Score Max 1.0308 +trainer/Advantage Score Min -1.99617 +trainer/V1 Predictions Mean -70.858 +trainer/V1 Predictions Std 19.0425 +trainer/V1 Predictions Max 0.166667 +trainer/V1 Predictions Min -86.3305 +trainer/VF Loss 0.040744 +expl/num steps total 147000 +expl/num paths total 149 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0817082 +expl/Actions Std 0.826972 +expl/Actions Max 2.40353 +expl/Actions Min -2.35631 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 144630 +eval/num paths total 147 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.151087 +eval/Actions Std 0.730204 +eval/Actions Max 0.998899 +eval/Actions Min -0.997982 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.66731e-06 +time/evaluation sampling (s) 3.21437 +time/exploration sampling (s) 3.68657 +time/logging (s) 0.00901081 +time/saving (s) 0.0127089 +time/training (s) 14.3601 +time/epoch (s) 21.2827 +time/total (s) 3222.62 +Epoch -854 +------------------------------ ---------------- +2022-05-15 18:56:29.389979 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -853 finished +------------------------------ ---------------- +epoch -853 +replay_buffer/size 999047 +trainer/num train calls 148000 +trainer/QF1 Loss 0.985764 +trainer/QF2 Loss 1.10978 +trainer/Policy Loss 22.8485 +trainer/Q1 Predictions Mean -71.1181 +trainer/Q1 Predictions Std 17.8311 +trainer/Q1 Predictions Max -3.0076 +trainer/Q1 Predictions Min -86.9919 +trainer/Q2 Predictions Mean -71.1235 +trainer/Q2 Predictions Std 17.8046 +trainer/Q2 Predictions Max -3.81788 +trainer/Q2 Predictions Min -87.3689 +trainer/Q Targets Mean -71.2464 +trainer/Q Targets Std 18.1382 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3173 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0116616 +trainer/policy/mean Std 0.68987 +trainer/policy/mean Max 0.997799 +trainer/policy/mean Min -0.995863 +trainer/policy/std Mean 0.438224 +trainer/policy/std Std 0.0215635 +trainer/policy/std Max 0.463168 +trainer/policy/std Min 0.403948 +trainer/Advantage Weights Mean 4.58069 +trainer/Advantage Weights Std 16.4095 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.95009e-14 +trainer/Advantage Score Mean -0.287418 +trainer/Advantage Score Std 0.498122 +trainer/Advantage Score Max 1.15506 +trainer/Advantage Score Min -3.04528 +trainer/V1 Predictions Mean -70.9777 +trainer/V1 Predictions Std 18.1631 +trainer/V1 Predictions Max -3.04525 +trainer/V1 Predictions Min -86.9953 +trainer/VF Loss 0.0500064 +expl/num steps total 148000 +expl/num paths total 150 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0728737 +expl/Actions Std 0.842168 +expl/Actions Max 2.99807 +expl/Actions Min -2.47618 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 145630 +eval/num paths total 148 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.144343 +eval/Actions Std 0.686513 +eval/Actions Max 0.999857 +eval/Actions Min -0.996172 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.93092e-06 +time/evaluation sampling (s) 3.20589 +time/exploration sampling (s) 3.22028 +time/logging (s) 0.00736031 +time/saving (s) 0.0105499 +time/training (s) 14.4074 +time/epoch (s) 20.8514 +time/total (s) 3243.47 +Epoch -853 +------------------------------ ---------------- +2022-05-15 18:56:49.092410 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -852 finished +------------------------------ ---------------- +epoch -852 +replay_buffer/size 999047 +trainer/num train calls 149000 +trainer/QF1 Loss 0.512872 +trainer/QF2 Loss 0.485064 +trainer/Policy Loss 37.7076 +trainer/Q1 Predictions Mean -72.2284 +trainer/Q1 Predictions Std 16.4505 +trainer/Q1 Predictions Max -0.721999 +trainer/Q1 Predictions Min -86.8615 +trainer/Q2 Predictions Mean -72.2916 +trainer/Q2 Predictions Std 16.467 +trainer/Q2 Predictions Max -0.546857 +trainer/Q2 Predictions Min -87.2264 +trainer/Q Targets Mean -72.5861 +trainer/Q Targets Std 16.3682 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4708 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0076271 +trainer/policy/mean Std 0.719984 +trainer/policy/mean Max 0.997587 +trainer/policy/mean Min -0.997943 +trainer/policy/std Mean 0.436843 +trainer/policy/std Std 0.0229821 +trainer/policy/std Max 0.464896 +trainer/policy/std Min 0.400622 +trainer/Advantage Weights Mean 7.44436 +trainer/Advantage Weights Std 22.217 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.07389e-10 +trainer/Advantage Score Mean -0.151713 +trainer/Advantage Score Std 0.419239 +trainer/Advantage Score Max 1.01018 +trainer/Advantage Score Min -2.29546 +trainer/V1 Predictions Mean -72.3397 +trainer/V1 Predictions Std 16.4144 +trainer/V1 Predictions Max -1.02183 +trainer/V1 Predictions Min -87.0038 +trainer/VF Loss 0.0436111 +expl/num steps total 149000 +expl/num paths total 151 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0221908 +expl/Actions Std 0.890186 +expl/Actions Max 2.59036 +expl/Actions Min -2.30472 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 146630 +eval/num paths total 149 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0452151 +eval/Actions Std 0.729429 +eval/Actions Max 0.997598 +eval/Actions Min -0.997736 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.69804e-06 +time/evaluation sampling (s) 3.20277 +time/exploration sampling (s) 3.17253 +time/logging (s) 0.0070966 +time/saving (s) 0.0104769 +time/training (s) 13.3034 +time/epoch (s) 19.6963 +time/total (s) 3263.17 +Epoch -852 +------------------------------ ---------------- +2022-05-15 18:57:09.828722 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -851 finished +------------------------------ ---------------- +epoch -851 +replay_buffer/size 999047 +trainer/num train calls 150000 +trainer/QF1 Loss 0.47944 +trainer/QF2 Loss 0.534123 +trainer/Policy Loss 29.6576 +trainer/Q1 Predictions Mean -68.9771 +trainer/Q1 Predictions Std 20.2212 +trainer/Q1 Predictions Max -0.707807 +trainer/Q1 Predictions Min -87.1077 +trainer/Q2 Predictions Mean -68.9562 +trainer/Q2 Predictions Std 20.1673 +trainer/Q2 Predictions Max -1.06585 +trainer/Q2 Predictions Min -86.3584 +trainer/Q Targets Mean -68.8909 +trainer/Q Targets Std 20.4868 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7664 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.024182 +trainer/policy/mean Std 0.673532 +trainer/policy/mean Max 0.998363 +trainer/policy/mean Min -0.998874 +trainer/policy/std Mean 0.438342 +trainer/policy/std Std 0.0228917 +trainer/policy/std Max 0.465828 +trainer/policy/std Min 0.401653 +trainer/Advantage Weights Mean 5.82069 +trainer/Advantage Weights Std 18.4694 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.32523e-11 +trainer/Advantage Score Mean -0.270769 +trainer/Advantage Score Std 0.505069 +trainer/Advantage Score Max 1.0901 +trainer/Advantage Score Min -2.44846 +trainer/V1 Predictions Mean -68.6329 +trainer/V1 Predictions Std 20.6043 +trainer/V1 Predictions Max 0.183923 +trainer/V1 Predictions Min -86.2641 +trainer/VF Loss 0.0563876 +expl/num steps total 150000 +expl/num paths total 152 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0586197 +expl/Actions Std 0.803596 +expl/Actions Max 2.89899 +expl/Actions Min -2.53172 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 147224 +eval/num paths total 150 +eval/path length Mean 594 +eval/path length Std 0 +eval/path length Max 594 +eval/path length Min 594 +eval/Rewards Mean 0.0016835 +eval/Rewards Std 0.0409959 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0492305 +eval/Actions Std 0.703874 +eval/Actions Max 0.999671 +eval/Actions Min -0.998267 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.02632e-05 +time/evaluation sampling (s) 3.01317 +time/exploration sampling (s) 3.12444 +time/logging (s) 0.00642078 +time/saving (s) 0.010486 +time/training (s) 14.5738 +time/epoch (s) 20.7283 +time/total (s) 3283.9 +Epoch -851 +------------------------------ ---------------- +2022-05-15 18:57:30.731914 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -850 finished +------------------------------ ---------------- +epoch -850 +replay_buffer/size 999047 +trainer/num train calls 151000 +trainer/QF1 Loss 0.611683 +trainer/QF2 Loss 0.552166 +trainer/Policy Loss 8.43069 +trainer/Q1 Predictions Mean -71.348 +trainer/Q1 Predictions Std 16.6318 +trainer/Q1 Predictions Max -1.5088 +trainer/Q1 Predictions Min -86.5849 +trainer/Q2 Predictions Mean -71.3503 +trainer/Q2 Predictions Std 16.6596 +trainer/Q2 Predictions Max -1.30116 +trainer/Q2 Predictions Min -86.5594 +trainer/Q Targets Mean -71.5614 +trainer/Q Targets Std 16.6277 +trainer/Q Targets Max -1.51838 +trainer/Q Targets Min -86.8838 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00116406 +trainer/policy/mean Std 0.710557 +trainer/policy/mean Max 0.998625 +trainer/policy/mean Min -0.998607 +trainer/policy/std Mean 0.438903 +trainer/policy/std Std 0.0229971 +trainer/policy/std Max 0.463037 +trainer/policy/std Min 0.401372 +trainer/Advantage Weights Mean 2.03474 +trainer/Advantage Weights Std 11.1332 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.44638e-10 +trainer/Advantage Score Mean -0.424247 +trainer/Advantage Score Std 0.435874 +trainer/Advantage Score Max 0.768707 +trainer/Advantage Score Min -2.26568 +trainer/V1 Predictions Mean -71.3266 +trainer/V1 Predictions Std 16.6639 +trainer/V1 Predictions Max -0.662154 +trainer/V1 Predictions Min -86.7553 +trainer/VF Loss 0.0426614 +expl/num steps total 151000 +expl/num paths total 153 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.19849 +expl/Actions Std 0.822742 +expl/Actions Max 2.77831 +expl/Actions Min -2.42449 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 148224 +eval/num paths total 151 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.374449 +eval/Actions Std 0.648727 +eval/Actions Max 0.998835 +eval/Actions Min -0.997239 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0537e-05 +time/evaluation sampling (s) 2.8326 +time/exploration sampling (s) 3.51835 +time/logging (s) 0.0100553 +time/saving (s) 0.0253203 +time/training (s) 14.5147 +time/epoch (s) 20.9011 +time/total (s) 3304.81 +Epoch -850 +------------------------------ ---------------- +2022-05-15 18:57:51.792032 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -849 finished +------------------------------ ---------------- +epoch -849 +replay_buffer/size 999047 +trainer/num train calls 152000 +trainer/QF1 Loss 0.533338 +trainer/QF2 Loss 0.581328 +trainer/Policy Loss 9.0022 +trainer/Q1 Predictions Mean -70.0768 +trainer/Q1 Predictions Std 19.9837 +trainer/Q1 Predictions Max -0.421618 +trainer/Q1 Predictions Min -86.7722 +trainer/Q2 Predictions Mean -70.046 +trainer/Q2 Predictions Std 20.033 +trainer/Q2 Predictions Max -0.761088 +trainer/Q2 Predictions Min -86.5038 +trainer/Q Targets Mean -70.2377 +trainer/Q Targets Std 19.9377 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6811 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0231275 +trainer/policy/mean Std 0.698107 +trainer/policy/mean Max 0.998013 +trainer/policy/mean Min -0.998782 +trainer/policy/std Mean 0.439589 +trainer/policy/std Std 0.0217763 +trainer/policy/std Max 0.461278 +trainer/policy/std Min 0.404299 +trainer/Advantage Weights Mean 2.31553 +trainer/Advantage Weights Std 12.3531 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.24735e-12 +trainer/Advantage Score Mean -0.426601 +trainer/Advantage Score Std 0.459243 +trainer/Advantage Score Max 1.10246 +trainer/Advantage Score Min -2.64532 +trainer/V1 Predictions Mean -69.9848 +trainer/V1 Predictions Std 19.9879 +trainer/V1 Predictions Max -0.11718 +trainer/V1 Predictions Min -86.698 +trainer/VF Loss 0.0504032 +expl/num steps total 152000 +expl/num paths total 154 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0348349 +expl/Actions Std 0.810985 +expl/Actions Max 2.70016 +expl/Actions Min -2.28028 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 149224 +eval/num paths total 152 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.264138 +eval/Actions Std 0.652531 +eval/Actions Max 0.998201 +eval/Actions Min -0.99749 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.34695e-06 +time/evaluation sampling (s) 3.07363 +time/exploration sampling (s) 3.57398 +time/logging (s) 0.0117389 +time/saving (s) 0.0162505 +time/training (s) 14.3759 +time/epoch (s) 21.0515 +time/total (s) 3325.87 +Epoch -849 +------------------------------ ---------------- +2022-05-15 18:58:12.557928 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -848 finished +------------------------------ ---------------- +epoch -848 +replay_buffer/size 999047 +trainer/num train calls 153000 +trainer/QF1 Loss 1.39003 +trainer/QF2 Loss 1.40523 +trainer/Policy Loss 5.40524 +trainer/Q1 Predictions Mean -72.1756 +trainer/Q1 Predictions Std 17.3298 +trainer/Q1 Predictions Max -2.56739 +trainer/Q1 Predictions Min -87.264 +trainer/Q2 Predictions Mean -72.1227 +trainer/Q2 Predictions Std 17.2775 +trainer/Q2 Predictions Max -2.56529 +trainer/Q2 Predictions Min -88.0588 +trainer/Q Targets Mean -71.5643 +trainer/Q Targets Std 17.6784 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2548 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00586857 +trainer/policy/mean Std 0.699196 +trainer/policy/mean Max 0.993126 +trainer/policy/mean Min -0.996476 +trainer/policy/std Mean 0.438248 +trainer/policy/std Std 0.0219876 +trainer/policy/std Max 0.463386 +trainer/policy/std Min 0.403094 +trainer/Advantage Weights Mean 1.39322 +trainer/Advantage Weights Std 9.36136 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.78926e-21 +trainer/Advantage Score Mean -0.647404 +trainer/Advantage Score Std 0.579672 +trainer/Advantage Score Max 0.803398 +trainer/Advantage Score Min -4.73285 +trainer/V1 Predictions Mean -71.3633 +trainer/V1 Predictions Std 17.5696 +trainer/V1 Predictions Max -1.88192 +trainer/V1 Predictions Min -87.1539 +trainer/VF Loss 0.0800827 +expl/num steps total 153000 +expl/num paths total 155 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.111577 +expl/Actions Std 0.836547 +expl/Actions Max 2.28007 +expl/Actions Min -2.33888 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 150224 +eval/num paths total 153 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0632581 +eval/Actions Std 0.781545 +eval/Actions Max 0.999672 +eval/Actions Min -0.999657 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.74181e-06 +time/evaluation sampling (s) 3.15243 +time/exploration sampling (s) 3.46615 +time/logging (s) 0.00918208 +time/saving (s) 0.0144385 +time/training (s) 14.1107 +time/epoch (s) 20.7529 +time/total (s) 3346.63 +Epoch -848 +------------------------------ ---------------- +2022-05-15 18:58:32.995912 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -847 finished +------------------------------ ---------------- +epoch -847 +replay_buffer/size 999047 +trainer/num train calls 154000 +trainer/QF1 Loss 0.601486 +trainer/QF2 Loss 0.659329 +trainer/Policy Loss 12.0586 +trainer/Q1 Predictions Mean -70.0371 +trainer/Q1 Predictions Std 19.0396 +trainer/Q1 Predictions Max -0.469884 +trainer/Q1 Predictions Min -86.8076 +trainer/Q2 Predictions Mean -70.0353 +trainer/Q2 Predictions Std 19.065 +trainer/Q2 Predictions Max -0.414424 +trainer/Q2 Predictions Min -86.8771 +trainer/Q Targets Mean -69.8306 +trainer/Q Targets Std 18.9855 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7692 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0144998 +trainer/policy/mean Std 0.697841 +trainer/policy/mean Max 0.998256 +trainer/policy/mean Min -0.999419 +trainer/policy/std Mean 0.437419 +trainer/policy/std Std 0.0229018 +trainer/policy/std Max 0.461956 +trainer/policy/std Min 0.402516 +trainer/Advantage Weights Mean 3.26223 +trainer/Advantage Weights Std 17.1654 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.41044e-12 +trainer/Advantage Score Mean -0.476807 +trainer/Advantage Score Std 0.419829 +trainer/Advantage Score Max 1.22125 +trainer/Advantage Score Min -2.72871 +trainer/V1 Predictions Mean -69.5396 +trainer/V1 Predictions Std 19.1824 +trainer/V1 Predictions Max 0.632282 +trainer/V1 Predictions Min -86.7131 +trainer/VF Loss 0.0533109 +expl/num steps total 154000 +expl/num paths total 156 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0490659 +expl/Actions Std 0.830869 +expl/Actions Max 2.52323 +expl/Actions Min -2.21898 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 151224 +eval/num paths total 154 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0531011 +eval/Actions Std 0.751735 +eval/Actions Max 0.999305 +eval/Actions Min -0.996983 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.15392e-06 +time/evaluation sampling (s) 3.07665 +time/exploration sampling (s) 3.40018 +time/logging (s) 0.00785883 +time/saving (s) 0.0128771 +time/training (s) 13.9337 +time/epoch (s) 20.4313 +time/total (s) 3367.06 +Epoch -847 +------------------------------ ---------------- +2022-05-15 18:58:53.957986 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -846 finished +------------------------------ ---------------- +epoch -846 +replay_buffer/size 999047 +trainer/num train calls 155000 +trainer/QF1 Loss 0.97635 +trainer/QF2 Loss 0.969285 +trainer/Policy Loss 16.5483 +trainer/Q1 Predictions Mean -70.6815 +trainer/Q1 Predictions Std 19.346 +trainer/Q1 Predictions Max -2.86093 +trainer/Q1 Predictions Min -87.0279 +trainer/Q2 Predictions Mean -70.6369 +trainer/Q2 Predictions Std 19.2831 +trainer/Q2 Predictions Max -3.56888 +trainer/Q2 Predictions Min -86.8965 +trainer/Q Targets Mean -70.3802 +trainer/Q Targets Std 19.3823 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8663 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.020516 +trainer/policy/mean Std 0.707364 +trainer/policy/mean Max 0.997826 +trainer/policy/mean Min -0.998058 +trainer/policy/std Mean 0.438117 +trainer/policy/std Std 0.022234 +trainer/policy/std Max 0.462373 +trainer/policy/std Min 0.401623 +trainer/Advantage Weights Mean 3.88354 +trainer/Advantage Weights Std 17.4686 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.48014e-14 +trainer/Advantage Score Mean -0.416667 +trainer/Advantage Score Std 0.501186 +trainer/Advantage Score Max 1.32515 +trainer/Advantage Score Min -3.13279 +trainer/V1 Predictions Mean -70.1251 +trainer/V1 Predictions Std 19.5048 +trainer/V1 Predictions Max -1.4816 +trainer/V1 Predictions Min -86.6004 +trainer/VF Loss 0.0658306 +expl/num steps total 155000 +expl/num paths total 157 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0639733 +expl/Actions Std 0.88076 +expl/Actions Max 2.42014 +expl/Actions Min -2.80407 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 152224 +eval/num paths total 155 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.155156 +eval/Actions Std 0.654097 +eval/Actions Max 0.999503 +eval/Actions Min -0.99974 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.65241e-06 +time/evaluation sampling (s) 3.366 +time/exploration sampling (s) 3.28325 +time/logging (s) 0.00898996 +time/saving (s) 0.0125869 +time/training (s) 14.2853 +time/epoch (s) 20.9561 +time/total (s) 3388.02 +Epoch -846 +------------------------------ ---------------- +2022-05-15 18:59:14.287672 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -845 finished +------------------------------ ---------------- +epoch -845 +replay_buffer/size 999047 +trainer/num train calls 156000 +trainer/QF1 Loss 0.434385 +trainer/QF2 Loss 0.459789 +trainer/Policy Loss 33.7409 +trainer/Q1 Predictions Mean -72.1256 +trainer/Q1 Predictions Std 17.9728 +trainer/Q1 Predictions Max -1.23782 +trainer/Q1 Predictions Min -87.3819 +trainer/Q2 Predictions Mean -72.0432 +trainer/Q2 Predictions Std 17.9897 +trainer/Q2 Predictions Max -0.888772 +trainer/Q2 Predictions Min -87.0862 +trainer/Q Targets Mean -72.1916 +trainer/Q Targets Std 17.7717 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0196 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0137066 +trainer/policy/mean Std 0.698186 +trainer/policy/mean Max 0.99764 +trainer/policy/mean Min -0.996982 +trainer/policy/std Mean 0.437139 +trainer/policy/std Std 0.021254 +trainer/policy/std Max 0.460887 +trainer/policy/std Min 0.40251 +trainer/Advantage Weights Mean 5.64613 +trainer/Advantage Weights Std 17.0286 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.65501e-09 +trainer/Advantage Score Mean -0.189364 +trainer/Advantage Score Std 0.422335 +trainer/Advantage Score Max 1.07944 +trainer/Advantage Score Min -1.88279 +trainer/V1 Predictions Mean -71.9174 +trainer/V1 Predictions Std 17.9058 +trainer/V1 Predictions Max -0.75648 +trainer/V1 Predictions Min -86.8864 +trainer/VF Loss 0.0397519 +expl/num steps total 156000 +expl/num paths total 159 +expl/path length Mean 500 +expl/path length Std 429 +expl/path length Max 929 +expl/path length Min 71 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0349532 +expl/Actions Std 0.819787 +expl/Actions Max 2.30634 +expl/Actions Min -2.2627 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 153224 +eval/num paths total 156 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0518042 +eval/Actions Std 0.680213 +eval/Actions Max 0.997911 +eval/Actions Min -0.997078 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.77115e-06 +time/evaluation sampling (s) 3.04739 +time/exploration sampling (s) 3.09906 +time/logging (s) 0.00760866 +time/saving (s) 0.011385 +time/training (s) 14.1572 +time/epoch (s) 20.3227 +time/total (s) 3408.35 +Epoch -845 +------------------------------ ---------------- +2022-05-15 18:59:34.366287 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -844 finished +------------------------------ ---------------- +epoch -844 +replay_buffer/size 999047 +trainer/num train calls 157000 +trainer/QF1 Loss 1.24024 +trainer/QF2 Loss 1.31862 +trainer/Policy Loss 9.20631 +trainer/Q1 Predictions Mean -71.7232 +trainer/Q1 Predictions Std 18.224 +trainer/Q1 Predictions Max -5.54746 +trainer/Q1 Predictions Min -88.5654 +trainer/Q2 Predictions Mean -71.6439 +trainer/Q2 Predictions Std 18.1876 +trainer/Q2 Predictions Max -4.77726 +trainer/Q2 Predictions Min -88.5268 +trainer/Q Targets Mean -71.8033 +trainer/Q Targets Std 18.2395 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8465 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0298068 +trainer/policy/mean Std 0.702833 +trainer/policy/mean Max 0.998273 +trainer/policy/mean Min -0.999286 +trainer/policy/std Mean 0.436687 +trainer/policy/std Std 0.0220299 +trainer/policy/std Max 0.46146 +trainer/policy/std Min 0.403084 +trainer/Advantage Weights Mean 2.10377 +trainer/Advantage Weights Std 10.8143 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.63843e-13 +trainer/Advantage Score Mean -0.388769 +trainer/Advantage Score Std 0.427152 +trainer/Advantage Score Max 0.658279 +trainer/Advantage Score Min -2.77774 +trainer/V1 Predictions Mean -71.514 +trainer/V1 Predictions Std 18.2212 +trainer/V1 Predictions Max -4.6393 +trainer/V1 Predictions Min -87.623 +trainer/VF Loss 0.0390727 +expl/num steps total 157000 +expl/num paths total 161 +expl/path length Mean 500 +expl/path length Std 451 +expl/path length Max 951 +expl/path length Min 49 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0327675 +expl/Actions Std 0.830715 +expl/Actions Max 2.26147 +expl/Actions Min -2.65396 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 154224 +eval/num paths total 157 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.054076 +eval/Actions Std 0.715468 +eval/Actions Max 0.999192 +eval/Actions Min -0.999734 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.9728e-06 +time/evaluation sampling (s) 2.67685 +time/exploration sampling (s) 3.41601 +time/logging (s) 0.00772841 +time/saving (s) 0.0106719 +time/training (s) 13.9598 +time/epoch (s) 20.0711 +time/total (s) 3428.43 +Epoch -844 +------------------------------ ---------------- +2022-05-15 18:59:54.799799 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -843 finished +------------------------------ ---------------- +epoch -843 +replay_buffer/size 999047 +trainer/num train calls 158000 +trainer/QF1 Loss 0.820981 +trainer/QF2 Loss 0.826867 +trainer/Policy Loss 12.9657 +trainer/Q1 Predictions Mean -72.9969 +trainer/Q1 Predictions Std 14.6458 +trainer/Q1 Predictions Max -4.52907 +trainer/Q1 Predictions Min -86.6177 +trainer/Q2 Predictions Mean -72.962 +trainer/Q2 Predictions Std 14.6497 +trainer/Q2 Predictions Max -4.33798 +trainer/Q2 Predictions Min -86.4429 +trainer/Q Targets Mean -73.0535 +trainer/Q Targets Std 14.9238 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7526 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0136894 +trainer/policy/mean Std 0.705034 +trainer/policy/mean Max 0.998088 +trainer/policy/mean Min -0.998297 +trainer/policy/std Mean 0.436811 +trainer/policy/std Std 0.0218127 +trainer/policy/std Max 0.460389 +trainer/policy/std Min 0.404272 +trainer/Advantage Weights Mean 3.38908 +trainer/Advantage Weights Std 15.5319 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.14119e-11 +trainer/Advantage Score Mean -0.377184 +trainer/Advantage Score Std 0.471087 +trainer/Advantage Score Max 1.06748 +trainer/Advantage Score Min -2.39075 +trainer/V1 Predictions Mean -72.8074 +trainer/V1 Predictions Std 14.83 +trainer/V1 Predictions Max -5.28651 +trainer/V1 Predictions Min -86.4394 +trainer/VF Loss 0.0509843 +expl/num steps total 158000 +expl/num paths total 162 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0768951 +expl/Actions Std 0.818249 +expl/Actions Max 2.71598 +expl/Actions Min -2.1258 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 155224 +eval/num paths total 158 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.158389 +eval/Actions Std 0.570316 +eval/Actions Max 0.999721 +eval/Actions Min -0.998909 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.77068e-06 +time/evaluation sampling (s) 2.79534 +time/exploration sampling (s) 3.73981 +time/logging (s) 0.00732021 +time/saving (s) 0.0157589 +time/training (s) 13.869 +time/epoch (s) 20.4273 +time/total (s) 3448.86 +Epoch -843 +------------------------------ ---------------- +2022-05-15 19:00:15.579972 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -842 finished +------------------------------ ---------------- +epoch -842 +replay_buffer/size 999047 +trainer/num train calls 159000 +trainer/QF1 Loss 1.32934 +trainer/QF2 Loss 1.54206 +trainer/Policy Loss 18.1562 +trainer/Q1 Predictions Mean -70.2576 +trainer/Q1 Predictions Std 19.4297 +trainer/Q1 Predictions Max -1.52724 +trainer/Q1 Predictions Min -86.2871 +trainer/Q2 Predictions Mean -70.2522 +trainer/Q2 Predictions Std 19.3512 +trainer/Q2 Predictions Max -1.52827 +trainer/Q2 Predictions Min -86.1203 +trainer/Q Targets Mean -69.9246 +trainer/Q Targets Std 19.9823 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5523 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0103344 +trainer/policy/mean Std 0.68707 +trainer/policy/mean Max 0.999206 +trainer/policy/mean Min -0.993516 +trainer/policy/std Mean 0.435812 +trainer/policy/std Std 0.0201172 +trainer/policy/std Max 0.455169 +trainer/policy/std Min 0.404715 +trainer/Advantage Weights Mean 2.61657 +trainer/Advantage Weights Std 12.3345 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.86079e-23 +trainer/Advantage Score Mean -0.406353 +trainer/Advantage Score Std 0.540988 +trainer/Advantage Score Max 0.754925 +trainer/Advantage Score Min -5.19084 +trainer/V1 Predictions Mean -69.7109 +trainer/V1 Predictions Std 19.9526 +trainer/V1 Predictions Max -0.236146 +trainer/V1 Predictions Min -86.6682 +trainer/VF Loss 0.053485 +expl/num steps total 159000 +expl/num paths total 163 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.044726 +expl/Actions Std 0.800784 +expl/Actions Max 2.45931 +expl/Actions Min -2.27655 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 156013 +eval/num paths total 159 +eval/path length Mean 789 +eval/path length Std 0 +eval/path length Max 789 +eval/path length Min 789 +eval/Rewards Mean 0.00126743 +eval/Rewards Std 0.0355784 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0236697 +eval/Actions Std 0.738563 +eval/Actions Max 0.998822 +eval/Actions Min -0.997891 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.80514e-06 +time/evaluation sampling (s) 2.86076 +time/exploration sampling (s) 3.77207 +time/logging (s) 0.00637899 +time/saving (s) 0.010215 +time/training (s) 14.1233 +time/epoch (s) 20.7727 +time/total (s) 3469.63 +Epoch -842 +------------------------------ ---------------- +2022-05-15 19:00:36.696781 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -841 finished +------------------------------ ---------------- +epoch -841 +replay_buffer/size 999047 +trainer/num train calls 160000 +trainer/QF1 Loss 1.29363 +trainer/QF2 Loss 1.30659 +trainer/Policy Loss 22.3142 +trainer/Q1 Predictions Mean -71.0409 +trainer/Q1 Predictions Std 18.393 +trainer/Q1 Predictions Max -0.27244 +trainer/Q1 Predictions Min -86.9829 +trainer/Q2 Predictions Mean -70.996 +trainer/Q2 Predictions Std 18.4336 +trainer/Q2 Predictions Max -0.294878 +trainer/Q2 Predictions Min -86.4736 +trainer/Q Targets Mean -71.6889 +trainer/Q Targets Std 18.3496 +trainer/Q Targets Max -0.637192 +trainer/Q Targets Min -87.0565 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -5.90784e-05 +trainer/policy/mean Std 0.699975 +trainer/policy/mean Max 0.999347 +trainer/policy/mean Min -0.997128 +trainer/policy/std Mean 0.437859 +trainer/policy/std Std 0.0210984 +trainer/policy/std Max 0.460091 +trainer/policy/std Min 0.403453 +trainer/Advantage Weights Mean 5.159 +trainer/Advantage Weights Std 19.1081 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.30162e-12 +trainer/Advantage Score Mean -0.277576 +trainer/Advantage Score Std 0.437032 +trainer/Advantage Score Max 1.49646 +trainer/Advantage Score Min -2.57902 +trainer/V1 Predictions Mean -71.4526 +trainer/V1 Predictions Std 18.5066 +trainer/V1 Predictions Max 0.748487 +trainer/V1 Predictions Min -87.2091 +trainer/VF Loss 0.0480895 +expl/num steps total 160000 +expl/num paths total 164 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0337917 +expl/Actions Std 0.820698 +expl/Actions Max 2.37668 +expl/Actions Min -2.28542 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 157013 +eval/num paths total 160 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.114353 +eval/Actions Std 0.597478 +eval/Actions Max 0.999028 +eval/Actions Min -0.997529 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.80002e-06 +time/evaluation sampling (s) 3.22964 +time/exploration sampling (s) 3.6771 +time/logging (s) 0.00896527 +time/saving (s) 0.0148713 +time/training (s) 14.182 +time/epoch (s) 21.1126 +time/total (s) 3490.75 +Epoch -841 +------------------------------ ---------------- +2022-05-15 19:00:57.756144 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -840 finished +------------------------------ ---------------- +epoch -840 +replay_buffer/size 999047 +trainer/num train calls 161000 +trainer/QF1 Loss 0.940931 +trainer/QF2 Loss 0.961605 +trainer/Policy Loss 6.70212 +trainer/Q1 Predictions Mean -70.2229 +trainer/Q1 Predictions Std 20.8285 +trainer/Q1 Predictions Max -1.00045 +trainer/Q1 Predictions Min -86.6385 +trainer/Q2 Predictions Mean -70.2409 +trainer/Q2 Predictions Std 20.8441 +trainer/Q2 Predictions Max -1.00013 +trainer/Q2 Predictions Min -86.9615 +trainer/Q Targets Mean -69.7929 +trainer/Q Targets Std 20.678 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2369 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00537579 +trainer/policy/mean Std 0.705493 +trainer/policy/mean Max 0.998858 +trainer/policy/mean Min -0.99847 +trainer/policy/std Mean 0.43715 +trainer/policy/std Std 0.0206712 +trainer/policy/std Max 0.459964 +trainer/policy/std Min 0.40211 +trainer/Advantage Weights Mean 1.05563 +trainer/Advantage Weights Std 8.84006 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.46602e-10 +trainer/Advantage Score Mean -0.539718 +trainer/Advantage Score Std 0.398916 +trainer/Advantage Score Max 0.663987 +trainer/Advantage Score Min -2.26433 +trainer/V1 Predictions Mean -69.4606 +trainer/V1 Predictions Std 20.7765 +trainer/V1 Predictions Max -0.854353 +trainer/V1 Predictions Min -86.2338 +trainer/VF Loss 0.0483155 +expl/num steps total 161000 +expl/num paths total 166 +expl/path length Mean 500 +expl/path length Std 401 +expl/path length Max 901 +expl/path length Min 99 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0279551 +expl/Actions Std 0.830443 +expl/Actions Max 2.39603 +expl/Actions Min -2.22589 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 157745 +eval/num paths total 161 +eval/path length Mean 732 +eval/path length Std 0 +eval/path length Max 732 +eval/path length Min 732 +eval/Rewards Mean 0.00136612 +eval/Rewards Std 0.0369358 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0427099 +eval/Actions Std 0.729489 +eval/Actions Max 0.999557 +eval/Actions Min -0.997999 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.70689e-06 +time/evaluation sampling (s) 3.21396 +time/exploration sampling (s) 3.83067 +time/logging (s) 0.00681293 +time/saving (s) 0.0116815 +time/training (s) 13.9857 +time/epoch (s) 21.0488 +time/total (s) 3511.81 +Epoch -840 +------------------------------ ---------------- +2022-05-15 19:01:17.996520 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -839 finished +------------------------------ ---------------- +epoch -839 +replay_buffer/size 999047 +trainer/num train calls 162000 +trainer/QF1 Loss 0.753013 +trainer/QF2 Loss 0.710799 +trainer/Policy Loss 4.89303 +trainer/Q1 Predictions Mean -70.856 +trainer/Q1 Predictions Std 18.4019 +trainer/Q1 Predictions Max -1.09695 +trainer/Q1 Predictions Min -86.8793 +trainer/Q2 Predictions Mean -70.8213 +trainer/Q2 Predictions Std 18.3393 +trainer/Q2 Predictions Max -0.379202 +trainer/Q2 Predictions Min -86.8095 +trainer/Q Targets Mean -70.4343 +trainer/Q Targets Std 18.3278 +trainer/Q Targets Max -0.57926 +trainer/Q Targets Min -86.2833 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0317002 +trainer/policy/mean Std 0.706857 +trainer/policy/mean Max 0.997706 +trainer/policy/mean Min -0.998454 +trainer/policy/std Mean 0.439562 +trainer/policy/std Std 0.0219235 +trainer/policy/std Max 0.46369 +trainer/policy/std Min 0.403065 +trainer/Advantage Weights Mean 0.782224 +trainer/Advantage Weights Std 6.88514 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.41554e-15 +trainer/Advantage Score Mean -0.562607 +trainer/Advantage Score Std 0.502254 +trainer/Advantage Score Max 0.801682 +trainer/Advantage Score Min -3.30536 +trainer/V1 Predictions Mean -70.1129 +trainer/V1 Predictions Std 18.5495 +trainer/V1 Predictions Max -0.00929973 +trainer/V1 Predictions Min -86.1636 +trainer/VF Loss 0.0599272 +expl/num steps total 162000 +expl/num paths total 167 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0625478 +expl/Actions Std 0.811043 +expl/Actions Max 2.31522 +expl/Actions Min -2.30775 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 158745 +eval/num paths total 162 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.120013 +eval/Actions Std 0.652257 +eval/Actions Max 0.998027 +eval/Actions Min -0.997823 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.14321e-06 +time/evaluation sampling (s) 3.07059 +time/exploration sampling (s) 3.2702 +time/logging (s) 0.00711524 +time/saving (s) 0.00996579 +time/training (s) 13.8759 +time/epoch (s) 20.2338 +time/total (s) 3532.04 +Epoch -839 +------------------------------ ---------------- +2022-05-15 19:01:37.843403 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -838 finished +------------------------------ ---------------- +epoch -838 +replay_buffer/size 999047 +trainer/num train calls 163000 +trainer/QF1 Loss 1.09712 +trainer/QF2 Loss 1.17133 +trainer/Policy Loss 5.65984 +trainer/Q1 Predictions Mean -71.1599 +trainer/Q1 Predictions Std 18.2475 +trainer/Q1 Predictions Max -0.648436 +trainer/Q1 Predictions Min -86.5143 +trainer/Q2 Predictions Mean -71.1783 +trainer/Q2 Predictions Std 18.2421 +trainer/Q2 Predictions Max -0.648545 +trainer/Q2 Predictions Min -86.9302 +trainer/Q Targets Mean -70.757 +trainer/Q Targets Std 18.3135 +trainer/Q Targets Max 0.172135 +trainer/Q Targets Min -86.2255 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00141511 +trainer/policy/mean Std 0.707935 +trainer/policy/mean Max 0.999114 +trainer/policy/mean Min -0.997703 +trainer/policy/std Mean 0.436374 +trainer/policy/std Std 0.0219582 +trainer/policy/std Max 0.460608 +trainer/policy/std Min 0.399214 +trainer/Advantage Weights Mean 0.944434 +trainer/Advantage Weights Std 8.86182 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.96535e-17 +trainer/Advantage Score Mean -0.750601 +trainer/Advantage Score Std 0.629009 +trainer/Advantage Score Max 0.673496 +trainer/Advantage Score Min -3.7358 +trainer/V1 Predictions Mean -70.5402 +trainer/V1 Predictions Std 18.4398 +trainer/V1 Predictions Max 2.22535 +trainer/V1 Predictions Min -86.1337 +trainer/VF Loss 0.0986801 +expl/num steps total 163000 +expl/num paths total 168 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0833308 +expl/Actions Std 0.825065 +expl/Actions Max 2.50511 +expl/Actions Min -2.5312 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 159745 +eval/num paths total 163 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.112696 +eval/Actions Std 0.690957 +eval/Actions Max 0.99929 +eval/Actions Min -0.998795 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.7488e-06 +time/evaluation sampling (s) 2.85651 +time/exploration sampling (s) 3.30267 +time/logging (s) 0.00682668 +time/saving (s) 0.013435 +time/training (s) 13.66 +time/epoch (s) 19.8394 +time/total (s) 3551.89 +Epoch -838 +------------------------------ ---------------- +2022-05-15 19:01:58.077990 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -837 finished +------------------------------ ---------------- +epoch -837 +replay_buffer/size 999047 +trainer/num train calls 164000 +trainer/QF1 Loss 1.32877 +trainer/QF2 Loss 1.44559 +trainer/Policy Loss 0.174949 +trainer/Q1 Predictions Mean -71.9149 +trainer/Q1 Predictions Std 16.9787 +trainer/Q1 Predictions Max -2.30541 +trainer/Q1 Predictions Min -85.7621 +trainer/Q2 Predictions Mean -71.9997 +trainer/Q2 Predictions Std 16.9591 +trainer/Q2 Predictions Max -2.31758 +trainer/Q2 Predictions Min -85.7324 +trainer/Q Targets Mean -71.5422 +trainer/Q Targets Std 17.2494 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.6127 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0282629 +trainer/policy/mean Std 0.703051 +trainer/policy/mean Max 0.996645 +trainer/policy/mean Min -0.999395 +trainer/policy/std Mean 0.436644 +trainer/policy/std Std 0.0237647 +trainer/policy/std Max 0.463652 +trainer/policy/std Min 0.398816 +trainer/Advantage Weights Mean 0.0400387 +trainer/Advantage Weights Std 0.231684 +trainer/Advantage Weights Max 3.41292 +trainer/Advantage Weights Min 1.3727e-15 +trainer/Advantage Score Mean -0.779472 +trainer/Advantage Score Std 0.506535 +trainer/Advantage Score Max 0.122757 +trainer/Advantage Score Min -3.4222 +trainer/V1 Predictions Mean -71.2304 +trainer/V1 Predictions Std 17.2426 +trainer/V1 Predictions Max -1.57654 +trainer/V1 Predictions Min -85.5249 +trainer/VF Loss 0.0864629 +expl/num steps total 164000 +expl/num paths total 169 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0267957 +expl/Actions Std 0.813086 +expl/Actions Max 2.20388 +expl/Actions Min -2.41465 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 160745 +eval/num paths total 164 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.125043 +eval/Actions Std 0.662925 +eval/Actions Max 0.998093 +eval/Actions Min -0.997785 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.71294e-06 +time/evaluation sampling (s) 2.76188 +time/exploration sampling (s) 3.56103 +time/logging (s) 0.0086027 +time/saving (s) 0.0110302 +time/training (s) 13.8882 +time/epoch (s) 20.2307 +time/total (s) 3572.12 +Epoch -837 +------------------------------ ---------------- +2022-05-15 19:02:18.249290 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -836 finished +------------------------------ ---------------- +epoch -836 +replay_buffer/size 999047 +trainer/num train calls 165000 +trainer/QF1 Loss 1.36668 +trainer/QF2 Loss 1.49823 +trainer/Policy Loss 11.6203 +trainer/Q1 Predictions Mean -70.8024 +trainer/Q1 Predictions Std 17.6333 +trainer/Q1 Predictions Max -3.68658 +trainer/Q1 Predictions Min -86.7605 +trainer/Q2 Predictions Mean -70.8926 +trainer/Q2 Predictions Std 17.6659 +trainer/Q2 Predictions Max -3.82175 +trainer/Q2 Predictions Min -87.342 +trainer/Q Targets Mean -70.2186 +trainer/Q Targets Std 17.8361 +trainer/Q Targets Max -3.9764 +trainer/Q Targets Min -87.3686 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000399609 +trainer/policy/mean Std 0.70549 +trainer/policy/mean Max 0.997823 +trainer/policy/mean Min -0.997997 +trainer/policy/std Mean 0.436826 +trainer/policy/std Std 0.0209093 +trainer/policy/std Max 0.461479 +trainer/policy/std Min 0.401195 +trainer/Advantage Weights Mean 2.27936 +trainer/Advantage Weights Std 12.6349 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.92803e-14 +trainer/Advantage Score Mean -0.489952 +trainer/Advantage Score Std 0.588516 +trainer/Advantage Score Max 0.6325 +trainer/Advantage Score Min -3.11619 +trainer/V1 Predictions Mean -69.8981 +trainer/V1 Predictions Std 18.1249 +trainer/V1 Predictions Max -2.21932 +trainer/V1 Predictions Min -87.4053 +trainer/VF Loss 0.0643364 +expl/num steps total 165000 +expl/num paths total 170 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0135174 +expl/Actions Std 0.902731 +expl/Actions Max 2.61299 +expl/Actions Min -2.57861 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 161745 +eval/num paths total 165 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.273162 +eval/Actions Std 0.730737 +eval/Actions Max 0.999248 +eval/Actions Min -0.997705 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.65939e-06 +time/evaluation sampling (s) 2.96696 +time/exploration sampling (s) 3.95841 +time/logging (s) 0.00864416 +time/saving (s) 0.0140107 +time/training (s) 13.2173 +time/epoch (s) 20.1654 +time/total (s) 3592.29 +Epoch -836 +------------------------------ ---------------- +2022-05-15 19:02:38.809022 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -835 finished +------------------------------ ---------------- +epoch -835 +replay_buffer/size 999047 +trainer/num train calls 166000 +trainer/QF1 Loss 19.5439 +trainer/QF2 Loss 19.8083 +trainer/Policy Loss 4.43996 +trainer/Q1 Predictions Mean -71.3577 +trainer/Q1 Predictions Std 19.1337 +trainer/Q1 Predictions Max -1.84618 +trainer/Q1 Predictions Min -87.4508 +trainer/Q2 Predictions Mean -71.3867 +trainer/Q2 Predictions Std 19.1212 +trainer/Q2 Predictions Max -1.75376 +trainer/Q2 Predictions Min -87.4605 +trainer/Q Targets Mean -71.2471 +trainer/Q Targets Std 18.9254 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7488 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0159672 +trainer/policy/mean Std 0.695857 +trainer/policy/mean Max 0.999535 +trainer/policy/mean Min -0.998234 +trainer/policy/std Mean 0.436822 +trainer/policy/std Std 0.020963 +trainer/policy/std Max 0.458979 +trainer/policy/std Min 0.398593 +trainer/Advantage Weights Mean 1.55072 +trainer/Advantage Weights Std 11.0043 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.33302e-12 +trainer/Advantage Score Mean -0.558303 +trainer/Advantage Score Std 0.507325 +trainer/Advantage Score Max 0.879517 +trainer/Advantage Score Min -2.73436 +trainer/V1 Predictions Mean -70.7437 +trainer/V1 Predictions Std 19.3721 +trainer/V1 Predictions Max -0.1411 +trainer/V1 Predictions Min -86.6222 +trainer/VF Loss 0.062425 +expl/num steps total 166000 +expl/num paths total 171 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.102195 +expl/Actions Std 0.869155 +expl/Actions Max 2.33097 +expl/Actions Min -2.19121 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 162745 +eval/num paths total 166 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.233114 +eval/Actions Std 0.705426 +eval/Actions Max 0.999503 +eval/Actions Min -0.9959 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.76975e-06 +time/evaluation sampling (s) 3.22925 +time/exploration sampling (s) 3.64038 +time/logging (s) 0.0105351 +time/saving (s) 0.0115782 +time/training (s) 13.6627 +time/epoch (s) 20.5544 +time/total (s) 3612.85 +Epoch -835 +------------------------------ ---------------- +2022-05-15 19:02:59.886889 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -834 finished +------------------------------ ---------------- +epoch -834 +replay_buffer/size 999047 +trainer/num train calls 167000 +trainer/QF1 Loss 0.78446 +trainer/QF2 Loss 0.746782 +trainer/Policy Loss 24.245 +trainer/Q1 Predictions Mean -72.3336 +trainer/Q1 Predictions Std 17.016 +trainer/Q1 Predictions Max -0.759717 +trainer/Q1 Predictions Min -87.6959 +trainer/Q2 Predictions Mean -72.2377 +trainer/Q2 Predictions Std 17.0579 +trainer/Q2 Predictions Max -0.539817 +trainer/Q2 Predictions Min -87.201 +trainer/Q Targets Mean -72.4904 +trainer/Q Targets Std 17.1297 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4442 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0410564 +trainer/policy/mean Std 0.700763 +trainer/policy/mean Max 0.999238 +trainer/policy/mean Min -0.995191 +trainer/policy/std Mean 0.437812 +trainer/policy/std Std 0.0217479 +trainer/policy/std Max 0.45922 +trainer/policy/std Min 0.398441 +trainer/Advantage Weights Mean 4.85278 +trainer/Advantage Weights Std 18.052 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.94311e-18 +trainer/Advantage Score Mean -0.286291 +trainer/Advantage Score Std 0.52769 +trainer/Advantage Score Max 1.39037 +trainer/Advantage Score Min -4.07822 +trainer/V1 Predictions Mean -72.2288 +trainer/V1 Predictions Std 17.2894 +trainer/V1 Predictions Max 0.185884 +trainer/V1 Predictions Min -85.936 +trainer/VF Loss 0.0560729 +expl/num steps total 167000 +expl/num paths total 172 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00635891 +expl/Actions Std 0.819921 +expl/Actions Max 2.44691 +expl/Actions Min -2.76639 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 163745 +eval/num paths total 167 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.127929 +eval/Actions Std 0.514008 +eval/Actions Max 0.998187 +eval/Actions Min -0.998598 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.96906e-06 +time/evaluation sampling (s) 3.18506 +time/exploration sampling (s) 3.78303 +time/logging (s) 0.00703927 +time/saving (s) 0.00946437 +time/training (s) 14.0812 +time/epoch (s) 21.0658 +time/total (s) 3633.92 +Epoch -834 +------------------------------ ---------------- +2022-05-15 19:03:20.974541 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -833 finished +------------------------------ ---------------- +epoch -833 +replay_buffer/size 999047 +trainer/num train calls 168000 +trainer/QF1 Loss 2.53749 +trainer/QF2 Loss 2.38745 +trainer/Policy Loss 70.848 +trainer/Q1 Predictions Mean -71.3813 +trainer/Q1 Predictions Std 17.0647 +trainer/Q1 Predictions Max -0.347236 +trainer/Q1 Predictions Min -86.4015 +trainer/Q2 Predictions Mean -71.4305 +trainer/Q2 Predictions Std 16.9622 +trainer/Q2 Predictions Max -0.740045 +trainer/Q2 Predictions Min -86.241 +trainer/Q Targets Mean -71.9512 +trainer/Q Targets Std 16.2647 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.98 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.042338 +trainer/policy/mean Std 0.704195 +trainer/policy/mean Max 0.999165 +trainer/policy/mean Min -0.997113 +trainer/policy/std Mean 0.437805 +trainer/policy/std Std 0.0221696 +trainer/policy/std Max 0.459994 +trainer/policy/std Min 0.400538 +trainer/Advantage Weights Mean 13.4304 +trainer/Advantage Weights Std 32.469 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.22205e-14 +trainer/Advantage Score Mean -0.383293 +trainer/Advantage Score Std 0.87682 +trainer/Advantage Score Max 3.21509 +trainer/Advantage Score Min -3.14378 +trainer/V1 Predictions Mean -71.6137 +trainer/V1 Predictions Std 16.2537 +trainer/V1 Predictions Max -4.66131 +trainer/V1 Predictions Min -85.964 +trainer/VF Loss 0.253656 +expl/num steps total 168000 +expl/num paths total 173 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0496286 +expl/Actions Std 0.824284 +expl/Actions Max 2.98877 +expl/Actions Min -2.19469 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 164692 +eval/num paths total 168 +eval/path length Mean 947 +eval/path length Std 0 +eval/path length Max 947 +eval/path length Min 947 +eval/Rewards Mean 0.00105597 +eval/Rewards Std 0.0324785 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0432579 +eval/Actions Std 0.714031 +eval/Actions Max 0.999845 +eval/Actions Min -0.99731 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.85404e-06 +time/evaluation sampling (s) 3.15395 +time/exploration sampling (s) 3.52124 +time/logging (s) 0.0077141 +time/saving (s) 0.0109555 +time/training (s) 14.3873 +time/epoch (s) 21.0811 +time/total (s) 3655.01 +Epoch -833 +------------------------------ ---------------- +2022-05-15 19:03:41.764362 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -832 finished +------------------------------ ---------------- +epoch -832 +replay_buffer/size 999047 +trainer/num train calls 169000 +trainer/QF1 Loss 0.820677 +trainer/QF2 Loss 0.910356 +trainer/Policy Loss 42.6265 +trainer/Q1 Predictions Mean -71.568 +trainer/Q1 Predictions Std 17.7587 +trainer/Q1 Predictions Max -0.461553 +trainer/Q1 Predictions Min -86.1562 +trainer/Q2 Predictions Mean -71.5201 +trainer/Q2 Predictions Std 17.7344 +trainer/Q2 Predictions Max -0.621941 +trainer/Q2 Predictions Min -86.1766 +trainer/Q Targets Mean -72.0861 +trainer/Q Targets Std 17.7361 +trainer/Q Targets Max 0.49848 +trainer/Q Targets Min -86.6809 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00201039 +trainer/policy/mean Std 0.713554 +trainer/policy/mean Max 0.998228 +trainer/policy/mean Min -0.999292 +trainer/policy/std Mean 0.436031 +trainer/policy/std Std 0.0224748 +trainer/policy/std Max 0.463344 +trainer/policy/std Min 0.400778 +trainer/Advantage Weights Mean 7.22013 +trainer/Advantage Weights Std 21.5701 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.69921e-13 +trainer/Advantage Score Mean -0.242598 +trainer/Advantage Score Std 0.549693 +trainer/Advantage Score Max 1.20701 +trainer/Advantage Score Min -2.80316 +trainer/V1 Predictions Mean -71.7278 +trainer/V1 Predictions Std 17.9325 +trainer/V1 Predictions Max 1.65826 +trainer/V1 Predictions Min -86.4406 +trainer/VF Loss 0.0645479 +expl/num steps total 169000 +expl/num paths total 174 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00331212 +expl/Actions Std 0.806336 +expl/Actions Max 2.3077 +expl/Actions Min -2.29071 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 165692 +eval/num paths total 169 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.262838 +eval/Actions Std 0.705772 +eval/Actions Max 0.999468 +eval/Actions Min -0.996146 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.85963e-06 +time/evaluation sampling (s) 3.34251 +time/exploration sampling (s) 3.16059 +time/logging (s) 0.00953479 +time/saving (s) 0.0138145 +time/training (s) 14.2589 +time/epoch (s) 20.7854 +time/total (s) 3675.8 +Epoch -832 +------------------------------ ---------------- +2022-05-15 19:04:02.416542 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -831 finished +------------------------------ ---------------- +epoch -831 +replay_buffer/size 999047 +trainer/num train calls 170000 +trainer/QF1 Loss 0.600095 +trainer/QF2 Loss 0.659804 +trainer/Policy Loss 12.9128 +trainer/Q1 Predictions Mean -71.8081 +trainer/Q1 Predictions Std 17.6358 +trainer/Q1 Predictions Max -0.750586 +trainer/Q1 Predictions Min -85.6586 +trainer/Q2 Predictions Mean -71.7412 +trainer/Q2 Predictions Std 17.6894 +trainer/Q2 Predictions Max -0.806066 +trainer/Q2 Predictions Min -85.7239 +trainer/Q Targets Mean -72.1548 +trainer/Q Targets Std 17.6006 +trainer/Q Targets Max 1.03037 +trainer/Q Targets Min -86.3407 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00313269 +trainer/policy/mean Std 0.710813 +trainer/policy/mean Max 0.999204 +trainer/policy/mean Min -0.998624 +trainer/policy/std Mean 0.436236 +trainer/policy/std Std 0.0231517 +trainer/policy/std Max 0.46478 +trainer/policy/std Min 0.399 +trainer/Advantage Weights Mean 3.5967 +trainer/Advantage Weights Std 15.7739 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.56612e-10 +trainer/Advantage Score Mean -0.355262 +trainer/Advantage Score Std 0.490764 +trainer/Advantage Score Max 2.22973 +trainer/Advantage Score Min -2.20835 +trainer/V1 Predictions Mean -71.904 +trainer/V1 Predictions Std 17.6389 +trainer/V1 Predictions Max -1.54043 +trainer/V1 Predictions Min -86.2009 +trainer/VF Loss 0.0813589 +expl/num steps total 170000 +expl/num paths total 175 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00190232 +expl/Actions Std 0.813235 +expl/Actions Max 2.31213 +expl/Actions Min -2.25643 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 166692 +eval/num paths total 170 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.290513 +eval/Actions Std 0.634164 +eval/Actions Max 0.999801 +eval/Actions Min -0.997499 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.01496e-06 +time/evaluation sampling (s) 2.71671 +time/exploration sampling (s) 3.49247 +time/logging (s) 0.0124609 +time/saving (s) 0.0188479 +time/training (s) 14.4064 +time/epoch (s) 20.6469 +time/total (s) 3696.45 +Epoch -831 +------------------------------ ---------------- +2022-05-15 19:04:22.579515 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -830 finished +------------------------------ ---------------- +epoch -830 +replay_buffer/size 999047 +trainer/num train calls 171000 +trainer/QF1 Loss 0.518962 +trainer/QF2 Loss 0.582306 +trainer/Policy Loss 12.1158 +trainer/Q1 Predictions Mean -72.9734 +trainer/Q1 Predictions Std 15.7933 +trainer/Q1 Predictions Max -1.73525 +trainer/Q1 Predictions Min -85.8145 +trainer/Q2 Predictions Mean -73.0315 +trainer/Q2 Predictions Std 15.7284 +trainer/Q2 Predictions Max -1.51617 +trainer/Q2 Predictions Min -86.0906 +trainer/Q Targets Mean -72.842 +trainer/Q Targets Std 15.9037 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9713 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00690328 +trainer/policy/mean Std 0.69918 +trainer/policy/mean Max 0.996863 +trainer/policy/mean Min -0.9979 +trainer/policy/std Mean 0.435989 +trainer/policy/std Std 0.022894 +trainer/policy/std Max 0.463489 +trainer/policy/std Min 0.395291 +trainer/Advantage Weights Mean 2.70102 +trainer/Advantage Weights Std 13.1798 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.13386e-11 +trainer/Advantage Score Mean -0.358001 +trainer/Advantage Score Std 0.460264 +trainer/Advantage Score Max 0.958031 +trainer/Advantage Score Min -2.45705 +trainer/V1 Predictions Mean -72.5512 +trainer/V1 Predictions Std 16.0415 +trainer/V1 Predictions Max -1.16836 +trainer/V1 Predictions Min -86.3655 +trainer/VF Loss 0.0439323 +expl/num steps total 171000 +expl/num paths total 176 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.222662 +expl/Actions Std 0.793222 +expl/Actions Max 2.48381 +expl/Actions Min -2.20836 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 167692 +eval/num paths total 171 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0579787 +eval/Actions Std 0.720927 +eval/Actions Max 0.99966 +eval/Actions Min -0.998815 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.84612e-06 +time/evaluation sampling (s) 2.906 +time/exploration sampling (s) 3.66648 +time/logging (s) 0.00779339 +time/saving (s) 0.0116869 +time/training (s) 13.5559 +time/epoch (s) 20.1479 +time/total (s) 3716.61 +Epoch -830 +------------------------------ ---------------- +2022-05-15 19:04:43.149031 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -829 finished +------------------------------ ---------------- +epoch -829 +replay_buffer/size 999047 +trainer/num train calls 172000 +trainer/QF1 Loss 0.924811 +trainer/QF2 Loss 1.00732 +trainer/Policy Loss 12.9221 +trainer/Q1 Predictions Mean -71.4132 +trainer/Q1 Predictions Std 18.8269 +trainer/Q1 Predictions Max -1.14514 +trainer/Q1 Predictions Min -87.3634 +trainer/Q2 Predictions Mean -71.3677 +trainer/Q2 Predictions Std 18.8318 +trainer/Q2 Predictions Max -1.06753 +trainer/Q2 Predictions Min -86.8441 +trainer/Q Targets Mean -70.9567 +trainer/Q Targets Std 19.1346 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9764 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00264711 +trainer/policy/mean Std 0.719112 +trainer/policy/mean Max 0.999399 +trainer/policy/mean Min -0.996774 +trainer/policy/std Mean 0.435454 +trainer/policy/std Std 0.0222305 +trainer/policy/std Max 0.461719 +trainer/policy/std Min 0.397592 +trainer/Advantage Weights Mean 2.07626 +trainer/Advantage Weights Std 11.4686 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.2273e-20 +trainer/Advantage Score Mean -0.541783 +trainer/Advantage Score Std 0.62437 +trainer/Advantage Score Max 0.536575 +trainer/Advantage Score Min -4.39442 +trainer/V1 Predictions Mean -70.7286 +trainer/V1 Predictions Std 19.1658 +trainer/V1 Predictions Max 0.574235 +trainer/V1 Predictions Min -86.8528 +trainer/VF Loss 0.0729576 +expl/num steps total 172000 +expl/num paths total 177 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0311749 +expl/Actions Std 0.820683 +expl/Actions Max 2.41285 +expl/Actions Min -2.36506 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 168692 +eval/num paths total 172 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0418064 +eval/Actions Std 0.80044 +eval/Actions Max 0.99853 +eval/Actions Min -0.998428 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.94577e-06 +time/evaluation sampling (s) 3.13282 +time/exploration sampling (s) 3.66875 +time/logging (s) 0.00859941 +time/saving (s) 0.0131501 +time/training (s) 13.7397 +time/epoch (s) 20.563 +time/total (s) 3737.18 +Epoch -829 +------------------------------ ---------------- +2022-05-15 19:05:04.445007 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -828 finished +------------------------------ ---------------- +epoch -828 +replay_buffer/size 999047 +trainer/num train calls 173000 +trainer/QF1 Loss 0.986925 +trainer/QF2 Loss 1.12128 +trainer/Policy Loss 13.1543 +trainer/Q1 Predictions Mean -70.7583 +trainer/Q1 Predictions Std 19.6163 +trainer/Q1 Predictions Max -0.825997 +trainer/Q1 Predictions Min -87.1958 +trainer/Q2 Predictions Mean -70.798 +trainer/Q2 Predictions Std 19.6464 +trainer/Q2 Predictions Max -0.543094 +trainer/Q2 Predictions Min -87.0744 +trainer/Q Targets Mean -70.4752 +trainer/Q Targets Std 19.863 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2489 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00929316 +trainer/policy/mean Std 0.702661 +trainer/policy/mean Max 0.999367 +trainer/policy/mean Min -0.994962 +trainer/policy/std Mean 0.43566 +trainer/policy/std Std 0.021934 +trainer/policy/std Max 0.45947 +trainer/policy/std Min 0.399266 +trainer/Advantage Weights Mean 3.51913 +trainer/Advantage Weights Std 13.9129 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.84254e-12 +trainer/Advantage Score Mean -0.358743 +trainer/Advantage Score Std 0.491357 +trainer/Advantage Score Max 1.09095 +trainer/Advantage Score Min -2.58659 +trainer/V1 Predictions Mean -70.2852 +trainer/V1 Predictions Std 19.8705 +trainer/V1 Predictions Max 0.369885 +trainer/V1 Predictions Min -87.1495 +trainer/VF Loss 0.0493045 +expl/num steps total 173000 +expl/num paths total 178 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0701766 +expl/Actions Std 0.810899 +expl/Actions Max 2.68104 +expl/Actions Min -2.52808 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 169692 +eval/num paths total 173 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0637592 +eval/Actions Std 0.727878 +eval/Actions Max 0.999669 +eval/Actions Min -0.997037 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.97092e-06 +time/evaluation sampling (s) 2.89138 +time/exploration sampling (s) 3.73643 +time/logging (s) 0.00830177 +time/saving (s) 0.0114337 +time/training (s) 14.6401 +time/epoch (s) 21.2877 +time/total (s) 3758.47 +Epoch -828 +------------------------------ ---------------- +2022-05-15 19:05:25.417275 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -827 finished +------------------------------ ---------------- +epoch -827 +replay_buffer/size 999047 +trainer/num train calls 174000 +trainer/QF1 Loss 0.80127 +trainer/QF2 Loss 0.77383 +trainer/Policy Loss 8.40546 +trainer/Q1 Predictions Mean -70.7961 +trainer/Q1 Predictions Std 19.0759 +trainer/Q1 Predictions Max -1.05663 +trainer/Q1 Predictions Min -86.4605 +trainer/Q2 Predictions Mean -70.7966 +trainer/Q2 Predictions Std 19.0644 +trainer/Q2 Predictions Max -0.650852 +trainer/Q2 Predictions Min -85.9414 +trainer/Q Targets Mean -70.9815 +trainer/Q Targets Std 19.314 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.542 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00702694 +trainer/policy/mean Std 0.715525 +trainer/policy/mean Max 0.998513 +trainer/policy/mean Min -0.998032 +trainer/policy/std Mean 0.434697 +trainer/policy/std Std 0.0213721 +trainer/policy/std Max 0.459259 +trainer/policy/std Min 0.398871 +trainer/Advantage Weights Mean 2.04396 +trainer/Advantage Weights Std 9.90298 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.77741e-18 +trainer/Advantage Score Mean -0.373392 +trainer/Advantage Score Std 0.54845 +trainer/Advantage Score Max 0.466614 +trainer/Advantage Score Min -4.01175 +trainer/V1 Predictions Mean -70.6828 +trainer/V1 Predictions Std 19.3621 +trainer/V1 Predictions Max -0.110438 +trainer/V1 Predictions Min -86.1771 +trainer/VF Loss 0.0489772 +expl/num steps total 174000 +expl/num paths total 180 +expl/path length Mean 500 +expl/path length Std 350 +expl/path length Max 850 +expl/path length Min 150 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0304348 +expl/Actions Std 0.837549 +expl/Actions Max 2.50551 +expl/Actions Min -2.41323 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 170692 +eval/num paths total 174 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0668835 +eval/Actions Std 0.628816 +eval/Actions Max 0.999756 +eval/Actions Min -0.994192 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.11201e-06 +time/evaluation sampling (s) 3.12112 +time/exploration sampling (s) 3.59313 +time/logging (s) 0.0130251 +time/saving (s) 0.0196188 +time/training (s) 14.2236 +time/epoch (s) 20.9705 +time/total (s) 3779.44 +Epoch -827 +------------------------------ ---------------- +2022-05-15 19:05:45.308295 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -826 finished +------------------------------ ---------------- +epoch -826 +replay_buffer/size 999047 +trainer/num train calls 175000 +trainer/QF1 Loss 0.70214 +trainer/QF2 Loss 0.780218 +trainer/Policy Loss 14.9761 +trainer/Q1 Predictions Mean -70.9166 +trainer/Q1 Predictions Std 18.7769 +trainer/Q1 Predictions Max -0.500969 +trainer/Q1 Predictions Min -86.9522 +trainer/Q2 Predictions Mean -70.8084 +trainer/Q2 Predictions Std 18.6991 +trainer/Q2 Predictions Max -0.598373 +trainer/Q2 Predictions Min -86.5861 +trainer/Q Targets Mean -71.0968 +trainer/Q Targets Std 18.7256 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7043 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00288593 +trainer/policy/mean Std 0.695511 +trainer/policy/mean Max 0.999079 +trainer/policy/mean Min -0.994328 +trainer/policy/std Mean 0.434946 +trainer/policy/std Std 0.0221865 +trainer/policy/std Max 0.459511 +trainer/policy/std Min 0.399217 +trainer/Advantage Weights Mean 3.54772 +trainer/Advantage Weights Std 14.998 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.02913e-11 +trainer/Advantage Score Mean -0.335527 +trainer/Advantage Score Std 0.481444 +trainer/Advantage Score Max 0.940315 +trainer/Advantage Score Min -2.52997 +trainer/V1 Predictions Mean -70.7966 +trainer/V1 Predictions Std 18.8685 +trainer/V1 Predictions Max 1.05858 +trainer/V1 Predictions Min -86.6841 +trainer/VF Loss 0.0447138 +expl/num steps total 175000 +expl/num paths total 181 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0509778 +expl/Actions Std 0.827882 +expl/Actions Max 2.2834 +expl/Actions Min -2.42174 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 171692 +eval/num paths total 175 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.116571 +eval/Actions Std 0.63036 +eval/Actions Max 0.999473 +eval/Actions Min -0.997754 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.03274e-05 +time/evaluation sampling (s) 3.09135 +time/exploration sampling (s) 3.29687 +time/logging (s) 0.00700788 +time/saving (s) 0.0100043 +time/training (s) 13.4678 +time/epoch (s) 19.8731 +time/total (s) 3799.32 +Epoch -826 +------------------------------ ---------------- +2022-05-15 19:06:05.989846 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -825 finished +------------------------------ ---------------- +epoch -825 +replay_buffer/size 999047 +trainer/num train calls 176000 +trainer/QF1 Loss 0.557894 +trainer/QF2 Loss 0.594797 +trainer/Policy Loss 14.0096 +trainer/Q1 Predictions Mean -72.5871 +trainer/Q1 Predictions Std 17.0128 +trainer/Q1 Predictions Max -0.424668 +trainer/Q1 Predictions Min -86.3204 +trainer/Q2 Predictions Mean -72.6382 +trainer/Q2 Predictions Std 16.9864 +trainer/Q2 Predictions Max -1.0665 +trainer/Q2 Predictions Min -86.4203 +trainer/Q Targets Mean -72.6132 +trainer/Q Targets Std 16.9136 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9587 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00478886 +trainer/policy/mean Std 0.719215 +trainer/policy/mean Max 0.999197 +trainer/policy/mean Min -0.998905 +trainer/policy/std Mean 0.433583 +trainer/policy/std Std 0.022068 +trainer/policy/std Max 0.462066 +trainer/policy/std Min 0.396801 +trainer/Advantage Weights Mean 3.99868 +trainer/Advantage Weights Std 17.304 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.00079e-11 +trainer/Advantage Score Mean -0.365574 +trainer/Advantage Score Std 0.500268 +trainer/Advantage Score Max 1.43768 +trainer/Advantage Score Min -2.53276 +trainer/V1 Predictions Mean -72.3204 +trainer/V1 Predictions Std 17.1148 +trainer/V1 Predictions Max -1.69092 +trainer/V1 Predictions Min -85.8189 +trainer/VF Loss 0.0565319 +expl/num steps total 176000 +expl/num paths total 182 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0388401 +expl/Actions Std 0.825429 +expl/Actions Max 2.38744 +expl/Actions Min -2.32599 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 172692 +eval/num paths total 176 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.149889 +eval/Actions Std 0.737834 +eval/Actions Max 0.998586 +eval/Actions Min -0.998672 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.83774e-06 +time/evaluation sampling (s) 3.19082 +time/exploration sampling (s) 3.00604 +time/logging (s) 0.00884567 +time/saving (s) 0.0127811 +time/training (s) 14.4588 +time/epoch (s) 20.6772 +time/total (s) 3820.01 +Epoch -825 +------------------------------ ---------------- +2022-05-15 19:06:26.130970 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -824 finished +------------------------------ ---------------- +epoch -824 +replay_buffer/size 999047 +trainer/num train calls 177000 +trainer/QF1 Loss 3.35146 +trainer/QF2 Loss 3.57507 +trainer/Policy Loss 4.66286 +trainer/Q1 Predictions Mean -71.7552 +trainer/Q1 Predictions Std 19.293 +trainer/Q1 Predictions Max -0.687107 +trainer/Q1 Predictions Min -86.9103 +trainer/Q2 Predictions Mean -71.8295 +trainer/Q2 Predictions Std 19.3425 +trainer/Q2 Predictions Max -0.519517 +trainer/Q2 Predictions Min -86.76 +trainer/Q Targets Mean -71.2484 +trainer/Q Targets Std 19.9128 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.39 +trainer/rewards Mean -0.976562 +trainer/rewards Std 0.151288 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0234375 +trainer/terminals Std 0.151288 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00495361 +trainer/policy/mean Std 0.707916 +trainer/policy/mean Max 0.998354 +trainer/policy/mean Min -0.999473 +trainer/policy/std Mean 0.431243 +trainer/policy/std Std 0.0218424 +trainer/policy/std Max 0.457754 +trainer/policy/std Min 0.396349 +trainer/Advantage Weights Mean 1.51196 +trainer/Advantage Weights Std 10.8039 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.25907e-16 +trainer/Advantage Score Mean -0.465417 +trainer/Advantage Score Std 0.494232 +trainer/Advantage Score Max 1.23268 +trainer/Advantage Score Min -3.51814 +trainer/V1 Predictions Mean -71.0798 +trainer/V1 Predictions Std 19.5815 +trainer/V1 Predictions Max -0.433893 +trainer/V1 Predictions Min -86.122 +trainer/VF Loss 0.0535216 +expl/num steps total 177000 +expl/num paths total 183 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.216698 +expl/Actions Std 0.820608 +expl/Actions Max 2.45485 +expl/Actions Min -2.48546 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 173692 +eval/num paths total 177 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0385275 +eval/Actions Std 0.733736 +eval/Actions Max 0.998208 +eval/Actions Min -0.99824 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.66498e-06 +time/evaluation sampling (s) 2.69106 +time/exploration sampling (s) 3.41225 +time/logging (s) 0.00836802 +time/saving (s) 0.0118509 +time/training (s) 14.0092 +time/epoch (s) 20.1327 +time/total (s) 3840.14 +Epoch -824 +------------------------------ ---------------- +2022-05-15 19:06:46.532602 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -823 finished +------------------------------ ---------------- +epoch -823 +replay_buffer/size 999047 +trainer/num train calls 178000 +trainer/QF1 Loss 0.827094 +trainer/QF2 Loss 0.833087 +trainer/Policy Loss 5.87903 +trainer/Q1 Predictions Mean -70.1553 +trainer/Q1 Predictions Std 19.624 +trainer/Q1 Predictions Max -1.55469 +trainer/Q1 Predictions Min -85.9795 +trainer/Q2 Predictions Mean -70.151 +trainer/Q2 Predictions Std 19.6607 +trainer/Q2 Predictions Max -1.01065 +trainer/Q2 Predictions Min -85.8306 +trainer/Q Targets Mean -69.6296 +trainer/Q Targets Std 19.937 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0956 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0430224 +trainer/policy/mean Std 0.705843 +trainer/policy/mean Max 0.99914 +trainer/policy/mean Min -0.998759 +trainer/policy/std Mean 0.432589 +trainer/policy/std Std 0.0215062 +trainer/policy/std Max 0.4601 +trainer/policy/std Min 0.399376 +trainer/Advantage Weights Mean 1.51936 +trainer/Advantage Weights Std 10.8546 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.717e-19 +trainer/Advantage Score Mean -0.645233 +trainer/Advantage Score Std 0.592813 +trainer/Advantage Score Max 1.8566 +trainer/Advantage Score Min -4.20057 +trainer/V1 Predictions Mean -69.3701 +trainer/V1 Predictions Std 19.9431 +trainer/V1 Predictions Max 0.914689 +trainer/V1 Predictions Min -85.8143 +trainer/VF Loss 0.0942143 +expl/num steps total 178000 +expl/num paths total 184 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0385304 +expl/Actions Std 0.83055 +expl/Actions Max 2.47356 +expl/Actions Min -2.48017 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 174692 +eval/num paths total 178 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0179443 +eval/Actions Std 0.704387 +eval/Actions Max 0.998781 +eval/Actions Min -0.999104 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.92389e-06 +time/evaluation sampling (s) 2.89851 +time/exploration sampling (s) 3.66351 +time/logging (s) 0.00836953 +time/saving (s) 0.0139853 +time/training (s) 13.8115 +time/epoch (s) 20.3959 +time/total (s) 3860.54 +Epoch -823 +------------------------------ ---------------- +2022-05-15 19:07:07.124552 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -822 finished +------------------------------ ---------------- +epoch -822 +replay_buffer/size 999047 +trainer/num train calls 179000 +trainer/QF1 Loss 0.775222 +trainer/QF2 Loss 0.798543 +trainer/Policy Loss 71.5158 +trainer/Q1 Predictions Mean -72.3072 +trainer/Q1 Predictions Std 16.4455 +trainer/Q1 Predictions Max -1.26759 +trainer/Q1 Predictions Min -85.8251 +trainer/Q2 Predictions Mean -72.3144 +trainer/Q2 Predictions Std 16.4337 +trainer/Q2 Predictions Max -1.04771 +trainer/Q2 Predictions Min -85.94 +trainer/Q Targets Mean -72.7321 +trainer/Q Targets Std 16.6257 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2013 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0103453 +trainer/policy/mean Std 0.700037 +trainer/policy/mean Max 0.997361 +trainer/policy/mean Min -0.999601 +trainer/policy/std Mean 0.433488 +trainer/policy/std Std 0.0217909 +trainer/policy/std Max 0.459464 +trainer/policy/std Min 0.397811 +trainer/Advantage Weights Mean 15.4436 +trainer/Advantage Weights Std 30.4404 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.07191e-10 +trainer/Advantage Score Mean -0.0756961 +trainer/Advantage Score Std 0.496528 +trainer/Advantage Score Max 0.91877 +trainer/Advantage Score Min -2.29564 +trainer/V1 Predictions Mean -72.4574 +trainer/V1 Predictions Std 16.749 +trainer/V1 Predictions Max 0.578228 +trainer/V1 Predictions Min -86.0778 +trainer/VF Loss 0.0706805 +expl/num steps total 179000 +expl/num paths total 185 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.073919 +expl/Actions Std 0.809786 +expl/Actions Max 2.30973 +expl/Actions Min -2.51281 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 175692 +eval/num paths total 179 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.371269 +eval/Actions Std 0.691215 +eval/Actions Max 0.999054 +eval/Actions Min -0.99932 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.31997e-06 +time/evaluation sampling (s) 3.09624 +time/exploration sampling (s) 3.5167 +time/logging (s) 0.00839315 +time/saving (s) 0.0125152 +time/training (s) 13.9498 +time/epoch (s) 20.5836 +time/total (s) 3881.13 +Epoch -822 +------------------------------ ---------------- +2022-05-15 19:07:27.499763 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -821 finished +------------------------------ ---------------- +epoch -821 +replay_buffer/size 999047 +trainer/num train calls 180000 +trainer/QF1 Loss 0.73347 +trainer/QF2 Loss 0.617324 +trainer/Policy Loss 9.63398 +trainer/Q1 Predictions Mean -72.0423 +trainer/Q1 Predictions Std 17.6962 +trainer/Q1 Predictions Max -0.472316 +trainer/Q1 Predictions Min -86.4588 +trainer/Q2 Predictions Mean -71.9989 +trainer/Q2 Predictions Std 17.7135 +trainer/Q2 Predictions Max -0.622882 +trainer/Q2 Predictions Min -86.368 +trainer/Q Targets Mean -71.7061 +trainer/Q Targets Std 17.5648 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2986 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00981932 +trainer/policy/mean Std 0.712067 +trainer/policy/mean Max 0.999597 +trainer/policy/mean Min -0.998962 +trainer/policy/std Mean 0.434687 +trainer/policy/std Std 0.0213751 +trainer/policy/std Max 0.45731 +trainer/policy/std Min 0.396729 +trainer/Advantage Weights Mean 3.12204 +trainer/Advantage Weights Std 15.4265 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.20617e-25 +trainer/Advantage Score Mean -0.455702 +trainer/Advantage Score Std 0.574972 +trainer/Advantage Score Max 0.855013 +trainer/Advantage Score Min -5.53448 +trainer/V1 Predictions Mean -71.4666 +trainer/V1 Predictions Std 17.621 +trainer/V1 Predictions Max -0.277479 +trainer/V1 Predictions Min -85.9967 +trainer/VF Loss 0.0644302 +expl/num steps total 180000 +expl/num paths total 186 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.205635 +expl/Actions Std 0.878159 +expl/Actions Max 2.56732 +expl/Actions Min -2.28898 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 176692 +eval/num paths total 180 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0384034 +eval/Actions Std 0.707252 +eval/Actions Max 0.999421 +eval/Actions Min -0.998895 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.61096e-06 +time/evaluation sampling (s) 3.06071 +time/exploration sampling (s) 3.66183 +time/logging (s) 0.00679291 +time/saving (s) 0.0105755 +time/training (s) 13.6273 +time/epoch (s) 20.3673 +time/total (s) 3901.5 +Epoch -821 +------------------------------ ---------------- +2022-05-15 19:07:47.624143 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -820 finished +------------------------------ ---------------- +epoch -820 +replay_buffer/size 999047 +trainer/num train calls 181000 +trainer/QF1 Loss 0.693397 +trainer/QF2 Loss 0.753527 +trainer/Policy Loss 8.06958 +trainer/Q1 Predictions Mean -73.6906 +trainer/Q1 Predictions Std 16.8657 +trainer/Q1 Predictions Max -2.18171 +trainer/Q1 Predictions Min -86.5637 +trainer/Q2 Predictions Mean -73.667 +trainer/Q2 Predictions Std 16.8115 +trainer/Q2 Predictions Max -2.96669 +trainer/Q2 Predictions Min -86.4517 +trainer/Q Targets Mean -73.3916 +trainer/Q Targets Std 16.7764 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6754 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0119661 +trainer/policy/mean Std 0.704922 +trainer/policy/mean Max 0.997312 +trainer/policy/mean Min -0.997874 +trainer/policy/std Mean 0.434882 +trainer/policy/std Std 0.0213735 +trainer/policy/std Max 0.457277 +trainer/policy/std Min 0.399241 +trainer/Advantage Weights Mean 1.95749 +trainer/Advantage Weights Std 9.18646 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.70986e-17 +trainer/Advantage Score Mean -0.409162 +trainer/Advantage Score Std 0.590157 +trainer/Advantage Score Max 0.922379 +trainer/Advantage Score Min -3.74018 +trainer/V1 Predictions Mean -73.0336 +trainer/V1 Predictions Std 17.1409 +trainer/V1 Predictions Max -1.74018 +trainer/V1 Predictions Min -86.6468 +trainer/VF Loss 0.0585637 +expl/num steps total 181000 +expl/num paths total 188 +expl/path length Mean 500 +expl/path length Std 335 +expl/path length Max 835 +expl/path length Min 165 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.040071 +expl/Actions Std 0.823352 +expl/Actions Max 2.53328 +expl/Actions Min -2.4943 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 177692 +eval/num paths total 181 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.138666 +eval/Actions Std 0.765826 +eval/Actions Max 0.99872 +eval/Actions Min -0.998572 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.75299e-06 +time/evaluation sampling (s) 3.1551 +time/exploration sampling (s) 3.47258 +time/logging (s) 0.00706015 +time/saving (s) 0.0095652 +time/training (s) 13.4726 +time/epoch (s) 20.117 +time/total (s) 3921.63 +Epoch -820 +------------------------------ ---------------- +2022-05-15 19:08:07.991832 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -819 finished +------------------------------ ---------------- +epoch -819 +replay_buffer/size 999047 +trainer/num train calls 182000 +trainer/QF1 Loss 1.40653 +trainer/QF2 Loss 1.50183 +trainer/Policy Loss 1.35721 +trainer/Q1 Predictions Mean -70.8218 +trainer/Q1 Predictions Std 19.2551 +trainer/Q1 Predictions Max -0.335621 +trainer/Q1 Predictions Min -86.338 +trainer/Q2 Predictions Mean -70.863 +trainer/Q2 Predictions Std 19.3067 +trainer/Q2 Predictions Max -0.514033 +trainer/Q2 Predictions Min -86.6232 +trainer/Q Targets Mean -69.9735 +trainer/Q Targets Std 19.4469 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.1042 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0203604 +trainer/policy/mean Std 0.724404 +trainer/policy/mean Max 0.999657 +trainer/policy/mean Min -0.998284 +trainer/policy/std Mean 0.433521 +trainer/policy/std Std 0.0221027 +trainer/policy/std Max 0.456897 +trainer/policy/std Min 0.394734 +trainer/Advantage Weights Mean 0.410841 +trainer/Advantage Weights Std 6.23901 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.94273e-19 +trainer/Advantage Score Mean -1.08017 +trainer/Advantage Score Std 0.647747 +trainer/Advantage Score Max 0.579672 +trainer/Advantage Score Min -4.21512 +trainer/V1 Predictions Mean -69.7405 +trainer/V1 Predictions Std 19.5368 +trainer/V1 Predictions Max 0.810777 +trainer/V1 Predictions Min -85.8445 +trainer/VF Loss 0.159711 +expl/num steps total 182000 +expl/num paths total 189 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.032971 +expl/Actions Std 0.860444 +expl/Actions Max 2.44616 +expl/Actions Min -2.56699 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 178692 +eval/num paths total 182 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.065217 +eval/Actions Std 0.732671 +eval/Actions Max 0.99905 +eval/Actions Min -0.997725 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.65008e-06 +time/evaluation sampling (s) 3.14973 +time/exploration sampling (s) 3.67208 +time/logging (s) 0.00826357 +time/saving (s) 0.0146717 +time/training (s) 13.5179 +time/epoch (s) 20.3627 +time/total (s) 3941.99 +Epoch -819 +------------------------------ ---------------- +2022-05-15 19:08:28.481588 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -818 finished +------------------------------ ---------------- +epoch -818 +replay_buffer/size 999047 +trainer/num train calls 183000 +trainer/QF1 Loss 1.36768 +trainer/QF2 Loss 1.40013 +trainer/Policy Loss 21.1746 +trainer/Q1 Predictions Mean -71.4596 +trainer/Q1 Predictions Std 17.6224 +trainer/Q1 Predictions Max -1.16132 +trainer/Q1 Predictions Min -85.9356 +trainer/Q2 Predictions Mean -71.4138 +trainer/Q2 Predictions Std 17.6215 +trainer/Q2 Predictions Max -0.883304 +trainer/Q2 Predictions Min -85.8218 +trainer/Q Targets Mean -71.5195 +trainer/Q Targets Std 18.012 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6063 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -3.96944e-05 +trainer/policy/mean Std 0.709815 +trainer/policy/mean Max 0.99764 +trainer/policy/mean Min -0.996633 +trainer/policy/std Mean 0.432392 +trainer/policy/std Std 0.0211432 +trainer/policy/std Max 0.457878 +trainer/policy/std Min 0.396196 +trainer/Advantage Weights Mean 4.94089 +trainer/Advantage Weights Std 16.7328 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.23138e-13 +trainer/Advantage Score Mean -0.268857 +trainer/Advantage Score Std 0.52139 +trainer/Advantage Score Max 1.38724 +trainer/Advantage Score Min -2.84911 +trainer/V1 Predictions Mean -71.3299 +trainer/V1 Predictions Std 17.9011 +trainer/V1 Predictions Max -1.31992 +trainer/V1 Predictions Min -87.1962 +trainer/VF Loss 0.0547806 +expl/num steps total 183000 +expl/num paths total 190 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.033262 +expl/Actions Std 0.821111 +expl/Actions Max 2.27631 +expl/Actions Min -2.19236 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 179692 +eval/num paths total 183 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00257014 +eval/Actions Std 0.698735 +eval/Actions Max 0.999023 +eval/Actions Min -0.999144 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.62214e-06 +time/evaluation sampling (s) 3.10709 +time/exploration sampling (s) 3.27692 +time/logging (s) 0.00692948 +time/saving (s) 0.00933063 +time/training (s) 14.0806 +time/epoch (s) 20.4809 +time/total (s) 3962.48 +Epoch -818 +------------------------------ ---------------- +2022-05-15 19:08:48.121623 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -817 finished +------------------------------ ---------------- +epoch -817 +replay_buffer/size 999047 +trainer/num train calls 184000 +trainer/QF1 Loss 0.665527 +trainer/QF2 Loss 0.640259 +trainer/Policy Loss 7.48661 +trainer/Q1 Predictions Mean -73.3465 +trainer/Q1 Predictions Std 15.6956 +trainer/Q1 Predictions Max -3.14415 +trainer/Q1 Predictions Min -86.4117 +trainer/Q2 Predictions Mean -73.2913 +trainer/Q2 Predictions Std 15.7479 +trainer/Q2 Predictions Max -2.47568 +trainer/Q2 Predictions Min -85.917 +trainer/Q Targets Mean -73.1442 +trainer/Q Targets Std 15.6949 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2562 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0111109 +trainer/policy/mean Std 0.711796 +trainer/policy/mean Max 0.997215 +trainer/policy/mean Min -0.995664 +trainer/policy/std Mean 0.434043 +trainer/policy/std Std 0.021596 +trainer/policy/std Max 0.456006 +trainer/policy/std Min 0.3981 +trainer/Advantage Weights Mean 1.86803 +trainer/Advantage Weights Std 11.3435 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.94405e-09 +trainer/Advantage Score Mean -0.4354 +trainer/Advantage Score Std 0.365933 +trainer/Advantage Score Max 0.833832 +trainer/Advantage Score Min -1.96435 +trainer/V1 Predictions Mean -72.9928 +trainer/V1 Predictions Std 15.67 +trainer/V1 Predictions Max -3.37289 +trainer/V1 Predictions Min -86.3453 +trainer/VF Loss 0.0382867 +expl/num steps total 184000 +expl/num paths total 191 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0565728 +expl/Actions Std 0.815362 +expl/Actions Max 2.46787 +expl/Actions Min -2.25632 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 180692 +eval/num paths total 184 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0715303 +eval/Actions Std 0.727634 +eval/Actions Max 0.998771 +eval/Actions Min -0.997591 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.44286e-06 +time/evaluation sampling (s) 2.80344 +time/exploration sampling (s) 3.11162 +time/logging (s) 0.00772469 +time/saving (s) 0.0119072 +time/training (s) 13.7005 +time/epoch (s) 19.6352 +time/total (s) 3982.12 +Epoch -817 +------------------------------ ---------------- +2022-05-15 19:09:07.579753 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -816 finished +------------------------------ ---------------- +epoch -816 +replay_buffer/size 999047 +trainer/num train calls 185000 +trainer/QF1 Loss 0.650896 +trainer/QF2 Loss 0.637334 +trainer/Policy Loss 14.5252 +trainer/Q1 Predictions Mean -70.4446 +trainer/Q1 Predictions Std 19.0497 +trainer/Q1 Predictions Max -0.392657 +trainer/Q1 Predictions Min -86.2173 +trainer/Q2 Predictions Mean -70.4301 +trainer/Q2 Predictions Std 18.996 +trainer/Q2 Predictions Max -0.502767 +trainer/Q2 Predictions Min -86.1161 +trainer/Q Targets Mean -70.531 +trainer/Q Targets Std 19.1139 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.387 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00901579 +trainer/policy/mean Std 0.712447 +trainer/policy/mean Max 0.999254 +trainer/policy/mean Min -0.999845 +trainer/policy/std Mean 0.433599 +trainer/policy/std Std 0.0212359 +trainer/policy/std Max 0.456656 +trainer/policy/std Min 0.397543 +trainer/Advantage Weights Mean 4.19071 +trainer/Advantage Weights Std 16.2213 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.61201e-09 +trainer/Advantage Score Mean -0.247467 +trainer/Advantage Score Std 0.404901 +trainer/Advantage Score Max 0.667439 +trainer/Advantage Score Min -2.02458 +trainer/V1 Predictions Mean -70.3662 +trainer/V1 Predictions Std 19.0603 +trainer/V1 Predictions Max 0.133821 +trainer/V1 Predictions Min -86.2909 +trainer/VF Loss 0.0337617 +expl/num steps total 185000 +expl/num paths total 193 +expl/path length Mean 500 +expl/path length Std 352 +expl/path length Max 852 +expl/path length Min 148 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0509099 +expl/Actions Std 0.819349 +expl/Actions Max 2.51081 +expl/Actions Min -2.3356 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 181692 +eval/num paths total 185 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.185203 +eval/Actions Std 0.625049 +eval/Actions Max 0.999141 +eval/Actions Min -0.997331 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.73809e-06 +time/evaluation sampling (s) 2.78843 +time/exploration sampling (s) 3.16282 +time/logging (s) 0.0073149 +time/saving (s) 0.00998665 +time/training (s) 13.4819 +time/epoch (s) 19.4505 +time/total (s) 4001.57 +Epoch -816 +------------------------------ ---------------- +2022-05-15 19:09:27.353506 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -815 finished +------------------------------ ---------------- +epoch -815 +replay_buffer/size 999047 +trainer/num train calls 186000 +trainer/QF1 Loss 1.57837 +trainer/QF2 Loss 1.65616 +trainer/Policy Loss 18.9625 +trainer/Q1 Predictions Mean -70.1819 +trainer/Q1 Predictions Std 19.0395 +trainer/Q1 Predictions Max -0.610321 +trainer/Q1 Predictions Min -86.5664 +trainer/Q2 Predictions Mean -70.2661 +trainer/Q2 Predictions Std 19.0402 +trainer/Q2 Predictions Max -0.734061 +trainer/Q2 Predictions Min -86.2181 +trainer/Q Targets Mean -70.2721 +trainer/Q Targets Std 19.2534 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7979 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.014048 +trainer/policy/mean Std 0.699477 +trainer/policy/mean Max 0.999011 +trainer/policy/mean Min -0.995633 +trainer/policy/std Mean 0.432632 +trainer/policy/std Std 0.021575 +trainer/policy/std Max 0.456551 +trainer/policy/std Min 0.39435 +trainer/Advantage Weights Mean 4.18025 +trainer/Advantage Weights Std 18.3911 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.98443e-12 +trainer/Advantage Score Mean -0.384337 +trainer/Advantage Score Std 0.539361 +trainer/Advantage Score Max 1.99797 +trainer/Advantage Score Min -2.55535 +trainer/V1 Predictions Mean -70.1508 +trainer/V1 Predictions Std 19.2189 +trainer/V1 Predictions Max 0.659665 +trainer/V1 Predictions Min -86.5995 +trainer/VF Loss 0.0747363 +expl/num steps total 186000 +expl/num paths total 195 +expl/path length Mean 500 +expl/path length Std 236 +expl/path length Max 736 +expl/path length Min 264 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0431882 +expl/Actions Std 0.815608 +expl/Actions Max 2.22115 +expl/Actions Min -2.26476 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 182692 +eval/num paths total 186 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.10531 +eval/Actions Std 0.676492 +eval/Actions Max 0.998299 +eval/Actions Min -0.996817 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.58395e-06 +time/evaluation sampling (s) 2.65343 +time/exploration sampling (s) 3.54997 +time/logging (s) 0.00698384 +time/saving (s) 0.00940165 +time/training (s) 13.5476 +time/epoch (s) 19.7674 +time/total (s) 4021.34 +Epoch -815 +------------------------------ ---------------- +2022-05-15 19:09:47.300854 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -814 finished +------------------------------ ---------------- +epoch -814 +replay_buffer/size 999047 +trainer/num train calls 187000 +trainer/QF1 Loss 0.808772 +trainer/QF2 Loss 0.823106 +trainer/Policy Loss 2.263 +trainer/Q1 Predictions Mean -70.2254 +trainer/Q1 Predictions Std 19.6507 +trainer/Q1 Predictions Max -0.409557 +trainer/Q1 Predictions Min -87.152 +trainer/Q2 Predictions Mean -70.1924 +trainer/Q2 Predictions Std 19.7041 +trainer/Q2 Predictions Max -0.664637 +trainer/Q2 Predictions Min -87.1185 +trainer/Q Targets Mean -69.9026 +trainer/Q Targets Std 19.5169 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2136 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0170731 +trainer/policy/mean Std 0.700609 +trainer/policy/mean Max 0.997584 +trainer/policy/mean Min -0.999213 +trainer/policy/std Mean 0.43429 +trainer/policy/std Std 0.021858 +trainer/policy/std Max 0.458133 +trainer/policy/std Min 0.399231 +trainer/Advantage Weights Mean 0.913143 +trainer/Advantage Weights Std 7.35681 +trainer/Advantage Weights Max 89.2962 +trainer/Advantage Weights Min 6.65145e-18 +trainer/Advantage Score Mean -0.568002 +trainer/Advantage Score Std 0.564252 +trainer/Advantage Score Max 0.449196 +trainer/Advantage Score Min -3.95517 +trainer/V1 Predictions Mean -69.6211 +trainer/V1 Predictions Std 19.7364 +trainer/V1 Predictions Max -0.466141 +trainer/V1 Predictions Min -86.1845 +trainer/VF Loss 0.0660282 +expl/num steps total 187000 +expl/num paths total 196 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.21595 +expl/Actions Std 0.820771 +expl/Actions Max 2.60898 +expl/Actions Min -2.14282 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 183692 +eval/num paths total 187 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0979368 +eval/Actions Std 0.656183 +eval/Actions Max 0.99925 +eval/Actions Min -0.99848 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.77674e-06 +time/evaluation sampling (s) 2.71964 +time/exploration sampling (s) 3.56697 +time/logging (s) 0.0083176 +time/saving (s) 0.0106359 +time/training (s) 13.6375 +time/epoch (s) 19.9431 +time/total (s) 4041.29 +Epoch -814 +------------------------------ ---------------- +2022-05-15 19:10:07.340137 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -813 finished +------------------------------ ---------------- +epoch -813 +replay_buffer/size 999047 +trainer/num train calls 188000 +trainer/QF1 Loss 0.535383 +trainer/QF2 Loss 0.546563 +trainer/Policy Loss 26.6649 +trainer/Q1 Predictions Mean -72.3724 +trainer/Q1 Predictions Std 17.1702 +trainer/Q1 Predictions Max -1.70191 +trainer/Q1 Predictions Min -85.7741 +trainer/Q2 Predictions Mean -72.4482 +trainer/Q2 Predictions Std 17.1094 +trainer/Q2 Predictions Max -1.91675 +trainer/Q2 Predictions Min -87.0538 +trainer/Q Targets Mean -72.5661 +trainer/Q Targets Std 17.2872 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3938 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00895236 +trainer/policy/mean Std 0.707356 +trainer/policy/mean Max 0.999011 +trainer/policy/mean Min -0.996612 +trainer/policy/std Mean 0.433576 +trainer/policy/std Std 0.0209743 +trainer/policy/std Max 0.457614 +trainer/policy/std Min 0.397997 +trainer/Advantage Weights Mean 3.55879 +trainer/Advantage Weights Std 15.5716 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.64004e-10 +trainer/Advantage Score Mean -0.264475 +trainer/Advantage Score Std 0.442546 +trainer/Advantage Score Max 1.94115 +trainer/Advantage Score Min -2.07599 +trainer/V1 Predictions Mean -72.3094 +trainer/V1 Predictions Std 17.3929 +trainer/V1 Predictions Max -0.631777 +trainer/V1 Predictions Min -86.0666 +trainer/VF Loss 0.0543459 +expl/num steps total 188000 +expl/num paths total 198 +expl/path length Mean 500 +expl/path length Std 394 +expl/path length Max 894 +expl/path length Min 106 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0563199 +expl/Actions Std 0.801624 +expl/Actions Max 2.56804 +expl/Actions Min -2.48166 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 184692 +eval/num paths total 188 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0600044 +eval/Actions Std 0.721261 +eval/Actions Max 0.998958 +eval/Actions Min -0.998294 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.25684e-06 +time/evaluation sampling (s) 3.02076 +time/exploration sampling (s) 3.76408 +time/logging (s) 0.00769351 +time/saving (s) 0.0121524 +time/training (s) 13.2279 +time/epoch (s) 20.0326 +time/total (s) 4061.33 +Epoch -813 +------------------------------ ---------------- +2022-05-15 19:10:27.531243 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -812 finished +------------------------------ ---------------- +epoch -812 +replay_buffer/size 999047 +trainer/num train calls 189000 +trainer/QF1 Loss 0.393789 +trainer/QF2 Loss 0.36642 +trainer/Policy Loss 34.3777 +trainer/Q1 Predictions Mean -73.2397 +trainer/Q1 Predictions Std 15.0949 +trainer/Q1 Predictions Max -1.56662 +trainer/Q1 Predictions Min -86.1793 +trainer/Q2 Predictions Mean -73.2316 +trainer/Q2 Predictions Std 15.1397 +trainer/Q2 Predictions Max -1.68296 +trainer/Q2 Predictions Min -86.3531 +trainer/Q Targets Mean -73.4447 +trainer/Q Targets Std 15.0902 +trainer/Q Targets Max -3.15507 +trainer/Q Targets Min -86.7315 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0398374 +trainer/policy/mean Std 0.703655 +trainer/policy/mean Max 0.99916 +trainer/policy/mean Min -0.999241 +trainer/policy/std Mean 0.432237 +trainer/policy/std Std 0.0211526 +trainer/policy/std Max 0.456522 +trainer/policy/std Min 0.396143 +trainer/Advantage Weights Mean 6.85383 +trainer/Advantage Weights Std 18.5031 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.38322e-15 +trainer/Advantage Score Mean -0.160769 +trainer/Advantage Score Std 0.486746 +trainer/Advantage Score Max 1.38175 +trainer/Advantage Score Min -3.42144 +trainer/V1 Predictions Mean -73.1903 +trainer/V1 Predictions Std 15.1631 +trainer/V1 Predictions Max -1.70806 +trainer/V1 Predictions Min -86.3695 +trainer/VF Loss 0.0536436 +expl/num steps total 189000 +expl/num paths total 199 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0432227 +expl/Actions Std 0.837052 +expl/Actions Max 2.46967 +expl/Actions Min -2.35726 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 185562 +eval/num paths total 189 +eval/path length Mean 870 +eval/path length Std 0 +eval/path length Max 870 +eval/path length Min 870 +eval/Rewards Mean 0.00114943 +eval/Rewards Std 0.0338837 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0366462 +eval/Actions Std 0.740878 +eval/Actions Max 0.999642 +eval/Actions Min -0.998878 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 7.91997e-06 +time/evaluation sampling (s) 2.98424 +time/exploration sampling (s) 3.79103 +time/logging (s) 0.00815886 +time/saving (s) 0.0126255 +time/training (s) 13.3884 +time/epoch (s) 20.1844 +time/total (s) 4081.51 +Epoch -812 +------------------------------ ---------------- +2022-05-15 19:10:47.619466 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -811 finished +------------------------------ ---------------- +epoch -811 +replay_buffer/size 999047 +trainer/num train calls 190000 +trainer/QF1 Loss 0.835411 +trainer/QF2 Loss 0.803488 +trainer/Policy Loss 17.8267 +trainer/Q1 Predictions Mean -71.8227 +trainer/Q1 Predictions Std 17.7141 +trainer/Q1 Predictions Max -0.770781 +trainer/Q1 Predictions Min -85.412 +trainer/Q2 Predictions Mean -71.7694 +trainer/Q2 Predictions Std 17.7708 +trainer/Q2 Predictions Max -0.891101 +trainer/Q2 Predictions Min -85.3204 +trainer/Q Targets Mean -71.602 +trainer/Q Targets Std 17.9746 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.4748 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0126158 +trainer/policy/mean Std 0.697068 +trainer/policy/mean Max 0.998884 +trainer/policy/mean Min -0.994386 +trainer/policy/std Mean 0.431977 +trainer/policy/std Std 0.0222703 +trainer/policy/std Max 0.458173 +trainer/policy/std Min 0.394995 +trainer/Advantage Weights Mean 5.64803 +trainer/Advantage Weights Std 20.1698 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.97959e-11 +trainer/Advantage Score Mean -0.361305 +trainer/Advantage Score Std 0.50602 +trainer/Advantage Score Max 1.87607 +trainer/Advantage Score Min -2.39473 +trainer/V1 Predictions Mean -71.4254 +trainer/V1 Predictions Std 17.9408 +trainer/V1 Predictions Max 0.384829 +trainer/V1 Predictions Min -85.1907 +trainer/VF Loss 0.0685988 +expl/num steps total 190000 +expl/num paths total 200 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.077369 +expl/Actions Std 0.816287 +expl/Actions Max 2.24515 +expl/Actions Min -2.2724 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 186562 +eval/num paths total 190 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.100258 +eval/Actions Std 0.726348 +eval/Actions Max 0.999355 +eval/Actions Min -0.998085 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.11062e-06 +time/evaluation sampling (s) 3.22355 +time/exploration sampling (s) 3.83513 +time/logging (s) 0.00687396 +time/saving (s) 0.0104548 +time/training (s) 13.0041 +time/epoch (s) 20.0801 +time/total (s) 4101.6 +Epoch -811 +------------------------------ ---------------- +2022-05-15 19:11:07.846649 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -810 finished +------------------------------ ---------------- +epoch -810 +replay_buffer/size 999047 +trainer/num train calls 191000 +trainer/QF1 Loss 0.834291 +trainer/QF2 Loss 0.817439 +trainer/Policy Loss 9.27058 +trainer/Q1 Predictions Mean -71.1909 +trainer/Q1 Predictions Std 18.9701 +trainer/Q1 Predictions Max -0.356334 +trainer/Q1 Predictions Min -86.6498 +trainer/Q2 Predictions Mean -71.0613 +trainer/Q2 Predictions Std 19.0045 +trainer/Q2 Predictions Max -0.56634 +trainer/Q2 Predictions Min -86.2179 +trainer/Q Targets Mean -71.1451 +trainer/Q Targets Std 18.5258 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2219 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0154729 +trainer/policy/mean Std 0.70534 +trainer/policy/mean Max 0.997567 +trainer/policy/mean Min -0.997636 +trainer/policy/std Mean 0.433631 +trainer/policy/std Std 0.0221011 +trainer/policy/std Max 0.461155 +trainer/policy/std Min 0.399857 +trainer/Advantage Weights Mean 2.55346 +trainer/Advantage Weights Std 13.1961 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.13223e-16 +trainer/Advantage Score Mean -0.453664 +trainer/Advantage Score Std 0.589093 +trainer/Advantage Score Max 2.16382 +trainer/Advantage Score Min -3.52058 +trainer/V1 Predictions Mean -70.7817 +trainer/V1 Predictions Std 18.8997 +trainer/V1 Predictions Max -1.31378 +trainer/V1 Predictions Min -86.1307 +trainer/VF Loss 0.0758147 +expl/num steps total 191000 +expl/num paths total 201 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0333674 +expl/Actions Std 0.826676 +expl/Actions Max 2.49251 +expl/Actions Min -2.25834 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 187562 +eval/num paths total 191 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.137658 +eval/Actions Std 0.709296 +eval/Actions Max 0.998311 +eval/Actions Min -0.998867 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.55089e-06 +time/evaluation sampling (s) 2.958 +time/exploration sampling (s) 3.6206 +time/logging (s) 0.00701436 +time/saving (s) 0.00964498 +time/training (s) 13.6263 +time/epoch (s) 20.2216 +time/total (s) 4121.82 +Epoch -810 +------------------------------ ---------------- +2022-05-15 19:11:28.264836 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -809 finished +------------------------------ --------------- +epoch -809 +replay_buffer/size 999047 +trainer/num train calls 192000 +trainer/QF1 Loss 0.755238 +trainer/QF2 Loss 0.646268 +trainer/Policy Loss 34.8624 +trainer/Q1 Predictions Mean -71.4048 +trainer/Q1 Predictions Std 17.9913 +trainer/Q1 Predictions Max -1.78128 +trainer/Q1 Predictions Min -86.6796 +trainer/Q2 Predictions Mean -71.461 +trainer/Q2 Predictions Std 17.9948 +trainer/Q2 Predictions Max -1.90305 +trainer/Q2 Predictions Min -85.8959 +trainer/Q Targets Mean -71.5967 +trainer/Q Targets Std 18.1151 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5691 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00189987 +trainer/policy/mean Std 0.709084 +trainer/policy/mean Max 0.997507 +trainer/policy/mean Min -0.997348 +trainer/policy/std Mean 0.434761 +trainer/policy/std Std 0.0219975 +trainer/policy/std Max 0.461018 +trainer/policy/std Min 0.401275 +trainer/Advantage Weights Mean 7.54132 +trainer/Advantage Weights Std 21.9975 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.3376e-07 +trainer/Advantage Score Mean -0.168257 +trainer/Advantage Score Std 0.45094 +trainer/Advantage Score Max 1.47887 +trainer/Advantage Score Min -1.58272 +trainer/V1 Predictions Mean -71.3725 +trainer/V1 Predictions Std 18.1588 +trainer/V1 Predictions Max -1.0523 +trainer/V1 Predictions Min -86.64 +trainer/VF Loss 0.058223 +expl/num steps total 192000 +expl/num paths total 202 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0537561 +expl/Actions Std 0.818277 +expl/Actions Max 2.4308 +expl/Actions Min -2.31376 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 188562 +eval/num paths total 192 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0672101 +eval/Actions Std 0.726275 +eval/Actions Max 0.9995 +eval/Actions Min -0.999117 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.6701e-06 +time/evaluation sampling (s) 3.02635 +time/exploration sampling (s) 3.26508 +time/logging (s) 0.00734315 +time/saving (s) 0.0101759 +time/training (s) 14.1029 +time/epoch (s) 20.4118 +time/total (s) 4142.24 +Epoch -809 +------------------------------ --------------- +2022-05-15 19:11:47.785243 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -808 finished +------------------------------ ---------------- +epoch -808 +replay_buffer/size 999047 +trainer/num train calls 193000 +trainer/QF1 Loss 0.910198 +trainer/QF2 Loss 1.05011 +trainer/Policy Loss 8.25816 +trainer/Q1 Predictions Mean -72.5112 +trainer/Q1 Predictions Std 17.8156 +trainer/Q1 Predictions Max -0.321231 +trainer/Q1 Predictions Min -86.439 +trainer/Q2 Predictions Mean -72.6083 +trainer/Q2 Predictions Std 17.7888 +trainer/Q2 Predictions Max -0.761054 +trainer/Q2 Predictions Min -86.5809 +trainer/Q Targets Mean -71.9273 +trainer/Q Targets Std 18.0684 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9511 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00636432 +trainer/policy/mean Std 0.718612 +trainer/policy/mean Max 0.998652 +trainer/policy/mean Min -0.997574 +trainer/policy/std Mean 0.434268 +trainer/policy/std Std 0.0220005 +trainer/policy/std Max 0.461384 +trainer/policy/std Min 0.397355 +trainer/Advantage Weights Mean 1.78661 +trainer/Advantage Weights Std 10.9853 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.57871e-17 +trainer/Advantage Score Mean -0.528584 +trainer/Advantage Score Std 0.486632 +trainer/Advantage Score Max 0.556297 +trainer/Advantage Score Min -3.68844 +trainer/V1 Predictions Mean -71.724 +trainer/V1 Predictions Std 17.9872 +trainer/V1 Predictions Max 0.449775 +trainer/V1 Predictions Min -85.8598 +trainer/VF Loss 0.0556919 +expl/num steps total 193000 +expl/num paths total 203 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0778716 +expl/Actions Std 0.806065 +expl/Actions Max 2.46883 +expl/Actions Min -2.24844 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 189562 +eval/num paths total 193 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0839636 +eval/Actions Std 0.731914 +eval/Actions Max 0.998873 +eval/Actions Min -0.997283 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.21306e-06 +time/evaluation sampling (s) 2.61628 +time/exploration sampling (s) 3.48675 +time/logging (s) 0.00753949 +time/saving (s) 0.0116409 +time/training (s) 13.3922 +time/epoch (s) 19.5144 +time/total (s) 4161.76 +Epoch -808 +------------------------------ ---------------- +2022-05-15 19:12:07.817828 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -807 finished +------------------------------ ---------------- +epoch -807 +replay_buffer/size 999047 +trainer/num train calls 194000 +trainer/QF1 Loss 0.942543 +trainer/QF2 Loss 1.00713 +trainer/Policy Loss 9.17101 +trainer/Q1 Predictions Mean -71.4724 +trainer/Q1 Predictions Std 19.0794 +trainer/Q1 Predictions Max -1.43223 +trainer/Q1 Predictions Min -86.6198 +trainer/Q2 Predictions Mean -71.5161 +trainer/Q2 Predictions Std 19.0295 +trainer/Q2 Predictions Max -1.46283 +trainer/Q2 Predictions Min -86.9044 +trainer/Q Targets Mean -70.8396 +trainer/Q Targets Std 19.3036 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9673 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00888624 +trainer/policy/mean Std 0.704527 +trainer/policy/mean Max 0.999005 +trainer/policy/mean Min -0.997998 +trainer/policy/std Mean 0.435216 +trainer/policy/std Std 0.0238142 +trainer/policy/std Max 0.462357 +trainer/policy/std Min 0.394003 +trainer/Advantage Weights Mean 2.02459 +trainer/Advantage Weights Std 12.4471 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.93922e-15 +trainer/Advantage Score Mean -0.43954 +trainer/Advantage Score Std 0.531428 +trainer/Advantage Score Max 1.08661 +trainer/Advantage Score Min -3.23483 +trainer/V1 Predictions Mean -70.572 +trainer/V1 Predictions Std 19.3077 +trainer/V1 Predictions Max -0.882718 +trainer/V1 Predictions Min -85.8101 +trainer/VF Loss 0.0570193 +expl/num steps total 194000 +expl/num paths total 205 +expl/path length Mean 500 +expl/path length Std 297 +expl/path length Max 797 +expl/path length Min 203 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0129668 +expl/Actions Std 0.824303 +expl/Actions Max 2.48538 +expl/Actions Min -2.40799 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 190562 +eval/num paths total 194 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.010575 +eval/Actions Std 0.738415 +eval/Actions Max 0.999206 +eval/Actions Min -0.999108 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.90992e-06 +time/evaluation sampling (s) 2.67512 +time/exploration sampling (s) 3.86895 +time/logging (s) 0.00699124 +time/saving (s) 0.00948122 +time/training (s) 13.4642 +time/epoch (s) 20.0247 +time/total (s) 4181.79 +Epoch -807 +------------------------------ ---------------- +2022-05-15 19:12:28.337036 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -806 finished +------------------------------ ---------------- +epoch -806 +replay_buffer/size 999047 +trainer/num train calls 195000 +trainer/QF1 Loss 0.776706 +trainer/QF2 Loss 0.866242 +trainer/Policy Loss 21.0253 +trainer/Q1 Predictions Mean -70.998 +trainer/Q1 Predictions Std 18.782 +trainer/Q1 Predictions Max -3.19188 +trainer/Q1 Predictions Min -85.8744 +trainer/Q2 Predictions Mean -71.0189 +trainer/Q2 Predictions Std 18.7163 +trainer/Q2 Predictions Max -3.47623 +trainer/Q2 Predictions Min -85.8052 +trainer/Q Targets Mean -71.1492 +trainer/Q Targets Std 18.9559 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.8766 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0100534 +trainer/policy/mean Std 0.710677 +trainer/policy/mean Max 0.999464 +trainer/policy/mean Min -0.997204 +trainer/policy/std Mean 0.434845 +trainer/policy/std Std 0.0228038 +trainer/policy/std Max 0.46135 +trainer/policy/std Min 0.397133 +trainer/Advantage Weights Mean 4.99027 +trainer/Advantage Weights Std 18.0589 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.15536e-12 +trainer/Advantage Score Mean -0.304181 +trainer/Advantage Score Std 0.484845 +trainer/Advantage Score Max 0.769131 +trainer/Advantage Score Min -2.74866 +trainer/V1 Predictions Mean -70.9397 +trainer/V1 Predictions Std 18.8528 +trainer/V1 Predictions Max -3.88242 +trainer/V1 Predictions Min -85.9451 +trainer/VF Loss 0.047169 +expl/num steps total 195000 +expl/num paths total 206 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0378157 +expl/Actions Std 0.829132 +expl/Actions Max 2.50162 +expl/Actions Min -2.45121 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 191562 +eval/num paths total 195 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.206566 +eval/Actions Std 0.728207 +eval/Actions Max 0.997088 +eval/Actions Min -0.998245 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.50991e-06 +time/evaluation sampling (s) 2.97597 +time/exploration sampling (s) 3.98518 +time/logging (s) 0.00731559 +time/saving (s) 0.0099575 +time/training (s) 13.5351 +time/epoch (s) 20.5135 +time/total (s) 4202.31 +Epoch -806 +------------------------------ ---------------- +2022-05-15 19:12:49.186623 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -805 finished +------------------------------ ---------------- +epoch -805 +replay_buffer/size 999047 +trainer/num train calls 196000 +trainer/QF1 Loss 0.515061 +trainer/QF2 Loss 0.600749 +trainer/Policy Loss 41.6508 +trainer/Q1 Predictions Mean -71.1607 +trainer/Q1 Predictions Std 18.2876 +trainer/Q1 Predictions Max -1.11279 +trainer/Q1 Predictions Min -86.4612 +trainer/Q2 Predictions Mean -71.1357 +trainer/Q2 Predictions Std 18.3314 +trainer/Q2 Predictions Max -0.996557 +trainer/Q2 Predictions Min -86.6779 +trainer/Q Targets Mean -71.1319 +trainer/Q Targets Std 18.1615 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3984 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0084276 +trainer/policy/mean Std 0.70124 +trainer/policy/mean Max 0.996959 +trainer/policy/mean Min -0.999762 +trainer/policy/std Mean 0.433312 +trainer/policy/std Std 0.0228019 +trainer/policy/std Max 0.456587 +trainer/policy/std Min 0.396169 +trainer/Advantage Weights Mean 11.0309 +trainer/Advantage Weights Std 27.1512 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.78944e-11 +trainer/Advantage Score Mean -0.109061 +trainer/Advantage Score Std 0.501744 +trainer/Advantage Score Max 2.05948 +trainer/Advantage Score Min -2.34131 +trainer/V1 Predictions Mean -70.8152 +trainer/V1 Predictions Std 18.2365 +trainer/V1 Predictions Max -1.94918 +trainer/V1 Predictions Min -86.2606 +trainer/VF Loss 0.0805061 +expl/num steps total 196000 +expl/num paths total 207 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0545804 +expl/Actions Std 0.811657 +expl/Actions Max 2.84174 +expl/Actions Min -2.28338 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 192562 +eval/num paths total 196 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.183578 +eval/Actions Std 0.733294 +eval/Actions Max 0.998984 +eval/Actions Min -0.99761 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.75904e-06 +time/evaluation sampling (s) 2.97914 +time/exploration sampling (s) 3.93226 +time/logging (s) 0.00689548 +time/saving (s) 0.0106327 +time/training (s) 13.914 +time/epoch (s) 20.8429 +time/total (s) 4223.15 +Epoch -805 +------------------------------ ---------------- +2022-05-15 19:13:08.300266 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -804 finished +------------------------------ ---------------- +epoch -804 +replay_buffer/size 999047 +trainer/num train calls 197000 +trainer/QF1 Loss 0.738971 +trainer/QF2 Loss 0.72219 +trainer/Policy Loss 23.2988 +trainer/Q1 Predictions Mean -71.0534 +trainer/Q1 Predictions Std 19.2337 +trainer/Q1 Predictions Max -0.511323 +trainer/Q1 Predictions Min -85.8073 +trainer/Q2 Predictions Mean -71.1393 +trainer/Q2 Predictions Std 19.2097 +trainer/Q2 Predictions Max -0.325595 +trainer/Q2 Predictions Min -85.9561 +trainer/Q Targets Mean -71.0356 +trainer/Q Targets Std 19.2014 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.8079 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0417436 +trainer/policy/mean Std 0.698164 +trainer/policy/mean Max 0.998883 +trainer/policy/mean Min -0.997237 +trainer/policy/std Mean 0.433625 +trainer/policy/std Std 0.0218876 +trainer/policy/std Max 0.455984 +trainer/policy/std Min 0.397394 +trainer/Advantage Weights Mean 3.03508 +trainer/Advantage Weights Std 14.9205 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.20915e-11 +trainer/Advantage Score Mean -0.375121 +trainer/Advantage Score Std 0.448551 +trainer/Advantage Score Max 1.53838 +trainer/Advantage Score Min -2.51385 +trainer/V1 Predictions Mean -70.7835 +trainer/V1 Predictions Std 19.2359 +trainer/V1 Predictions Max 0.943445 +trainer/V1 Predictions Min -85.6048 +trainer/VF Loss 0.0540015 +expl/num steps total 197000 +expl/num paths total 209 +expl/path length Mean 500 +expl/path length Std 297 +expl/path length Max 797 +expl/path length Min 203 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0432388 +expl/Actions Std 0.819285 +expl/Actions Max 2.27036 +expl/Actions Min -2.25805 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 193562 +eval/num paths total 197 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0755426 +eval/Actions Std 0.693036 +eval/Actions Max 0.999389 +eval/Actions Min -0.999164 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78139e-06 +time/evaluation sampling (s) 2.93755 +time/exploration sampling (s) 3.63055 +time/logging (s) 0.00931765 +time/saving (s) 0.0138533 +time/training (s) 12.5191 +time/epoch (s) 19.1103 +time/total (s) 4242.27 +Epoch -804 +------------------------------ ---------------- +2022-05-15 19:13:28.135596 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -803 finished +------------------------------ ---------------- +epoch -803 +replay_buffer/size 999047 +trainer/num train calls 198000 +trainer/QF1 Loss 1.1708 +trainer/QF2 Loss 1.12814 +trainer/Policy Loss 53.3527 +trainer/Q1 Predictions Mean -68.8365 +trainer/Q1 Predictions Std 21.138 +trainer/Q1 Predictions Max -0.854556 +trainer/Q1 Predictions Min -85.8951 +trainer/Q2 Predictions Mean -68.8824 +trainer/Q2 Predictions Std 21.1798 +trainer/Q2 Predictions Max -0.631852 +trainer/Q2 Predictions Min -85.8705 +trainer/Q Targets Mean -68.9238 +trainer/Q Targets Std 21.3644 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0871 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0157316 +trainer/policy/mean Std 0.701263 +trainer/policy/mean Max 0.998841 +trainer/policy/mean Min -0.998295 +trainer/policy/std Mean 0.432049 +trainer/policy/std Std 0.0221537 +trainer/policy/std Max 0.454913 +trainer/policy/std Min 0.396913 +trainer/Advantage Weights Mean 9.07025 +trainer/Advantage Weights Std 23.6005 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.00235e-14 +trainer/Advantage Score Mean -0.183137 +trainer/Advantage Score Std 0.570923 +trainer/Advantage Score Max 2.36774 +trainer/Advantage Score Min -3.22338 +trainer/V1 Predictions Mean -68.745 +trainer/V1 Predictions Std 21.2129 +trainer/V1 Predictions Max -0.678142 +trainer/V1 Predictions Min -85.9464 +trainer/VF Loss 0.0889073 +expl/num steps total 198000 +expl/num paths total 210 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0268655 +expl/Actions Std 0.817334 +expl/Actions Max 2.33353 +expl/Actions Min -2.52771 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 194562 +eval/num paths total 198 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0940767 +eval/Actions Std 0.663851 +eval/Actions Max 0.999741 +eval/Actions Min -0.998946 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.37254e-06 +time/evaluation sampling (s) 3.11163 +time/exploration sampling (s) 3.35521 +time/logging (s) 0.00679969 +time/saving (s) 0.00944353 +time/training (s) 13.3405 +time/epoch (s) 19.8236 +time/total (s) 4262.1 +Epoch -803 +------------------------------ ---------------- +2022-05-15 19:13:48.663064 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -802 finished +------------------------------ ---------------- +epoch -802 +replay_buffer/size 999047 +trainer/num train calls 199000 +trainer/QF1 Loss 0.534981 +trainer/QF2 Loss 0.525015 +trainer/Policy Loss 9.47063 +trainer/Q1 Predictions Mean -73.1195 +trainer/Q1 Predictions Std 15.0745 +trainer/Q1 Predictions Max -1.51886 +trainer/Q1 Predictions Min -86.2798 +trainer/Q2 Predictions Mean -73.1752 +trainer/Q2 Predictions Std 15.023 +trainer/Q2 Predictions Max -1.59477 +trainer/Q2 Predictions Min -86.3393 +trainer/Q Targets Mean -72.982 +trainer/Q Targets Std 15.054 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4945 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00986635 +trainer/policy/mean Std 0.704734 +trainer/policy/mean Max 0.998478 +trainer/policy/mean Min -0.992702 +trainer/policy/std Mean 0.433484 +trainer/policy/std Std 0.0234576 +trainer/policy/std Max 0.461503 +trainer/policy/std Min 0.394115 +trainer/Advantage Weights Mean 2.47565 +trainer/Advantage Weights Std 12.678 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.28143e-11 +trainer/Advantage Score Mean -0.4068 +trainer/Advantage Score Std 0.452592 +trainer/Advantage Score Max 1.17225 +trainer/Advantage Score Min -2.50805 +trainer/V1 Predictions Mean -72.6899 +trainer/V1 Predictions Std 15.2406 +trainer/V1 Predictions Max -1.16293 +trainer/V1 Predictions Min -86.3561 +trainer/VF Loss 0.0478565 +expl/num steps total 199000 +expl/num paths total 211 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0597123 +expl/Actions Std 0.809359 +expl/Actions Max 2.70261 +expl/Actions Min -2.46378 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 195562 +eval/num paths total 199 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.144746 +eval/Actions Std 0.719991 +eval/Actions Max 0.998888 +eval/Actions Min -0.994424 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.91178e-06 +time/evaluation sampling (s) 2.84986 +time/exploration sampling (s) 3.24048 +time/logging (s) 0.0099244 +time/saving (s) 0.0146514 +time/training (s) 14.4097 +time/epoch (s) 20.5246 +time/total (s) 4282.63 +Epoch -802 +------------------------------ ---------------- +2022-05-15 19:14:09.137014 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -801 finished +------------------------------ ---------------- +epoch -801 +replay_buffer/size 999047 +trainer/num train calls 200000 +trainer/QF1 Loss 0.415542 +trainer/QF2 Loss 0.442678 +trainer/Policy Loss 13.1171 +trainer/Q1 Predictions Mean -72.0136 +trainer/Q1 Predictions Std 17.1142 +trainer/Q1 Predictions Max -1.53452 +trainer/Q1 Predictions Min -86.7237 +trainer/Q2 Predictions Mean -72.0836 +trainer/Q2 Predictions Std 17.1022 +trainer/Q2 Predictions Max -1.3962 +trainer/Q2 Predictions Min -86.0967 +trainer/Q Targets Mean -72.0627 +trainer/Q Targets Std 17.0395 +trainer/Q Targets Max -2.53648 +trainer/Q Targets Min -85.9095 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000791235 +trainer/policy/mean Std 0.702812 +trainer/policy/mean Max 0.997773 +trainer/policy/mean Min -0.99751 +trainer/policy/std Mean 0.432974 +trainer/policy/std Std 0.0221785 +trainer/policy/std Max 0.455406 +trainer/policy/std Min 0.394814 +trainer/Advantage Weights Mean 2.4369 +trainer/Advantage Weights Std 13.1937 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.50062e-18 +trainer/Advantage Score Mean -0.465495 +trainer/Advantage Score Std 0.489346 +trainer/Advantage Score Max 1.11997 +trainer/Advantage Score Min -3.93064 +trainer/V1 Predictions Mean -71.7718 +trainer/V1 Predictions Std 17.2647 +trainer/V1 Predictions Max -0.416489 +trainer/V1 Predictions Min -85.7908 +trainer/VF Loss 0.0573955 +expl/num steps total 200000 +expl/num paths total 213 +expl/path length Mean 500 +expl/path length Std 450 +expl/path length Max 950 +expl/path length Min 50 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0576204 +expl/Actions Std 0.822483 +expl/Actions Max 2.37254 +expl/Actions Min -2.28911 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 196562 +eval/num paths total 200 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0953027 +eval/Actions Std 0.730088 +eval/Actions Max 0.999184 +eval/Actions Min -0.996854 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.1637e-06 +time/evaluation sampling (s) 2.74181 +time/exploration sampling (s) 3.96511 +time/logging (s) 0.00846847 +time/saving (s) 0.0145653 +time/training (s) 13.7348 +time/epoch (s) 20.4647 +time/total (s) 4303.1 +Epoch -801 +------------------------------ ---------------- +2022-05-15 19:14:29.949709 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -800 finished +------------------------------ ---------------- +epoch -800 +replay_buffer/size 999047 +trainer/num train calls 201000 +trainer/QF1 Loss 0.482244 +trainer/QF2 Loss 0.530283 +trainer/Policy Loss 18.4993 +trainer/Q1 Predictions Mean -72.2134 +trainer/Q1 Predictions Std 16.8773 +trainer/Q1 Predictions Max -0.660469 +trainer/Q1 Predictions Min -86.4415 +trainer/Q2 Predictions Mean -72.1415 +trainer/Q2 Predictions Std 16.7949 +trainer/Q2 Predictions Max -0.338399 +trainer/Q2 Predictions Min -86.3268 +trainer/Q Targets Mean -72.1871 +trainer/Q Targets Std 16.8454 +trainer/Q Targets Max -1.00441 +trainer/Q Targets Min -86.6885 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00981852 +trainer/policy/mean Std 0.710745 +trainer/policy/mean Max 0.999231 +trainer/policy/mean Min -0.997565 +trainer/policy/std Mean 0.432339 +trainer/policy/std Std 0.0209207 +trainer/policy/std Max 0.455041 +trainer/policy/std Min 0.397136 +trainer/Advantage Weights Mean 3.50439 +trainer/Advantage Weights Std 14.6155 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.29706e-14 +trainer/Advantage Score Mean -0.327766 +trainer/Advantage Score Std 0.54976 +trainer/Advantage Score Max 1.25162 +trainer/Advantage Score Min -3.19761 +trainer/V1 Predictions Mean -71.8647 +trainer/V1 Predictions Std 17.0007 +trainer/V1 Predictions Max -0.373934 +trainer/V1 Predictions Min -86.6101 +trainer/VF Loss 0.0568233 +expl/num steps total 201000 +expl/num paths total 214 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0459975 +expl/Actions Std 0.862004 +expl/Actions Max 2.46603 +expl/Actions Min -2.37944 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 197562 +eval/num paths total 201 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0281192 +eval/Actions Std 0.709713 +eval/Actions Max 0.998985 +eval/Actions Min -0.999648 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.91085e-06 +time/evaluation sampling (s) 3.18743 +time/exploration sampling (s) 3.90875 +time/logging (s) 0.0069402 +time/saving (s) 0.0199063 +time/training (s) 13.6797 +time/epoch (s) 20.8027 +time/total (s) 4323.9 +Epoch -800 +------------------------------ ---------------- +2022-05-15 19:14:50.419713 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -799 finished +------------------------------ ---------------- +epoch -799 +replay_buffer/size 999047 +trainer/num train calls 202000 +trainer/QF1 Loss 0.903248 +trainer/QF2 Loss 0.948621 +trainer/Policy Loss 5.30177 +trainer/Q1 Predictions Mean -72.7682 +trainer/Q1 Predictions Std 16.9443 +trainer/Q1 Predictions Max -0.324515 +trainer/Q1 Predictions Min -85.6698 +trainer/Q2 Predictions Mean -72.7497 +trainer/Q2 Predictions Std 16.9404 +trainer/Q2 Predictions Max -0.468941 +trainer/Q2 Predictions Min -85.7084 +trainer/Q Targets Mean -72.747 +trainer/Q Targets Std 16.7296 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.3114 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0223428 +trainer/policy/mean Std 0.704647 +trainer/policy/mean Max 0.999224 +trainer/policy/mean Min -0.998773 +trainer/policy/std Mean 0.432078 +trainer/policy/std Std 0.0209634 +trainer/policy/std Max 0.455351 +trainer/policy/std Min 0.396199 +trainer/Advantage Weights Mean 1.55695 +trainer/Advantage Weights Std 10.8671 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.92429e-10 +trainer/Advantage Score Mean -0.464197 +trainer/Advantage Score Std 0.45468 +trainer/Advantage Score Max 3.15878 +trainer/Advantage Score Min -2.23713 +trainer/V1 Predictions Mean -72.4551 +trainer/V1 Predictions Std 16.9184 +trainer/V1 Predictions Max -0.471167 +trainer/V1 Predictions Min -85.1623 +trainer/VF Loss 0.0848839 +expl/num steps total 202000 +expl/num paths total 215 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0344639 +expl/Actions Std 0.816859 +expl/Actions Max 2.59031 +expl/Actions Min -2.33666 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 198562 +eval/num paths total 202 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.114113 +eval/Actions Std 0.712819 +eval/Actions Max 0.99913 +eval/Actions Min -0.998135 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.19723e-06 +time/evaluation sampling (s) 3.10298 +time/exploration sampling (s) 3.82129 +time/logging (s) 0.00819887 +time/saving (s) 0.0128323 +time/training (s) 13.5198 +time/epoch (s) 20.4651 +time/total (s) 4344.37 +Epoch -799 +------------------------------ ---------------- +2022-05-15 19:15:10.209563 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -798 finished +------------------------------ ---------------- +epoch -798 +replay_buffer/size 999047 +trainer/num train calls 203000 +trainer/QF1 Loss 0.426277 +trainer/QF2 Loss 0.423549 +trainer/Policy Loss 7.03279 +trainer/Q1 Predictions Mean -73.1125 +trainer/Q1 Predictions Std 15.9211 +trainer/Q1 Predictions Max -5.81734 +trainer/Q1 Predictions Min -86.4747 +trainer/Q2 Predictions Mean -73.0736 +trainer/Q2 Predictions Std 15.9283 +trainer/Q2 Predictions Max -5.36288 +trainer/Q2 Predictions Min -86.4753 +trainer/Q Targets Mean -73.1776 +trainer/Q Targets Std 16.1077 +trainer/Q Targets Max -5.58024 +trainer/Q Targets Min -86.8069 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0187275 +trainer/policy/mean Std 0.70513 +trainer/policy/mean Max 0.99773 +trainer/policy/mean Min -0.995073 +trainer/policy/std Mean 0.432434 +trainer/policy/std Std 0.0216031 +trainer/policy/std Max 0.454739 +trainer/policy/std Min 0.397818 +trainer/Advantage Weights Mean 1.44733 +trainer/Advantage Weights Std 10.5887 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.71078e-15 +trainer/Advantage Score Mean -0.497672 +trainer/Advantage Score Std 0.524881 +trainer/Advantage Score Max 1.13498 +trainer/Advantage Score Min -3.40018 +trainer/V1 Predictions Mean -72.9126 +trainer/V1 Predictions Std 16.2014 +trainer/V1 Predictions Max -3.06012 +trainer/V1 Predictions Min -86.1651 +trainer/VF Loss 0.0585179 +expl/num steps total 203000 +expl/num paths total 216 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.135875 +expl/Actions Std 0.830449 +expl/Actions Max 2.54012 +expl/Actions Min -2.56953 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 199562 +eval/num paths total 203 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.233168 +eval/Actions Std 0.668894 +eval/Actions Max 0.999253 +eval/Actions Min -0.998445 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.99094e-06 +time/evaluation sampling (s) 3.08356 +time/exploration sampling (s) 3.86507 +time/logging (s) 0.00683342 +time/saving (s) 0.00940445 +time/training (s) 12.816 +time/epoch (s) 19.7809 +time/total (s) 4364.16 +Epoch -798 +------------------------------ ---------------- +2022-05-15 19:15:30.740798 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -797 finished +------------------------------ ---------------- +epoch -797 +replay_buffer/size 999047 +trainer/num train calls 204000 +trainer/QF1 Loss 0.814208 +trainer/QF2 Loss 0.890778 +trainer/Policy Loss 13.0215 +trainer/Q1 Predictions Mean -70.7682 +trainer/Q1 Predictions Std 19.4262 +trainer/Q1 Predictions Max -2.02271 +trainer/Q1 Predictions Min -86.4675 +trainer/Q2 Predictions Mean -70.738 +trainer/Q2 Predictions Std 19.3831 +trainer/Q2 Predictions Max -2.37872 +trainer/Q2 Predictions Min -86.2948 +trainer/Q Targets Mean -70.6609 +trainer/Q Targets Std 19.2613 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.8814 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0184625 +trainer/policy/mean Std 0.711086 +trainer/policy/mean Max 0.999185 +trainer/policy/mean Min -0.999393 +trainer/policy/std Mean 0.432181 +trainer/policy/std Std 0.0218535 +trainer/policy/std Max 0.457602 +trainer/policy/std Min 0.394349 +trainer/Advantage Weights Mean 3.50558 +trainer/Advantage Weights Std 15.3668 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.91053e-12 +trainer/Advantage Score Mean -0.458786 +trainer/Advantage Score Std 0.498947 +trainer/Advantage Score Max 1.2839 +trainer/Advantage Score Min -2.5698 +trainer/V1 Predictions Mean -70.3264 +trainer/V1 Predictions Std 19.5107 +trainer/V1 Predictions Max -1.05354 +trainer/V1 Predictions Min -85.7369 +trainer/VF Loss 0.059722 +expl/num steps total 204000 +expl/num paths total 217 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0264958 +expl/Actions Std 0.829134 +expl/Actions Max 2.58673 +expl/Actions Min -2.20255 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 200562 +eval/num paths total 204 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0853862 +eval/Actions Std 0.721261 +eval/Actions Max 0.999179 +eval/Actions Min -0.998863 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.42703e-06 +time/evaluation sampling (s) 2.96263 +time/exploration sampling (s) 3.33038 +time/logging (s) 0.00911371 +time/saving (s) 0.0124525 +time/training (s) 14.2129 +time/epoch (s) 20.5275 +time/total (s) 4384.69 +Epoch -797 +------------------------------ ---------------- +2022-05-15 19:15:50.481447 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -796 finished +------------------------------ ---------------- +epoch -796 +replay_buffer/size 999047 +trainer/num train calls 205000 +trainer/QF1 Loss 0.906338 +trainer/QF2 Loss 0.903489 +trainer/Policy Loss 17.2553 +trainer/Q1 Predictions Mean -71.665 +trainer/Q1 Predictions Std 17.4788 +trainer/Q1 Predictions Max -1.07367 +trainer/Q1 Predictions Min -85.7764 +trainer/Q2 Predictions Mean -71.7664 +trainer/Q2 Predictions Std 17.4734 +trainer/Q2 Predictions Max -0.916559 +trainer/Q2 Predictions Min -86.5123 +trainer/Q Targets Mean -71.8296 +trainer/Q Targets Std 17.4366 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4594 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00862841 +trainer/policy/mean Std 0.702003 +trainer/policy/mean Max 0.99755 +trainer/policy/mean Min -0.996525 +trainer/policy/std Mean 0.432149 +trainer/policy/std Std 0.0209117 +trainer/policy/std Max 0.454876 +trainer/policy/std Min 0.393791 +trainer/Advantage Weights Mean 3.17491 +trainer/Advantage Weights Std 14.9833 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.65657e-15 +trainer/Advantage Score Mean -0.378978 +trainer/Advantage Score Std 0.510993 +trainer/Advantage Score Max 1.29968 +trainer/Advantage Score Min -3.22711 +trainer/V1 Predictions Mean -71.5638 +trainer/V1 Predictions Std 17.4731 +trainer/V1 Predictions Max 0.231943 +trainer/V1 Predictions Min -86.2711 +trainer/VF Loss 0.0537345 +expl/num steps total 205000 +expl/num paths total 218 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0254105 +expl/Actions Std 0.824723 +expl/Actions Max 2.46468 +expl/Actions Min -2.40664 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 201562 +eval/num paths total 205 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.173055 +eval/Actions Std 0.700381 +eval/Actions Max 0.998791 +eval/Actions Min -0.998975 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 6.38328e-06 +time/evaluation sampling (s) 2.71476 +time/exploration sampling (s) 3.51176 +time/logging (s) 0.0100795 +time/saving (s) 0.0146498 +time/training (s) 13.484 +time/epoch (s) 19.7352 +time/total (s) 4404.43 +Epoch -796 +------------------------------ ---------------- +2022-05-15 19:16:10.829870 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -795 finished +------------------------------ ---------------- +epoch -795 +replay_buffer/size 999047 +trainer/num train calls 206000 +trainer/QF1 Loss 1.13296 +trainer/QF2 Loss 1.11452 +trainer/Policy Loss 77.3469 +trainer/Q1 Predictions Mean -69.59 +trainer/Q1 Predictions Std 20.1044 +trainer/Q1 Predictions Max -0.27663 +trainer/Q1 Predictions Min -85.5204 +trainer/Q2 Predictions Mean -69.6223 +trainer/Q2 Predictions Std 20.1093 +trainer/Q2 Predictions Max -0.325297 +trainer/Q2 Predictions Min -85.7831 +trainer/Q Targets Mean -70.3972 +trainer/Q Targets Std 20.1468 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3981 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0124493 +trainer/policy/mean Std 0.70715 +trainer/policy/mean Max 0.99922 +trainer/policy/mean Min -0.999126 +trainer/policy/std Mean 0.432198 +trainer/policy/std Std 0.0206486 +trainer/policy/std Max 0.454117 +trainer/policy/std Min 0.396983 +trainer/Advantage Weights Mean 16.4783 +trainer/Advantage Weights Std 31.3797 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.91869e-07 +trainer/Advantage Score Mean -0.0265154 +trainer/Advantage Score Std 0.471775 +trainer/Advantage Score Max 1.68469 +trainer/Advantage Score Min -1.434 +trainer/V1 Predictions Mean -70.1359 +trainer/V1 Predictions Std 20.2151 +trainer/V1 Predictions Max -0.242397 +trainer/V1 Predictions Min -86.3007 +trainer/VF Loss 0.097837 +expl/num steps total 206000 +expl/num paths total 219 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0589493 +expl/Actions Std 0.803521 +expl/Actions Max 2.44127 +expl/Actions Min -2.61366 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 202562 +eval/num paths total 206 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0778231 +eval/Actions Std 0.635992 +eval/Actions Max 0.999202 +eval/Actions Min -0.997268 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.63611e-06 +time/evaluation sampling (s) 2.76967 +time/exploration sampling (s) 3.98113 +time/logging (s) 0.00724519 +time/saving (s) 0.0103671 +time/training (s) 13.5685 +time/epoch (s) 20.3369 +time/total (s) 4424.77 +Epoch -795 +------------------------------ ---------------- +2022-05-15 19:16:31.669213 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -794 finished +------------------------------ ---------------- +epoch -794 +replay_buffer/size 999047 +trainer/num train calls 207000 +trainer/QF1 Loss 1.70437 +trainer/QF2 Loss 1.87936 +trainer/Policy Loss 4.65768 +trainer/Q1 Predictions Mean -70.9011 +trainer/Q1 Predictions Std 19.2691 +trainer/Q1 Predictions Max -1.62153 +trainer/Q1 Predictions Min -86.2623 +trainer/Q2 Predictions Mean -70.9738 +trainer/Q2 Predictions Std 19.2599 +trainer/Q2 Predictions Max -2.36921 +trainer/Q2 Predictions Min -86.6717 +trainer/Q Targets Mean -70.1041 +trainer/Q Targets Std 19.4971 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9352 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0401977 +trainer/policy/mean Std 0.699886 +trainer/policy/mean Max 0.99942 +trainer/policy/mean Min -0.998601 +trainer/policy/std Mean 0.431109 +trainer/policy/std Std 0.021247 +trainer/policy/std Max 0.455477 +trainer/policy/std Min 0.39774 +trainer/Advantage Weights Mean 1.15461 +trainer/Advantage Weights Std 10.4835 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.98615e-15 +trainer/Advantage Score Mean -1.05162 +trainer/Advantage Score Std 0.633115 +trainer/Advantage Score Max 2.10593 +trainer/Advantage Score Min -3.27493 +trainer/V1 Predictions Mean -69.8705 +trainer/V1 Predictions Std 19.3522 +trainer/V1 Predictions Max -1.06082 +trainer/V1 Predictions Min -86.1746 +trainer/VF Loss 0.165977 +expl/num steps total 207000 +expl/num paths total 220 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0370989 +expl/Actions Std 0.902785 +expl/Actions Max 2.55487 +expl/Actions Min -2.67164 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 203562 +eval/num paths total 207 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0638935 +eval/Actions Std 0.7355 +eval/Actions Max 0.999635 +eval/Actions Min -0.998276 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.01915e-06 +time/evaluation sampling (s) 3.09713 +time/exploration sampling (s) 3.97067 +time/logging (s) 0.0073051 +time/saving (s) 0.0104287 +time/training (s) 13.7477 +time/epoch (s) 20.8333 +time/total (s) 4445.61 +Epoch -794 +------------------------------ ---------------- +2022-05-15 19:16:52.284871 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -793 finished +------------------------------ ---------------- +epoch -793 +replay_buffer/size 999047 +trainer/num train calls 208000 +trainer/QF1 Loss 0.741688 +trainer/QF2 Loss 0.840421 +trainer/Policy Loss 10.3374 +trainer/Q1 Predictions Mean -73.1859 +trainer/Q1 Predictions Std 16.0752 +trainer/Q1 Predictions Max -3.25077 +trainer/Q1 Predictions Min -86.2988 +trainer/Q2 Predictions Mean -73.0972 +trainer/Q2 Predictions Std 16.2005 +trainer/Q2 Predictions Max -3.34165 +trainer/Q2 Predictions Min -86.0925 +trainer/Q Targets Mean -72.9922 +trainer/Q Targets Std 16.2635 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.167 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00117213 +trainer/policy/mean Std 0.704917 +trainer/policy/mean Max 0.997787 +trainer/policy/mean Min -0.996432 +trainer/policy/std Mean 0.431924 +trainer/policy/std Std 0.0203355 +trainer/policy/std Max 0.455603 +trainer/policy/std Min 0.399788 +trainer/Advantage Weights Mean 2.69768 +trainer/Advantage Weights Std 14.843 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.07823e-09 +trainer/Advantage Score Mean -0.432316 +trainer/Advantage Score Std 0.396818 +trainer/Advantage Score Max 0.978058 +trainer/Advantage Score Min -1.99918 +trainer/V1 Predictions Mean -72.819 +trainer/V1 Predictions Std 16.1683 +trainer/V1 Predictions Max -3.12764 +trainer/V1 Predictions Min -86.0055 +trainer/VF Loss 0.0456681 +expl/num steps total 208000 +expl/num paths total 221 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.229619 +expl/Actions Std 0.889727 +expl/Actions Max 2.42998 +expl/Actions Min -2.29596 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 204562 +eval/num paths total 208 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.122592 +eval/Actions Std 0.666842 +eval/Actions Max 0.999341 +eval/Actions Min -0.997375 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.61329e-06 +time/evaluation sampling (s) 2.93631 +time/exploration sampling (s) 3.9203 +time/logging (s) 0.00862126 +time/saving (s) 0.0116044 +time/training (s) 13.7335 +time/epoch (s) 20.6103 +time/total (s) 4466.22 +Epoch -793 +------------------------------ ---------------- +2022-05-15 19:17:12.991062 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -792 finished +------------------------------ ---------------- +epoch -792 +replay_buffer/size 999047 +trainer/num train calls 209000 +trainer/QF1 Loss 0.627354 +trainer/QF2 Loss 0.6531 +trainer/Policy Loss 10.7336 +trainer/Q1 Predictions Mean -70.6236 +trainer/Q1 Predictions Std 19.8709 +trainer/Q1 Predictions Max -0.762923 +trainer/Q1 Predictions Min -85.4119 +trainer/Q2 Predictions Mean -70.6289 +trainer/Q2 Predictions Std 19.83 +trainer/Q2 Predictions Max -1.10486 +trainer/Q2 Predictions Min -85.3534 +trainer/Q Targets Mean -70.498 +trainer/Q Targets Std 20.0854 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9333 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.024772 +trainer/policy/mean Std 0.702754 +trainer/policy/mean Max 0.999565 +trainer/policy/mean Min -0.997871 +trainer/policy/std Mean 0.43089 +trainer/policy/std Std 0.0217611 +trainer/policy/std Max 0.453525 +trainer/policy/std Min 0.395453 +trainer/Advantage Weights Mean 2.50802 +trainer/Advantage Weights Std 12.7666 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.12572e-12 +trainer/Advantage Score Mean -0.422818 +trainer/Advantage Score Std 0.468975 +trainer/Advantage Score Max 0.822128 +trainer/Advantage Score Min -2.64914 +trainer/V1 Predictions Mean -70.2229 +trainer/V1 Predictions Std 20.2099 +trainer/V1 Predictions Max -0.0326522 +trainer/V1 Predictions Min -85.7772 +trainer/VF Loss 0.0480027 +expl/num steps total 209000 +expl/num paths total 223 +expl/path length Mean 500 +expl/path length Std 458 +expl/path length Max 958 +expl/path length Min 42 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0459366 +expl/Actions Std 0.852215 +expl/Actions Max 2.45452 +expl/Actions Min -2.32669 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 205562 +eval/num paths total 209 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.240101 +eval/Actions Std 0.69072 +eval/Actions Max 0.998837 +eval/Actions Min -0.998164 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.77115e-06 +time/evaluation sampling (s) 3.1536 +time/exploration sampling (s) 3.40573 +time/logging (s) 0.0088144 +time/saving (s) 0.0169499 +time/training (s) 14.114 +time/epoch (s) 20.6991 +time/total (s) 4486.93 +Epoch -792 +------------------------------ ---------------- +2022-05-15 19:17:32.376203 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -791 finished +------------------------------ ---------------- +epoch -791 +replay_buffer/size 999047 +trainer/num train calls 210000 +trainer/QF1 Loss 0.704493 +trainer/QF2 Loss 0.680142 +trainer/Policy Loss 10.38 +trainer/Q1 Predictions Mean -72.3984 +trainer/Q1 Predictions Std 17.3761 +trainer/Q1 Predictions Max -0.544812 +trainer/Q1 Predictions Min -85.5229 +trainer/Q2 Predictions Mean -72.3971 +trainer/Q2 Predictions Std 17.4348 +trainer/Q2 Predictions Max -0.361855 +trainer/Q2 Predictions Min -85.6358 +trainer/Q Targets Mean -72.3294 +trainer/Q Targets Std 17.1851 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7379 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00573677 +trainer/policy/mean Std 0.700516 +trainer/policy/mean Max 0.999855 +trainer/policy/mean Min -0.995569 +trainer/policy/std Mean 0.430862 +trainer/policy/std Std 0.0207762 +trainer/policy/std Max 0.45217 +trainer/policy/std Min 0.397359 +trainer/Advantage Weights Mean 2.13344 +trainer/Advantage Weights Std 12.6387 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.24299e-15 +trainer/Advantage Score Mean -0.43577 +trainer/Advantage Score Std 0.468626 +trainer/Advantage Score Max 1.00985 +trainer/Advantage Score Min -3.23149 +trainer/V1 Predictions Mean -72.0333 +trainer/V1 Predictions Std 17.4672 +trainer/V1 Predictions Max 0.492127 +trainer/V1 Predictions Min -85.6932 +trainer/VF Loss 0.0488008 +expl/num steps total 210000 +expl/num paths total 224 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0361679 +expl/Actions Std 0.830546 +expl/Actions Max 2.36332 +expl/Actions Min -2.60193 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 206259 +eval/num paths total 210 +eval/path length Mean 697 +eval/path length Std 0 +eval/path length Max 697 +eval/path length Min 697 +eval/Rewards Mean 0.00143472 +eval/Rewards Std 0.0378505 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0306679 +eval/Actions Std 0.739582 +eval/Actions Max 0.998696 +eval/Actions Min -0.997968 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.92901e-06 +time/evaluation sampling (s) 2.64108 +time/exploration sampling (s) 3.61108 +time/logging (s) 0.00687966 +time/saving (s) 0.0122372 +time/training (s) 13.1032 +time/epoch (s) 19.3744 +time/total (s) 4506.31 +Epoch -791 +------------------------------ ---------------- +2022-05-15 19:17:52.458180 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -790 finished +------------------------------ ---------------- +epoch -790 +replay_buffer/size 999047 +trainer/num train calls 211000 +trainer/QF1 Loss 0.755649 +trainer/QF2 Loss 0.859019 +trainer/Policy Loss 16.5142 +trainer/Q1 Predictions Mean -72.4007 +trainer/Q1 Predictions Std 17.3072 +trainer/Q1 Predictions Max -0.834391 +trainer/Q1 Predictions Min -86.2402 +trainer/Q2 Predictions Mean -72.391 +trainer/Q2 Predictions Std 17.3092 +trainer/Q2 Predictions Max -0.498617 +trainer/Q2 Predictions Min -87.2614 +trainer/Q Targets Mean -72.2395 +trainer/Q Targets Std 16.9797 +trainer/Q Targets Max -1.41842 +trainer/Q Targets Min -86.1232 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00733335 +trainer/policy/mean Std 0.700019 +trainer/policy/mean Max 0.99957 +trainer/policy/mean Min -0.998471 +trainer/policy/std Mean 0.430395 +trainer/policy/std Std 0.0206038 +trainer/policy/std Max 0.453433 +trainer/policy/std Min 0.396835 +trainer/Advantage Weights Mean 4.082 +trainer/Advantage Weights Std 18.5041 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.10384e-19 +trainer/Advantage Score Mean -0.402478 +trainer/Advantage Score Std 0.563645 +trainer/Advantage Score Max 1.76717 +trainer/Advantage Score Min -4.36503 +trainer/V1 Predictions Mean -71.8838 +trainer/V1 Predictions Std 17.3284 +trainer/V1 Predictions Max 1.01089 +trainer/V1 Predictions Min -86.3674 +trainer/VF Loss 0.0766851 +expl/num steps total 211000 +expl/num paths total 225 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.157436 +expl/Actions Std 0.815231 +expl/Actions Max 2.76566 +expl/Actions Min -2.33269 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 207259 +eval/num paths total 211 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.180238 +eval/Actions Std 0.629777 +eval/Actions Max 0.999397 +eval/Actions Min -0.992397 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.57092e-06 +time/evaluation sampling (s) 2.80511 +time/exploration sampling (s) 3.92125 +time/logging (s) 0.00754091 +time/saving (s) 0.0117281 +time/training (s) 13.3298 +time/epoch (s) 20.0754 +time/total (s) 4526.38 +Epoch -790 +------------------------------ ---------------- +2022-05-15 19:18:12.107473 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -789 finished +------------------------------ ---------------- +epoch -789 +replay_buffer/size 999047 +trainer/num train calls 212000 +trainer/QF1 Loss 0.751106 +trainer/QF2 Loss 0.860463 +trainer/Policy Loss 24.7237 +trainer/Q1 Predictions Mean -69.902 +trainer/Q1 Predictions Std 18.9416 +trainer/Q1 Predictions Max -0.538838 +trainer/Q1 Predictions Min -86.4038 +trainer/Q2 Predictions Mean -69.886 +trainer/Q2 Predictions Std 18.9688 +trainer/Q2 Predictions Max -0.341034 +trainer/Q2 Predictions Min -86.4096 +trainer/Q Targets Mean -70.0436 +trainer/Q Targets Std 19.1099 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4978 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00241619 +trainer/policy/mean Std 0.711806 +trainer/policy/mean Max 0.997489 +trainer/policy/mean Min -0.998292 +trainer/policy/std Mean 0.429577 +trainer/policy/std Std 0.0203596 +trainer/policy/std Max 0.451209 +trainer/policy/std Min 0.396289 +trainer/Advantage Weights Mean 5.05799 +trainer/Advantage Weights Std 18.2671 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.02642e-13 +trainer/Advantage Score Mean -0.297694 +trainer/Advantage Score Std 0.547872 +trainer/Advantage Score Max 0.815731 +trainer/Advantage Score Min -2.99075 +trainer/V1 Predictions Mean -69.7894 +trainer/V1 Predictions Std 19.2223 +trainer/V1 Predictions Max 1.22799 +trainer/V1 Predictions Min -86.3633 +trainer/VF Loss 0.0547401 +expl/num steps total 212000 +expl/num paths total 226 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00915191 +expl/Actions Std 0.813426 +expl/Actions Max 2.16842 +expl/Actions Min -2.24887 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 208259 +eval/num paths total 212 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.126983 +eval/Actions Std 0.628079 +eval/Actions Max 0.999328 +eval/Actions Min -0.999809 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.90433e-06 +time/evaluation sampling (s) 2.95978 +time/exploration sampling (s) 3.7972 +time/logging (s) 0.00696456 +time/saving (s) 0.00971416 +time/training (s) 12.8676 +time/epoch (s) 19.6413 +time/total (s) 4546.03 +Epoch -789 +------------------------------ ---------------- +2022-05-15 19:18:32.521314 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -788 finished +------------------------------ ---------------- +epoch -788 +replay_buffer/size 999047 +trainer/num train calls 213000 +trainer/QF1 Loss 0.661063 +trainer/QF2 Loss 0.882816 +trainer/Policy Loss 19.6023 +trainer/Q1 Predictions Mean -70.0631 +trainer/Q1 Predictions Std 20.3206 +trainer/Q1 Predictions Max -0.276524 +trainer/Q1 Predictions Min -85.9437 +trainer/Q2 Predictions Mean -69.9803 +trainer/Q2 Predictions Std 20.3512 +trainer/Q2 Predictions Max -0.326953 +trainer/Q2 Predictions Min -85.806 +trainer/Q Targets Mean -70.2695 +trainer/Q Targets Std 20.148 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0808 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00364656 +trainer/policy/mean Std 0.704659 +trainer/policy/mean Max 0.997114 +trainer/policy/mean Min -0.996402 +trainer/policy/std Mean 0.430496 +trainer/policy/std Std 0.020688 +trainer/policy/std Max 0.453364 +trainer/policy/std Min 0.395254 +trainer/Advantage Weights Mean 4.68209 +trainer/Advantage Weights Std 18.4506 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.66024e-13 +trainer/Advantage Score Mean -0.380449 +trainer/Advantage Score Std 0.520274 +trainer/Advantage Score Max 2.19475 +trainer/Advantage Score Min -2.94266 +trainer/V1 Predictions Mean -69.9006 +trainer/V1 Predictions Std 20.3807 +trainer/V1 Predictions Max 1.95891 +trainer/V1 Predictions Min -85.816 +trainer/VF Loss 0.0741853 +expl/num steps total 213000 +expl/num paths total 227 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0471241 +expl/Actions Std 0.830059 +expl/Actions Max 2.59782 +expl/Actions Min -2.69742 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 209259 +eval/num paths total 213 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0623738 +eval/Actions Std 0.741345 +eval/Actions Max 0.999366 +eval/Actions Min -0.998005 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.92994e-06 +time/evaluation sampling (s) 2.93851 +time/exploration sampling (s) 3.93325 +time/logging (s) 0.00774187 +time/saving (s) 0.011086 +time/training (s) 13.5179 +time/epoch (s) 20.4085 +time/total (s) 4566.44 +Epoch -788 +------------------------------ ---------------- +2022-05-15 19:18:53.318739 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -787 finished +------------------------------ ---------------- +epoch -787 +replay_buffer/size 999047 +trainer/num train calls 214000 +trainer/QF1 Loss 15.4051 +trainer/QF2 Loss 15.5135 +trainer/Policy Loss 15.6735 +trainer/Q1 Predictions Mean -73.087 +trainer/Q1 Predictions Std 17.0921 +trainer/Q1 Predictions Max -1.9629 +trainer/Q1 Predictions Min -86.3407 +trainer/Q2 Predictions Mean -73.1591 +trainer/Q2 Predictions Std 17.138 +trainer/Q2 Predictions Max -1.7283 +trainer/Q2 Predictions Min -86.2467 +trainer/Q Targets Mean -72.7707 +trainer/Q Targets Std 17.2867 +trainer/Q Targets Max -4.21367 +trainer/Q Targets Min -86.2626 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0126989 +trainer/policy/mean Std 0.710083 +trainer/policy/mean Max 0.99873 +trainer/policy/mean Min -0.997007 +trainer/policy/std Mean 0.432457 +trainer/policy/std Std 0.020395 +trainer/policy/std Max 0.453689 +trainer/policy/std Min 0.397754 +trainer/Advantage Weights Mean 3.15279 +trainer/Advantage Weights Std 15.9473 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.47255e-10 +trainer/Advantage Score Mean -0.432446 +trainer/Advantage Score Std 0.445342 +trainer/Advantage Score Max 1.23558 +trainer/Advantage Score Min -2.10146 +trainer/V1 Predictions Mean -72.7963 +trainer/V1 Predictions Std 17.1061 +trainer/V1 Predictions Max -0.781375 +trainer/V1 Predictions Min -86.0234 +trainer/VF Loss 0.0527171 +expl/num steps total 214000 +expl/num paths total 228 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0445604 +expl/Actions Std 0.830944 +expl/Actions Max 2.39521 +expl/Actions Min -2.37063 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 210259 +eval/num paths total 214 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.130061 +eval/Actions Std 0.684765 +eval/Actions Max 0.998986 +eval/Actions Min -0.99868 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.55415e-06 +time/evaluation sampling (s) 3.03807 +time/exploration sampling (s) 3.7053 +time/logging (s) 0.00783563 +time/saving (s) 0.0101055 +time/training (s) 14.029 +time/epoch (s) 20.7903 +time/total (s) 4587.24 +Epoch -787 +------------------------------ ---------------- +2022-05-15 19:19:13.006137 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -786 finished +------------------------------ ---------------- +epoch -786 +replay_buffer/size 999047 +trainer/num train calls 215000 +trainer/QF1 Loss 1.11249 +trainer/QF2 Loss 1.13967 +trainer/Policy Loss 11.5403 +trainer/Q1 Predictions Mean -71.298 +trainer/Q1 Predictions Std 20.4218 +trainer/Q1 Predictions Max -0.586307 +trainer/Q1 Predictions Min -86.7721 +trainer/Q2 Predictions Mean -71.3473 +trainer/Q2 Predictions Std 20.5037 +trainer/Q2 Predictions Max -0.39958 +trainer/Q2 Predictions Min -87.162 +trainer/Q Targets Mean -71.0503 +trainer/Q Targets Std 20.1768 +trainer/Q Targets Max 0.878585 +trainer/Q Targets Min -86.0453 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0178903 +trainer/policy/mean Std 0.711627 +trainer/policy/mean Max 0.997712 +trainer/policy/mean Min -0.99631 +trainer/policy/std Mean 0.430625 +trainer/policy/std Std 0.0217385 +trainer/policy/std Max 0.45576 +trainer/policy/std Min 0.39381 +trainer/Advantage Weights Mean 2.5188 +trainer/Advantage Weights Std 12.3264 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.46011e-17 +trainer/Advantage Score Mean -0.412817 +trainer/Advantage Score Std 0.64369 +trainer/Advantage Score Max 0.648582 +trainer/Advantage Score Min -3.76488 +trainer/V1 Predictions Mean -70.7358 +trainer/V1 Predictions Std 20.5176 +trainer/V1 Predictions Max 1.42479 +trainer/V1 Predictions Min -86.5054 +trainer/VF Loss 0.0648453 +expl/num steps total 215000 +expl/num paths total 229 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0349188 +expl/Actions Std 0.810062 +expl/Actions Max 2.48288 +expl/Actions Min -2.34213 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 211259 +eval/num paths total 215 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.31592 +eval/Actions Std 0.595688 +eval/Actions Max 0.998477 +eval/Actions Min -0.99814 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.88198e-06 +time/evaluation sampling (s) 3.05016 +time/exploration sampling (s) 3.24967 +time/logging (s) 0.00772372 +time/saving (s) 0.0115446 +time/training (s) 13.3615 +time/epoch (s) 19.6806 +time/total (s) 4606.92 +Epoch -786 +------------------------------ ---------------- +2022-05-15 19:19:33.015672 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -785 finished +------------------------------ ---------------- +epoch -785 +replay_buffer/size 999047 +trainer/num train calls 216000 +trainer/QF1 Loss 1.18796 +trainer/QF2 Loss 1.13374 +trainer/Policy Loss 8.51122 +trainer/Q1 Predictions Mean -73.1897 +trainer/Q1 Predictions Std 17.7999 +trainer/Q1 Predictions Max -1.6161 +trainer/Q1 Predictions Min -86.2639 +trainer/Q2 Predictions Mean -73.1527 +trainer/Q2 Predictions Std 17.7333 +trainer/Q2 Predictions Max -1.72487 +trainer/Q2 Predictions Min -86.1497 +trainer/Q Targets Mean -72.5176 +trainer/Q Targets Std 17.4568 +trainer/Q Targets Max -2.84085 +trainer/Q Targets Min -85.5824 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0299565 +trainer/policy/mean Std 0.708044 +trainer/policy/mean Max 0.998987 +trainer/policy/mean Min -0.998258 +trainer/policy/std Mean 0.429917 +trainer/policy/std Std 0.021302 +trainer/policy/std Max 0.452963 +trainer/policy/std Min 0.394724 +trainer/Advantage Weights Mean 1.39016 +trainer/Advantage Weights Std 10.2585 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.73501e-18 +trainer/Advantage Score Mean -0.596511 +trainer/Advantage Score Std 0.502424 +trainer/Advantage Score Max 1.43562 +trainer/Advantage Score Min -4.08955 +trainer/V1 Predictions Mean -72.2175 +trainer/V1 Predictions Std 17.7152 +trainer/V1 Predictions Max -0.702944 +trainer/V1 Predictions Min -85.4372 +trainer/VF Loss 0.0714793 +expl/num steps total 216000 +expl/num paths total 230 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0535007 +expl/Actions Std 0.820351 +expl/Actions Max 2.45624 +expl/Actions Min -2.33259 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 212259 +eval/num paths total 216 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00310036 +eval/Actions Std 0.720637 +eval/Actions Max 0.999807 +eval/Actions Min -0.999705 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.88012e-06 +time/evaluation sampling (s) 2.60949 +time/exploration sampling (s) 3.57578 +time/logging (s) 0.0100295 +time/saving (s) 0.0143906 +time/training (s) 13.7949 +time/epoch (s) 20.0046 +time/total (s) 4626.93 +Epoch -785 +------------------------------ ---------------- +2022-05-15 19:19:52.963129 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -784 finished +------------------------------ ---------------- +epoch -784 +replay_buffer/size 999047 +trainer/num train calls 217000 +trainer/QF1 Loss 1.32713 +trainer/QF2 Loss 1.50501 +trainer/Policy Loss 42.7683 +trainer/Q1 Predictions Mean -71.0656 +trainer/Q1 Predictions Std 18.3367 +trainer/Q1 Predictions Max -0.737541 +trainer/Q1 Predictions Min -86.6376 +trainer/Q2 Predictions Mean -71.0832 +trainer/Q2 Predictions Std 18.3017 +trainer/Q2 Predictions Max -0.72877 +trainer/Q2 Predictions Min -86.7431 +trainer/Q Targets Mean -71.7092 +trainer/Q Targets Std 18.3735 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4619 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00649576 +trainer/policy/mean Std 0.706903 +trainer/policy/mean Max 0.999572 +trainer/policy/mean Min -0.99841 +trainer/policy/std Mean 0.431232 +trainer/policy/std Std 0.0201964 +trainer/policy/std Max 0.455405 +trainer/policy/std Min 0.395443 +trainer/Advantage Weights Mean 12.039 +trainer/Advantage Weights Std 25.8273 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.47595e-15 +trainer/Advantage Score Mean -0.0511777 +trainer/Advantage Score Std 0.515244 +trainer/Advantage Score Max 2.1263 +trainer/Advantage Score Min -3.41495 +trainer/V1 Predictions Mean -71.4527 +trainer/V1 Predictions Std 18.3968 +trainer/V1 Predictions Max -1.00219 +trainer/V1 Predictions Min -86.5363 +trainer/VF Loss 0.0925052 +expl/num steps total 217000 +expl/num paths total 232 +expl/path length Mean 500 +expl/path length Std 498 +expl/path length Max 998 +expl/path length Min 2 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0531484 +expl/Actions Std 0.832997 +expl/Actions Max 2.63254 +expl/Actions Min -2.58272 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 213259 +eval/num paths total 217 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.16475 +eval/Actions Std 0.690247 +eval/Actions Max 0.998703 +eval/Actions Min -0.997981 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.45428e-06 +time/evaluation sampling (s) 2.92624 +time/exploration sampling (s) 3.99115 +time/logging (s) 0.00671968 +time/saving (s) 0.0100623 +time/training (s) 13.0022 +time/epoch (s) 19.9364 +time/total (s) 4646.87 +Epoch -784 +------------------------------ ---------------- +2022-05-15 19:20:13.365493 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -783 finished +------------------------------ ---------------- +epoch -783 +replay_buffer/size 999047 +trainer/num train calls 218000 +trainer/QF1 Loss 0.711732 +trainer/QF2 Loss 0.718116 +trainer/Policy Loss 13.4755 +trainer/Q1 Predictions Mean -71.9438 +trainer/Q1 Predictions Std 17.4751 +trainer/Q1 Predictions Max -0.508798 +trainer/Q1 Predictions Min -86.2872 +trainer/Q2 Predictions Mean -71.9827 +trainer/Q2 Predictions Std 17.5053 +trainer/Q2 Predictions Max -0.808359 +trainer/Q2 Predictions Min -86.1913 +trainer/Q Targets Mean -72.0647 +trainer/Q Targets Std 17.553 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3612 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0380994 +trainer/policy/mean Std 0.712481 +trainer/policy/mean Max 0.999037 +trainer/policy/mean Min -0.998958 +trainer/policy/std Mean 0.429831 +trainer/policy/std Std 0.0210746 +trainer/policy/std Max 0.45348 +trainer/policy/std Min 0.396067 +trainer/Advantage Weights Mean 3.63363 +trainer/Advantage Weights Std 16.5902 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.07895e-20 +trainer/Advantage Score Mean -0.435444 +trainer/Advantage Score Std 0.597185 +trainer/Advantage Score Max 1.50626 +trainer/Advantage Score Min -4.39624 +trainer/V1 Predictions Mean -71.8526 +trainer/V1 Predictions Std 17.5374 +trainer/V1 Predictions Max 0.924603 +trainer/V1 Predictions Min -86.1208 +trainer/VF Loss 0.0763117 +expl/num steps total 218000 +expl/num paths total 233 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.149799 +expl/Actions Std 0.782741 +expl/Actions Max 2.47985 +expl/Actions Min -2.36737 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 214259 +eval/num paths total 218 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0482207 +eval/Actions Std 0.722501 +eval/Actions Max 0.999568 +eval/Actions Min -0.999451 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.84612e-06 +time/evaluation sampling (s) 3.18584 +time/exploration sampling (s) 3.88189 +time/logging (s) 0.00758699 +time/saving (s) 0.0116377 +time/training (s) 13.3103 +time/epoch (s) 20.3972 +time/total (s) 4667.27 +Epoch -783 +------------------------------ ---------------- +2022-05-15 19:20:33.898620 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -782 finished +------------------------------ ---------------- +epoch -782 +replay_buffer/size 999047 +trainer/num train calls 219000 +trainer/QF1 Loss 0.397536 +trainer/QF2 Loss 0.449036 +trainer/Policy Loss 5.82778 +trainer/Q1 Predictions Mean -73.8265 +trainer/Q1 Predictions Std 15.1683 +trainer/Q1 Predictions Max -0.961115 +trainer/Q1 Predictions Min -86.1943 +trainer/Q2 Predictions Mean -73.8021 +trainer/Q2 Predictions Std 15.1247 +trainer/Q2 Predictions Max -1.16773 +trainer/Q2 Predictions Min -86.0809 +trainer/Q Targets Mean -73.6654 +trainer/Q Targets Std 15.2454 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7531 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00240454 +trainer/policy/mean Std 0.702914 +trainer/policy/mean Max 0.999383 +trainer/policy/mean Min -0.996007 +trainer/policy/std Mean 0.431055 +trainer/policy/std Std 0.0216146 +trainer/policy/std Max 0.457434 +trainer/policy/std Min 0.395408 +trainer/Advantage Weights Mean 1.68859 +trainer/Advantage Weights Std 9.92866 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.95138e-14 +trainer/Advantage Score Mean -0.469575 +trainer/Advantage Score Std 0.502205 +trainer/Advantage Score Max 1.65474 +trainer/Advantage Score Min -3.00444 +trainer/V1 Predictions Mean -73.4022 +trainer/V1 Predictions Std 15.3624 +trainer/V1 Predictions Max -1.10618 +trainer/V1 Predictions Min -85.6663 +trainer/VF Loss 0.0602592 +expl/num steps total 219000 +expl/num paths total 234 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.126912 +expl/Actions Std 0.824094 +expl/Actions Max 2.36752 +expl/Actions Min -2.25889 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 215259 +eval/num paths total 219 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0523442 +eval/Actions Std 0.713103 +eval/Actions Max 0.999153 +eval/Actions Min -0.997764 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.70968e-06 +time/evaluation sampling (s) 3.14026 +time/exploration sampling (s) 3.75424 +time/logging (s) 0.00763255 +time/saving (s) 0.012145 +time/training (s) 13.6107 +time/epoch (s) 20.525 +time/total (s) 4687.8 +Epoch -782 +------------------------------ ---------------- +2022-05-15 19:20:53.868415 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -781 finished +------------------------------ ---------------- +epoch -781 +replay_buffer/size 999047 +trainer/num train calls 220000 +trainer/QF1 Loss 0.726769 +trainer/QF2 Loss 0.721592 +trainer/Policy Loss 20.1442 +trainer/Q1 Predictions Mean -72.7035 +trainer/Q1 Predictions Std 15.9461 +trainer/Q1 Predictions Max -0.493461 +trainer/Q1 Predictions Min -86.5634 +trainer/Q2 Predictions Mean -72.6272 +trainer/Q2 Predictions Std 15.933 +trainer/Q2 Predictions Max -0.506348 +trainer/Q2 Predictions Min -86.4689 +trainer/Q Targets Mean -72.4648 +trainer/Q Targets Std 15.7657 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4408 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0123113 +trainer/policy/mean Std 0.714062 +trainer/policy/mean Max 0.999485 +trainer/policy/mean Min -0.998077 +trainer/policy/std Mean 0.431543 +trainer/policy/std Std 0.0214252 +trainer/policy/std Max 0.455052 +trainer/policy/std Min 0.396352 +trainer/Advantage Weights Mean 3.99863 +trainer/Advantage Weights Std 16.8522 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.8364e-14 +trainer/Advantage Score Mean -0.376165 +trainer/Advantage Score Std 0.521968 +trainer/Advantage Score Max 0.907654 +trainer/Advantage Score Min -3.03139 +trainer/V1 Predictions Mean -72.279 +trainer/V1 Predictions Std 15.8783 +trainer/V1 Predictions Max -1.18087 +trainer/V1 Predictions Min -86.3191 +trainer/VF Loss 0.056773 +expl/num steps total 220000 +expl/num paths total 235 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0435147 +expl/Actions Std 0.82691 +expl/Actions Max 2.3189 +expl/Actions Min -2.51116 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 216259 +eval/num paths total 220 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.296431 +eval/Actions Std 0.645933 +eval/Actions Max 0.999382 +eval/Actions Min -0.998248 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.99168e-06 +time/evaluation sampling (s) 3.19823 +time/exploration sampling (s) 3.63414 +time/logging (s) 0.00818527 +time/saving (s) 0.0130888 +time/training (s) 13.1088 +time/epoch (s) 19.9625 +time/total (s) 4707.77 +Epoch -781 +------------------------------ ---------------- +2022-05-15 19:21:14.379166 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -780 finished +------------------------------ ---------------- +epoch -780 +replay_buffer/size 999047 +trainer/num train calls 221000 +trainer/QF1 Loss 0.35444 +trainer/QF2 Loss 0.495381 +trainer/Policy Loss 9.65105 +trainer/Q1 Predictions Mean -73.4693 +trainer/Q1 Predictions Std 15.9198 +trainer/Q1 Predictions Max -0.383222 +trainer/Q1 Predictions Min -87.1153 +trainer/Q2 Predictions Mean -73.5518 +trainer/Q2 Predictions Std 15.9407 +trainer/Q2 Predictions Max -0.550115 +trainer/Q2 Predictions Min -86.8664 +trainer/Q Targets Mean -73.3475 +trainer/Q Targets Std 15.7863 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0312 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0408649 +trainer/policy/mean Std 0.710522 +trainer/policy/mean Max 0.9988 +trainer/policy/mean Min -0.998183 +trainer/policy/std Mean 0.431133 +trainer/policy/std Std 0.0208659 +trainer/policy/std Max 0.453992 +trainer/policy/std Min 0.397486 +trainer/Advantage Weights Mean 0.773926 +trainer/Advantage Weights Std 6.4242 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.85263e-16 +trainer/Advantage Score Mean -0.462505 +trainer/Advantage Score Std 0.479763 +trainer/Advantage Score Max 0.667793 +trainer/Advantage Score Min -3.52618 +trainer/V1 Predictions Mean -73.0211 +trainer/V1 Predictions Std 16.1437 +trainer/V1 Predictions Max 0.53102 +trainer/V1 Predictions Min -86.7402 +trainer/VF Loss 0.0469214 +expl/num steps total 221000 +expl/num paths total 236 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0623147 +expl/Actions Std 0.818948 +expl/Actions Max 2.64593 +expl/Actions Min -2.39181 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 217259 +eval/num paths total 221 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0313338 +eval/Actions Std 0.733869 +eval/Actions Max 0.999277 +eval/Actions Min -0.99956 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.7339e-06 +time/evaluation sampling (s) 3.07677 +time/exploration sampling (s) 3.2338 +time/logging (s) 0.00758311 +time/saving (s) 0.0103605 +time/training (s) 14.174 +time/epoch (s) 20.5025 +time/total (s) 4728.28 +Epoch -780 +------------------------------ ---------------- +2022-05-15 19:21:34.605006 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -779 finished +------------------------------ ---------------- +epoch -779 +replay_buffer/size 999047 +trainer/num train calls 222000 +trainer/QF1 Loss 9.2476 +trainer/QF2 Loss 9.22975 +trainer/Policy Loss 31.5772 +trainer/Q1 Predictions Mean -70.7473 +trainer/Q1 Predictions Std 19.4699 +trainer/Q1 Predictions Max -0.33929 +trainer/Q1 Predictions Min -86.2489 +trainer/Q2 Predictions Mean -70.7502 +trainer/Q2 Predictions Std 19.4612 +trainer/Q2 Predictions Max -0.645792 +trainer/Q2 Predictions Min -86.3245 +trainer/Q Targets Mean -70.9318 +trainer/Q Targets Std 19.299 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0534 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0238456 +trainer/policy/mean Std 0.720985 +trainer/policy/mean Max 0.997238 +trainer/policy/mean Min -0.998211 +trainer/policy/std Mean 0.428533 +trainer/policy/std Std 0.021632 +trainer/policy/std Max 0.451154 +trainer/policy/std Min 0.392342 +trainer/Advantage Weights Mean 5.34861 +trainer/Advantage Weights Std 19.9642 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.01088e-12 +trainer/Advantage Score Mean -0.344367 +trainer/Advantage Score Std 0.481398 +trainer/Advantage Score Max 1.2796 +trainer/Advantage Score Min -2.76202 +trainer/V1 Predictions Mean -70.8688 +trainer/V1 Predictions Std 19.0815 +trainer/V1 Predictions Max 0.181916 +trainer/V1 Predictions Min -86.4328 +trainer/VF Loss 0.0656645 +expl/num steps total 222000 +expl/num paths total 237 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0213155 +expl/Actions Std 0.815142 +expl/Actions Max 2.37084 +expl/Actions Min -2.30272 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 218259 +eval/num paths total 222 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.450953 +eval/Actions Std 0.635193 +eval/Actions Max 0.999437 +eval/Actions Min -0.998551 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.99189e-06 +time/evaluation sampling (s) 2.70269 +time/exploration sampling (s) 3.89361 +time/logging (s) 0.00750709 +time/saving (s) 0.0104713 +time/training (s) 13.6041 +time/epoch (s) 20.2184 +time/total (s) 4748.5 +Epoch -779 +------------------------------ ---------------- +2022-05-15 19:21:55.681127 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -778 finished +------------------------------ ---------------- +epoch -778 +replay_buffer/size 999047 +trainer/num train calls 223000 +trainer/QF1 Loss 0.611094 +trainer/QF2 Loss 0.643967 +trainer/Policy Loss 30.7603 +trainer/Q1 Predictions Mean -73.0711 +trainer/Q1 Predictions Std 16.2306 +trainer/Q1 Predictions Max -0.562546 +trainer/Q1 Predictions Min -86.4273 +trainer/Q2 Predictions Mean -72.9696 +trainer/Q2 Predictions Std 16.1756 +trainer/Q2 Predictions Max -0.60712 +trainer/Q2 Predictions Min -86.0703 +trainer/Q Targets Mean -73.19 +trainer/Q Targets Std 16.1601 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4837 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000837395 +trainer/policy/mean Std 0.717691 +trainer/policy/mean Max 0.999904 +trainer/policy/mean Min -0.996563 +trainer/policy/std Mean 0.430653 +trainer/policy/std Std 0.0226469 +trainer/policy/std Max 0.454761 +trainer/policy/std Min 0.394321 +trainer/Advantage Weights Mean 5.43612 +trainer/Advantage Weights Std 20.1174 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.617e-09 +trainer/Advantage Score Mean -0.299664 +trainer/Advantage Score Std 0.45476 +trainer/Advantage Score Max 1.2884 +trainer/Advantage Score Min -2.02427 +trainer/V1 Predictions Mean -72.9453 +trainer/V1 Predictions Std 16.2142 +trainer/V1 Predictions Max -1.2651 +trainer/V1 Predictions Min -86.7991 +trainer/VF Loss 0.0591287 +expl/num steps total 223000 +expl/num paths total 238 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0410012 +expl/Actions Std 0.807763 +expl/Actions Max 2.50546 +expl/Actions Min -2.41865 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 218815 +eval/num paths total 223 +eval/path length Mean 556 +eval/path length Std 0 +eval/path length Max 556 +eval/path length Min 556 +eval/Rewards Mean 0.00179856 +eval/Rewards Std 0.0423713 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0446512 +eval/Actions Std 0.729219 +eval/Actions Max 0.999671 +eval/Actions Min -0.998642 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 9.64804e-06 +time/evaluation sampling (s) 3.08808 +time/exploration sampling (s) 3.95262 +time/logging (s) 0.00605182 +time/saving (s) 0.0102759 +time/training (s) 14.0105 +time/epoch (s) 21.0676 +time/total (s) 4769.57 +Epoch -778 +------------------------------ ---------------- +2022-05-15 19:22:16.415110 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -777 finished +------------------------------ ---------------- +epoch -777 +replay_buffer/size 999047 +trainer/num train calls 224000 +trainer/QF1 Loss 0.371381 +trainer/QF2 Loss 0.441232 +trainer/Policy Loss 9.15965 +trainer/Q1 Predictions Mean -70.7152 +trainer/Q1 Predictions Std 18.9866 +trainer/Q1 Predictions Max -0.302718 +trainer/Q1 Predictions Min -86.2526 +trainer/Q2 Predictions Mean -70.7349 +trainer/Q2 Predictions Std 19.0292 +trainer/Q2 Predictions Max -0.355232 +trainer/Q2 Predictions Min -86.1743 +trainer/Q Targets Mean -70.6707 +trainer/Q Targets Std 19.1699 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0442 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0393261 +trainer/policy/mean Std 0.711225 +trainer/policy/mean Max 0.998933 +trainer/policy/mean Min -0.995868 +trainer/policy/std Mean 0.4293 +trainer/policy/std Std 0.0216926 +trainer/policy/std Max 0.454966 +trainer/policy/std Min 0.397057 +trainer/Advantage Weights Mean 1.94852 +trainer/Advantage Weights Std 7.91467 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.00212e-12 +trainer/Advantage Score Mean -0.448449 +trainer/Advantage Score Std 0.554777 +trainer/Advantage Score Max 0.465784 +trainer/Advantage Score Min -2.58389 +trainer/V1 Predictions Mean -70.3838 +trainer/V1 Predictions Std 19.2553 +trainer/V1 Predictions Max 0.982972 +trainer/V1 Predictions Min -85.9119 +trainer/VF Loss 0.0566328 +expl/num steps total 224000 +expl/num paths total 240 +expl/path length Mean 500 +expl/path length Std 395 +expl/path length Max 895 +expl/path length Min 105 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0312692 +expl/Actions Std 0.82432 +expl/Actions Max 2.45976 +expl/Actions Min -2.37316 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 219815 +eval/num paths total 224 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.146546 +eval/Actions Std 0.725505 +eval/Actions Max 0.99888 +eval/Actions Min -0.999208 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.65986e-06 +time/evaluation sampling (s) 3.08714 +time/exploration sampling (s) 4.00545 +time/logging (s) 0.0109423 +time/saving (s) 0.0154125 +time/training (s) 13.6139 +time/epoch (s) 20.7328 +time/total (s) 4790.31 +Epoch -777 +------------------------------ ---------------- +2022-05-15 19:22:36.671094 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -776 finished +------------------------------ ---------------- +epoch -776 +replay_buffer/size 999047 +trainer/num train calls 225000 +trainer/QF1 Loss 0.579683 +trainer/QF2 Loss 0.625176 +trainer/Policy Loss 15.9645 +trainer/Q1 Predictions Mean -70.6047 +trainer/Q1 Predictions Std 19.3387 +trainer/Q1 Predictions Max -1.40486 +trainer/Q1 Predictions Min -85.8391 +trainer/Q2 Predictions Mean -70.6098 +trainer/Q2 Predictions Std 19.4105 +trainer/Q2 Predictions Max -1.25777 +trainer/Q2 Predictions Min -85.9333 +trainer/Q Targets Mean -70.5495 +trainer/Q Targets Std 19.0645 +trainer/Q Targets Max -1.62093 +trainer/Q Targets Min -85.6265 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00794804 +trainer/policy/mean Std 0.720781 +trainer/policy/mean Max 0.998827 +trainer/policy/mean Min -0.999128 +trainer/policy/std Mean 0.43 +trainer/policy/std Std 0.0216505 +trainer/policy/std Max 0.451944 +trainer/policy/std Min 0.397218 +trainer/Advantage Weights Mean 4.2882 +trainer/Advantage Weights Std 18.6436 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.42016e-12 +trainer/Advantage Score Mean -0.422432 +trainer/Advantage Score Std 0.447151 +trainer/Advantage Score Max 1.07193 +trainer/Advantage Score Min -2.72803 +trainer/V1 Predictions Mean -70.2919 +trainer/V1 Predictions Std 19.1159 +trainer/V1 Predictions Max -2.28103 +trainer/V1 Predictions Min -85.8263 +trainer/VF Loss 0.0558538 +expl/num steps total 225000 +expl/num paths total 241 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.237744 +expl/Actions Std 0.87232 +expl/Actions Max 2.34503 +expl/Actions Min -2.52708 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 220815 +eval/num paths total 225 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.27624 +eval/Actions Std 0.593163 +eval/Actions Max 0.998985 +eval/Actions Min -0.997941 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.84985e-06 +time/evaluation sampling (s) 3.24201 +time/exploration sampling (s) 3.56965 +time/logging (s) 0.00681913 +time/saving (s) 0.00934377 +time/training (s) 13.4129 +time/epoch (s) 20.2408 +time/total (s) 4810.56 +Epoch -776 +------------------------------ ---------------- +2022-05-15 19:22:56.706740 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -775 finished +------------------------------ ---------------- +epoch -775 +replay_buffer/size 999047 +trainer/num train calls 226000 +trainer/QF1 Loss 11.8407 +trainer/QF2 Loss 12.3496 +trainer/Policy Loss 39.4859 +trainer/Q1 Predictions Mean -71.7546 +trainer/Q1 Predictions Std 18.3652 +trainer/Q1 Predictions Max -0.919322 +trainer/Q1 Predictions Min -86.1999 +trainer/Q2 Predictions Mean -71.7836 +trainer/Q2 Predictions Std 18.4086 +trainer/Q2 Predictions Max -0.562539 +trainer/Q2 Predictions Min -86.0687 +trainer/Q Targets Mean -71.889 +trainer/Q Targets Std 18.3378 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.1564 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000131055 +trainer/policy/mean Std 0.69054 +trainer/policy/mean Max 0.997789 +trainer/policy/mean Min -0.996493 +trainer/policy/std Mean 0.431168 +trainer/policy/std Std 0.0215637 +trainer/policy/std Max 0.455208 +trainer/policy/std Min 0.396196 +trainer/Advantage Weights Mean 6.91488 +trainer/Advantage Weights Std 22.1848 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.46762e-13 +trainer/Advantage Score Mean -0.288046 +trainer/Advantage Score Std 0.492841 +trainer/Advantage Score Max 1.30444 +trainer/Advantage Score Min -2.955 +trainer/V1 Predictions Mean -71.4896 +trainer/V1 Predictions Std 18.3189 +trainer/V1 Predictions Max -0.510117 +trainer/V1 Predictions Min -85.8665 +trainer/VF Loss 0.0590489 +expl/num steps total 226000 +expl/num paths total 242 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0453296 +expl/Actions Std 0.844992 +expl/Actions Max 2.44913 +expl/Actions Min -2.21723 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 221815 +eval/num paths total 226 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.163302 +eval/Actions Std 0.635132 +eval/Actions Max 0.998831 +eval/Actions Min -0.998861 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.75113e-06 +time/evaluation sampling (s) 2.96378 +time/exploration sampling (s) 3.2521 +time/logging (s) 0.00948398 +time/saving (s) 0.00951357 +time/training (s) 13.7973 +time/epoch (s) 20.0322 +time/total (s) 4830.59 +Epoch -775 +------------------------------ ---------------- +2022-05-15 19:23:16.798852 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -774 finished +------------------------------ ---------------- +epoch -774 +replay_buffer/size 999047 +trainer/num train calls 227000 +trainer/QF1 Loss 1.06557 +trainer/QF2 Loss 0.888492 +trainer/Policy Loss 13.9621 +trainer/Q1 Predictions Mean -72.2235 +trainer/Q1 Predictions Std 17.315 +trainer/Q1 Predictions Max -2.20845 +trainer/Q1 Predictions Min -86.0541 +trainer/Q2 Predictions Mean -72.219 +trainer/Q2 Predictions Std 17.3305 +trainer/Q2 Predictions Max -1.86746 +trainer/Q2 Predictions Min -86.3898 +trainer/Q Targets Mean -72.4685 +trainer/Q Targets Std 17.0348 +trainer/Q Targets Max -4.61972 +trainer/Q Targets Min -86.2185 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00713538 +trainer/policy/mean Std 0.716321 +trainer/policy/mean Max 0.999497 +trainer/policy/mean Min -0.998204 +trainer/policy/std Mean 0.428903 +trainer/policy/std Std 0.0215009 +trainer/policy/std Max 0.45253 +trainer/policy/std Min 0.393468 +trainer/Advantage Weights Mean 2.83804 +trainer/Advantage Weights Std 13.9718 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.87251e-11 +trainer/Advantage Score Mean -0.393937 +trainer/Advantage Score Std 0.443043 +trainer/Advantage Score Max 1.01503 +trainer/Advantage Score Min -2.39745 +trainer/V1 Predictions Mean -72.1753 +trainer/V1 Predictions Std 17.2512 +trainer/V1 Predictions Max -3.55032 +trainer/V1 Predictions Min -86.1684 +trainer/VF Loss 0.0472968 +expl/num steps total 227000 +expl/num paths total 244 +expl/path length Mean 500 +expl/path length Std 377 +expl/path length Max 877 +expl/path length Min 123 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0411317 +expl/Actions Std 0.832155 +expl/Actions Max 2.17877 +expl/Actions Min -2.17488 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 222815 +eval/num paths total 227 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0643059 +eval/Actions Std 0.741016 +eval/Actions Max 0.998729 +eval/Actions Min -0.998752 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.76696e-06 +time/evaluation sampling (s) 2.67109 +time/exploration sampling (s) 3.74564 +time/logging (s) 0.00892391 +time/saving (s) 0.0148095 +time/training (s) 13.6443 +time/epoch (s) 20.0848 +time/total (s) 4850.68 +Epoch -774 +------------------------------ ---------------- +2022-05-15 19:23:36.499604 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -773 finished +------------------------------ ---------------- +epoch -773 +replay_buffer/size 999047 +trainer/num train calls 228000 +trainer/QF1 Loss 1.693 +trainer/QF2 Loss 1.84311 +trainer/Policy Loss 0.0537131 +trainer/Q1 Predictions Mean -71.7966 +trainer/Q1 Predictions Std 20.0578 +trainer/Q1 Predictions Max -0.282376 +trainer/Q1 Predictions Min -87.0555 +trainer/Q2 Predictions Mean -71.8941 +trainer/Q2 Predictions Std 20.0434 +trainer/Q2 Predictions Max -0.32515 +trainer/Q2 Predictions Min -87.3545 +trainer/Q Targets Mean -70.927 +trainer/Q Targets Std 20.0344 +trainer/Q Targets Max 0.649682 +trainer/Q Targets Min -86.1875 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00672589 +trainer/policy/mean Std 0.720129 +trainer/policy/mean Max 0.999501 +trainer/policy/mean Min -0.99871 +trainer/policy/std Mean 0.428053 +trainer/policy/std Std 0.0213125 +trainer/policy/std Max 0.455977 +trainer/policy/std Min 0.392144 +trainer/Advantage Weights Mean 0.0164289 +trainer/Advantage Weights Std 0.0860328 +trainer/Advantage Weights Max 1.16413 +trainer/Advantage Weights Min 3.95418e-19 +trainer/Advantage Score Mean -0.932772 +trainer/Advantage Score Std 0.608162 +trainer/Advantage Score Max 0.0151978 +trainer/Advantage Score Min -4.23743 +trainer/V1 Predictions Mean -70.6522 +trainer/V1 Predictions Std 20.2991 +trainer/V1 Predictions Max 2.50778 +trainer/V1 Predictions Min -86.0503 +trainer/VF Loss 0.123993 +expl/num steps total 228000 +expl/num paths total 246 +expl/path length Mean 500 +expl/path length Std 420 +expl/path length Max 920 +expl/path length Min 80 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0568151 +expl/Actions Std 0.836007 +expl/Actions Max 2.6521 +expl/Actions Min -2.38337 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 223463 +eval/num paths total 228 +eval/path length Mean 648 +eval/path length Std 0 +eval/path length Max 648 +eval/path length Min 648 +eval/Rewards Mean 0.00154321 +eval/Rewards Std 0.0392534 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0528217 +eval/Actions Std 0.756077 +eval/Actions Max 0.999305 +eval/Actions Min -0.997656 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.1041e-06 +time/evaluation sampling (s) 3.16881 +time/exploration sampling (s) 3.95459 +time/logging (s) 0.0104459 +time/saving (s) 0.0178507 +time/training (s) 12.5416 +time/epoch (s) 19.6933 +time/total (s) 4870.38 +Epoch -773 +------------------------------ ---------------- +2022-05-15 19:23:57.173018 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -772 finished +------------------------------ ---------------- +epoch -772 +replay_buffer/size 999047 +trainer/num train calls 229000 +trainer/QF1 Loss 0.686295 +trainer/QF2 Loss 0.64988 +trainer/Policy Loss 13.1697 +trainer/Q1 Predictions Mean -70.8252 +trainer/Q1 Predictions Std 19.1577 +trainer/Q1 Predictions Max -1.09082 +trainer/Q1 Predictions Min -86.0717 +trainer/Q2 Predictions Mean -70.8144 +trainer/Q2 Predictions Std 19.2274 +trainer/Q2 Predictions Max -0.837521 +trainer/Q2 Predictions Min -85.8943 +trainer/Q Targets Mean -70.7066 +trainer/Q Targets Std 19.5024 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4431 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0169215 +trainer/policy/mean Std 0.716606 +trainer/policy/mean Max 0.999369 +trainer/policy/mean Min -0.999806 +trainer/policy/std Mean 0.428869 +trainer/policy/std Std 0.0210404 +trainer/policy/std Max 0.453856 +trainer/policy/std Min 0.392108 +trainer/Advantage Weights Mean 2.31964 +trainer/Advantage Weights Std 12.7541 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.83247e-15 +trainer/Advantage Score Mean -0.462662 +trainer/Advantage Score Std 0.525244 +trainer/Advantage Score Max 1.07925 +trainer/Advantage Score Min -3.31953 +trainer/V1 Predictions Mean -70.4473 +trainer/V1 Predictions Std 19.5301 +trainer/V1 Predictions Max 0.319902 +trainer/V1 Predictions Min -86.0435 +trainer/VF Loss 0.0590923 +expl/num steps total 229000 +expl/num paths total 247 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.220697 +expl/Actions Std 0.860421 +expl/Actions Max 2.49496 +expl/Actions Min -2.46002 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 224463 +eval/num paths total 229 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0370144 +eval/Actions Std 0.732589 +eval/Actions Max 0.999353 +eval/Actions Min -0.998201 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.70596e-06 +time/evaluation sampling (s) 3.10356 +time/exploration sampling (s) 4.04974 +time/logging (s) 0.00699139 +time/saving (s) 0.0107138 +time/training (s) 13.4886 +time/epoch (s) 20.6596 +time/total (s) 4891.05 +Epoch -772 +------------------------------ ---------------- +2022-05-15 19:24:17.022486 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -771 finished +------------------------------ ---------------- +epoch -771 +replay_buffer/size 999047 +trainer/num train calls 230000 +trainer/QF1 Loss 0.603044 +trainer/QF2 Loss 0.613681 +trainer/Policy Loss 7.13753 +trainer/Q1 Predictions Mean -71.6409 +trainer/Q1 Predictions Std 17.6305 +trainer/Q1 Predictions Max -1.81613 +trainer/Q1 Predictions Min -86.0603 +trainer/Q2 Predictions Mean -71.635 +trainer/Q2 Predictions Std 17.6925 +trainer/Q2 Predictions Max -1.93794 +trainer/Q2 Predictions Min -86.0505 +trainer/Q Targets Mean -71.5 +trainer/Q Targets Std 17.6091 +trainer/Q Targets Max -3.51692 +trainer/Q Targets Min -85.9852 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0264109 +trainer/policy/mean Std 0.69776 +trainer/policy/mean Max 0.999737 +trainer/policy/mean Min -0.999712 +trainer/policy/std Mean 0.429794 +trainer/policy/std Std 0.0216829 +trainer/policy/std Max 0.45223 +trainer/policy/std Min 0.392081 +trainer/Advantage Weights Mean 1.3033 +trainer/Advantage Weights Std 8.50842 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.88087e-15 +trainer/Advantage Score Mean -0.461078 +trainer/Advantage Score Std 0.535536 +trainer/Advantage Score Max 1.16376 +trainer/Advantage Score Min -3.31827 +trainer/V1 Predictions Mean -71.2076 +trainer/V1 Predictions Std 17.9013 +trainer/V1 Predictions Max -1.06384 +trainer/V1 Predictions Min -85.8977 +trainer/VF Loss 0.056531 +expl/num steps total 230000 +expl/num paths total 249 +expl/path length Mean 500 +expl/path length Std 452 +expl/path length Max 952 +expl/path length Min 48 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0475353 +expl/Actions Std 0.818612 +expl/Actions Max 2.42003 +expl/Actions Min -2.26896 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 225463 +eval/num paths total 230 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.243497 +eval/Actions Std 0.628718 +eval/Actions Max 0.999401 +eval/Actions Min -0.997458 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.96161e-06 +time/evaluation sampling (s) 3.06763 +time/exploration sampling (s) 3.83819 +time/logging (s) 0.00738708 +time/saving (s) 0.0100428 +time/training (s) 12.9202 +time/epoch (s) 19.8435 +time/total (s) 4910.89 +Epoch -771 +------------------------------ ---------------- +2022-05-15 19:24:37.052189 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -770 finished +------------------------------ ---------------- +epoch -770 +replay_buffer/size 999047 +trainer/num train calls 231000 +trainer/QF1 Loss 1.02058 +trainer/QF2 Loss 1.01775 +trainer/Policy Loss 43.2438 +trainer/Q1 Predictions Mean -70.3599 +trainer/Q1 Predictions Std 19.3336 +trainer/Q1 Predictions Max -0.79157 +trainer/Q1 Predictions Min -86.1275 +trainer/Q2 Predictions Mean -70.3319 +trainer/Q2 Predictions Std 19.3427 +trainer/Q2 Predictions Max -0.710712 +trainer/Q2 Predictions Min -86.4152 +trainer/Q Targets Mean -70.7986 +trainer/Q Targets Std 19.5008 +trainer/Q Targets Max 0.566683 +trainer/Q Targets Min -86.2221 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0205007 +trainer/policy/mean Std 0.713183 +trainer/policy/mean Max 0.999065 +trainer/policy/mean Min -0.997435 +trainer/policy/std Mean 0.429443 +trainer/policy/std Std 0.0209754 +trainer/policy/std Max 0.453413 +trainer/policy/std Min 0.392259 +trainer/Advantage Weights Mean 9.5746 +trainer/Advantage Weights Std 26.0312 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.57075e-14 +trainer/Advantage Score Mean -0.243312 +trainer/Advantage Score Std 0.579236 +trainer/Advantage Score Max 2.18989 +trainer/Advantage Score Min -3.07165 +trainer/V1 Predictions Mean -70.6107 +trainer/V1 Predictions Std 19.3958 +trainer/V1 Predictions Max 0.801827 +trainer/V1 Predictions Min -85.9361 +trainer/VF Loss 0.091969 +expl/num steps total 231000 +expl/num paths total 250 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0229995 +expl/Actions Std 0.81935 +expl/Actions Max 2.55098 +expl/Actions Min -2.44152 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 226131 +eval/num paths total 231 +eval/path length Mean 668 +eval/path length Std 0 +eval/path length Max 668 +eval/path length Min 668 +eval/Rewards Mean 0.00149701 +eval/Rewards Std 0.0386622 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.000522263 +eval/Actions Std 0.720768 +eval/Actions Max 0.998822 +eval/Actions Min -0.998917 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.54577e-06 +time/evaluation sampling (s) 3.02542 +time/exploration sampling (s) 3.35911 +time/logging (s) 0.00702249 +time/saving (s) 0.0101054 +time/training (s) 13.6209 +time/epoch (s) 20.0226 +time/total (s) 4930.92 +Epoch -770 +------------------------------ ---------------- +2022-05-15 19:24:56.660812 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -769 finished +------------------------------ ---------------- +epoch -769 +replay_buffer/size 999047 +trainer/num train calls 232000 +trainer/QF1 Loss 0.809759 +trainer/QF2 Loss 0.869754 +trainer/Policy Loss 20.8866 +trainer/Q1 Predictions Mean -69.0486 +trainer/Q1 Predictions Std 20.0892 +trainer/Q1 Predictions Max -1.44355 +trainer/Q1 Predictions Min -85.96 +trainer/Q2 Predictions Mean -69.0769 +trainer/Q2 Predictions Std 20.0904 +trainer/Q2 Predictions Max -1.6096 +trainer/Q2 Predictions Min -86.0687 +trainer/Q Targets Mean -69.2508 +trainer/Q Targets Std 20.1195 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0701 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0100722 +trainer/policy/mean Std 0.717665 +trainer/policy/mean Max 0.999361 +trainer/policy/mean Min -0.997778 +trainer/policy/std Mean 0.43015 +trainer/policy/std Std 0.0216038 +trainer/policy/std Max 0.454161 +trainer/policy/std Min 0.39384 +trainer/Advantage Weights Mean 5.49454 +trainer/Advantage Weights Std 19.0072 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.87727e-19 +trainer/Advantage Score Mean -0.307717 +trainer/Advantage Score Std 0.619091 +trainer/Advantage Score Max 1.21653 +trainer/Advantage Score Min -4.31193 +trainer/V1 Predictions Mean -68.9799 +trainer/V1 Predictions Std 20.2249 +trainer/V1 Predictions Max -1.52878 +trainer/V1 Predictions Min -85.9898 +trainer/VF Loss 0.0711103 +expl/num steps total 232000 +expl/num paths total 251 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0383958 +expl/Actions Std 0.819828 +expl/Actions Max 2.52308 +expl/Actions Min -2.60604 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 226886 +eval/num paths total 232 +eval/path length Mean 755 +eval/path length Std 0 +eval/path length Max 755 +eval/path length Min 755 +eval/Rewards Mean 0.0013245 +eval/Rewards Std 0.0363696 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0305682 +eval/Actions Std 0.734969 +eval/Actions Max 0.999543 +eval/Actions Min -0.998174 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.6999e-06 +time/evaluation sampling (s) 2.70258 +time/exploration sampling (s) 3.43417 +time/logging (s) 0.00629946 +time/saving (s) 0.00933416 +time/training (s) 13.4491 +time/epoch (s) 19.6014 +time/total (s) 4950.53 +Epoch -769 +------------------------------ ---------------- +2022-05-15 19:25:16.921827 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -768 finished +------------------------------ ---------------- +epoch -768 +replay_buffer/size 999047 +trainer/num train calls 233000 +trainer/QF1 Loss 0.979134 +trainer/QF2 Loss 0.900466 +trainer/Policy Loss 13.1715 +trainer/Q1 Predictions Mean -71.6861 +trainer/Q1 Predictions Std 19.0135 +trainer/Q1 Predictions Max -0.446559 +trainer/Q1 Predictions Min -85.9085 +trainer/Q2 Predictions Mean -71.675 +trainer/Q2 Predictions Std 19.123 +trainer/Q2 Predictions Max -0.414266 +trainer/Q2 Predictions Min -85.8892 +trainer/Q Targets Mean -71.5003 +trainer/Q Targets Std 19.1914 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9302 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00582956 +trainer/policy/mean Std 0.713816 +trainer/policy/mean Max 0.998989 +trainer/policy/mean Min -0.998756 +trainer/policy/std Mean 0.429861 +trainer/policy/std Std 0.0215689 +trainer/policy/std Max 0.455913 +trainer/policy/std Min 0.393637 +trainer/Advantage Weights Mean 3.28582 +trainer/Advantage Weights Std 15.9661 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.96564e-16 +trainer/Advantage Score Mean -0.458121 +trainer/Advantage Score Std 0.583165 +trainer/Advantage Score Max 2.32308 +trainer/Advantage Score Min -3.54637 +trainer/V1 Predictions Mean -71.2487 +trainer/V1 Predictions Std 19.223 +trainer/V1 Predictions Max 0.26409 +trainer/V1 Predictions Min -85.8738 +trainer/VF Loss 0.0840611 +expl/num steps total 233000 +expl/num paths total 252 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.262796 +expl/Actions Std 0.905244 +expl/Actions Max 2.45376 +expl/Actions Min -2.53749 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 227596 +eval/num paths total 233 +eval/path length Mean 710 +eval/path length Std 0 +eval/path length Max 710 +eval/path length Min 710 +eval/Rewards Mean 0.00140845 +eval/Rewards Std 0.0375029 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0282357 +eval/Actions Std 0.719613 +eval/Actions Max 0.998515 +eval/Actions Min -0.997245 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.69618e-06 +time/evaluation sampling (s) 2.52975 +time/exploration sampling (s) 3.99143 +time/logging (s) 0.00766089 +time/saving (s) 0.0146035 +time/training (s) 13.713 +time/epoch (s) 20.2565 +time/total (s) 4970.79 +Epoch -768 +------------------------------ ---------------- +2022-05-15 19:25:37.909063 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -767 finished +------------------------------ ---------------- +epoch -767 +replay_buffer/size 999047 +trainer/num train calls 234000 +trainer/QF1 Loss 0.89802 +trainer/QF2 Loss 1.01704 +trainer/Policy Loss 1.15698 +trainer/Q1 Predictions Mean -71.7767 +trainer/Q1 Predictions Std 16.3036 +trainer/Q1 Predictions Max -0.697783 +trainer/Q1 Predictions Min -86.5814 +trainer/Q2 Predictions Mean -71.721 +trainer/Q2 Predictions Std 16.1996 +trainer/Q2 Predictions Max -0.480446 +trainer/Q2 Predictions Min -86.6274 +trainer/Q Targets Mean -71.663 +trainer/Q Targets Std 16.8727 +trainer/Q Targets Max 0.438189 +trainer/Q Targets Min -86.1125 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0295178 +trainer/policy/mean Std 0.711777 +trainer/policy/mean Max 0.999533 +trainer/policy/mean Min -0.997152 +trainer/policy/std Mean 0.430368 +trainer/policy/std Std 0.0220514 +trainer/policy/std Max 0.456041 +trainer/policy/std Min 0.395057 +trainer/Advantage Weights Mean 0.261446 +trainer/Advantage Weights Std 1.10277 +trainer/Advantage Weights Max 15.8193 +trainer/Advantage Weights Min 7.09078e-20 +trainer/Advantage Score Mean -0.718847 +trainer/Advantage Score Std 0.746656 +trainer/Advantage Score Max 0.276123 +trainer/Advantage Score Min -4.40929 +trainer/V1 Predictions Mean -71.3809 +trainer/V1 Predictions Std 16.9487 +trainer/V1 Predictions Max 1.95636 +trainer/V1 Predictions Min -86.1855 +trainer/VF Loss 0.107901 +expl/num steps total 234000 +expl/num paths total 253 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0232981 +expl/Actions Std 0.817753 +expl/Actions Max 2.43435 +expl/Actions Min -2.15973 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 228596 +eval/num paths total 234 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0600224 +eval/Actions Std 0.640432 +eval/Actions Max 0.999582 +eval/Actions Min -0.999735 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.84872e-06 +time/evaluation sampling (s) 3.08107 +time/exploration sampling (s) 3.93801 +time/logging (s) 0.00714211 +time/saving (s) 0.0101317 +time/training (s) 13.9418 +time/epoch (s) 20.9781 +time/total (s) 4991.77 +Epoch -767 +------------------------------ ---------------- +2022-05-15 19:25:58.371526 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -766 finished +------------------------------ ---------------- +epoch -766 +replay_buffer/size 999047 +trainer/num train calls 235000 +trainer/QF1 Loss 0.64437 +trainer/QF2 Loss 0.702803 +trainer/Policy Loss 26.1219 +trainer/Q1 Predictions Mean -71.8923 +trainer/Q1 Predictions Std 19.1825 +trainer/Q1 Predictions Max -0.578816 +trainer/Q1 Predictions Min -86.6115 +trainer/Q2 Predictions Mean -71.8872 +trainer/Q2 Predictions Std 19.1044 +trainer/Q2 Predictions Max -0.669751 +trainer/Q2 Predictions Min -86.4463 +trainer/Q Targets Mean -71.7911 +trainer/Q Targets Std 19.0139 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.235 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0223177 +trainer/policy/mean Std 0.705119 +trainer/policy/mean Max 0.998043 +trainer/policy/mean Min -0.99958 +trainer/policy/std Mean 0.430629 +trainer/policy/std Std 0.0218085 +trainer/policy/std Max 0.456169 +trainer/policy/std Min 0.392788 +trainer/Advantage Weights Mean 6.79156 +trainer/Advantage Weights Std 19.073 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.70965e-13 +trainer/Advantage Score Mean -0.161603 +trainer/Advantage Score Std 0.474967 +trainer/Advantage Score Max 0.765648 +trainer/Advantage Score Min -2.93973 +trainer/V1 Predictions Mean -71.5801 +trainer/V1 Predictions Std 18.9836 +trainer/V1 Predictions Max -1.31414 +trainer/V1 Predictions Min -86.1045 +trainer/VF Loss 0.0455482 +expl/num steps total 235000 +expl/num paths total 254 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0535155 +expl/Actions Std 0.836558 +expl/Actions Max 2.48379 +expl/Actions Min -2.30065 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 229596 +eval/num paths total 235 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0355319 +eval/Actions Std 0.734218 +eval/Actions Max 0.99938 +eval/Actions Min -0.998789 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.37394e-06 +time/evaluation sampling (s) 3.0387 +time/exploration sampling (s) 3.85994 +time/logging (s) 0.00706784 +time/saving (s) 0.0108484 +time/training (s) 13.5395 +time/epoch (s) 20.456 +time/total (s) 5012.23 +Epoch -766 +------------------------------ ---------------- +2022-05-15 19:26:19.237710 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -765 finished +------------------------------ ---------------- +epoch -765 +replay_buffer/size 999047 +trainer/num train calls 236000 +trainer/QF1 Loss 0.868417 +trainer/QF2 Loss 0.967265 +trainer/Policy Loss 11.8234 +trainer/Q1 Predictions Mean -71.3326 +trainer/Q1 Predictions Std 19.7454 +trainer/Q1 Predictions Max -0.997598 +trainer/Q1 Predictions Min -87.1116 +trainer/Q2 Predictions Mean -71.3148 +trainer/Q2 Predictions Std 19.8435 +trainer/Q2 Predictions Max -0.763647 +trainer/Q2 Predictions Min -87.1546 +trainer/Q Targets Mean -71.182 +trainer/Q Targets Std 20.0914 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5699 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0419874 +trainer/policy/mean Std 0.723548 +trainer/policy/mean Max 0.998881 +trainer/policy/mean Min -0.997602 +trainer/policy/std Mean 0.431753 +trainer/policy/std Std 0.0214415 +trainer/policy/std Max 0.457549 +trainer/policy/std Min 0.39358 +trainer/Advantage Weights Mean 2.3637 +trainer/Advantage Weights Std 14.7865 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.56679e-16 +trainer/Advantage Score Mean -0.786203 +trainer/Advantage Score Std 0.563455 +trainer/Advantage Score Max 0.912316 +trainer/Advantage Score Min -3.63923 +trainer/V1 Predictions Mean -71.0147 +trainer/V1 Predictions Std 19.8584 +trainer/V1 Predictions Max 0.250696 +trainer/V1 Predictions Min -86.6665 +trainer/VF Loss 0.100885 +expl/num steps total 236000 +expl/num paths total 255 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.209672 +expl/Actions Std 0.861336 +expl/Actions Max 2.4069 +expl/Actions Min -2.4699 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 230298 +eval/num paths total 236 +eval/path length Mean 702 +eval/path length Std 0 +eval/path length Max 702 +eval/path length Min 702 +eval/Rewards Mean 0.0014245 +eval/Rewards Std 0.0377157 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.052322 +eval/Actions Std 0.722725 +eval/Actions Max 0.999666 +eval/Actions Min -0.999393 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.0306e-05 +time/evaluation sampling (s) 3.15882 +time/exploration sampling (s) 4.04397 +time/logging (s) 0.0115748 +time/saving (s) 0.0215768 +time/training (s) 13.6283 +time/epoch (s) 20.8643 +time/total (s) 5033.1 +Epoch -765 +------------------------------ ---------------- +2022-05-15 19:26:39.620498 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -764 finished +------------------------------ ---------------- +epoch -764 +replay_buffer/size 999047 +trainer/num train calls 237000 +trainer/QF1 Loss 0.400415 +trainer/QF2 Loss 0.43381 +trainer/Policy Loss 4.78123 +trainer/Q1 Predictions Mean -72.2958 +trainer/Q1 Predictions Std 16.654 +trainer/Q1 Predictions Max -0.710957 +trainer/Q1 Predictions Min -85.8328 +trainer/Q2 Predictions Mean -72.1984 +trainer/Q2 Predictions Std 16.6295 +trainer/Q2 Predictions Max -0.760337 +trainer/Q2 Predictions Min -85.805 +trainer/Q Targets Mean -72.3794 +trainer/Q Targets Std 16.7325 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7326 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0135796 +trainer/policy/mean Std 0.714543 +trainer/policy/mean Max 0.996332 +trainer/policy/mean Min -0.997938 +trainer/policy/std Mean 0.4316 +trainer/policy/std Std 0.0210963 +trainer/policy/std Max 0.456202 +trainer/policy/std Min 0.395057 +trainer/Advantage Weights Mean 1.0757 +trainer/Advantage Weights Std 5.47458 +trainer/Advantage Weights Max 77.9985 +trainer/Advantage Weights Min 3.08851e-14 +trainer/Advantage Score Mean -0.378081 +trainer/Advantage Score Std 0.528926 +trainer/Advantage Score Max 0.435669 +trainer/Advantage Score Min -3.11085 +trainer/V1 Predictions Mean -72.0577 +trainer/V1 Predictions Std 17.0329 +trainer/V1 Predictions Max 1.14413 +trainer/V1 Predictions Min -85.6039 +trainer/VF Loss 0.0448967 +expl/num steps total 237000 +expl/num paths total 256 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0198867 +expl/Actions Std 0.814854 +expl/Actions Max 2.24229 +expl/Actions Min -2.36093 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 231298 +eval/num paths total 237 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.174244 +eval/Actions Std 0.7108 +eval/Actions Max 0.998827 +eval/Actions Min -0.998152 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.6999e-06 +time/evaluation sampling (s) 3.01147 +time/exploration sampling (s) 3.27166 +time/logging (s) 0.0082999 +time/saving (s) 0.0207344 +time/training (s) 14.0605 +time/epoch (s) 20.3727 +time/total (s) 5053.47 +Epoch -764 +------------------------------ ---------------- +2022-05-15 19:26:59.224690 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -763 finished +------------------------------ ---------------- +epoch -763 +replay_buffer/size 999047 +trainer/num train calls 238000 +trainer/QF1 Loss 0.5307 +trainer/QF2 Loss 0.557816 +trainer/Policy Loss 16.3659 +trainer/Q1 Predictions Mean -71.2855 +trainer/Q1 Predictions Std 18.5777 +trainer/Q1 Predictions Max -0.306201 +trainer/Q1 Predictions Min -85.5948 +trainer/Q2 Predictions Mean -71.3118 +trainer/Q2 Predictions Std 18.6703 +trainer/Q2 Predictions Max -0.39257 +trainer/Q2 Predictions Min -85.7708 +trainer/Q Targets Mean -71.3908 +trainer/Q Targets Std 18.5626 +trainer/Q Targets Max 0.507082 +trainer/Q Targets Min -85.899 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0235126 +trainer/policy/mean Std 0.714463 +trainer/policy/mean Max 0.998233 +trainer/policy/mean Min -0.998164 +trainer/policy/std Mean 0.4285 +trainer/policy/std Std 0.0224885 +trainer/policy/std Max 0.45239 +trainer/policy/std Min 0.390058 +trainer/Advantage Weights Mean 4.97594 +trainer/Advantage Weights Std 18.9228 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.03245e-15 +trainer/Advantage Score Mean -0.319108 +trainer/Advantage Score Std 0.516843 +trainer/Advantage Score Max 1.4085 +trainer/Advantage Score Min -3.27416 +trainer/V1 Predictions Mean -71.1114 +trainer/V1 Predictions Std 18.6563 +trainer/V1 Predictions Max 0.603735 +trainer/V1 Predictions Min -85.7569 +trainer/VF Loss 0.0636997 +expl/num steps total 238000 +expl/num paths total 257 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0613914 +expl/Actions Std 0.806232 +expl/Actions Max 2.69889 +expl/Actions Min -2.35975 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 231936 +eval/num paths total 238 +eval/path length Mean 638 +eval/path length Std 0 +eval/path length Max 638 +eval/path length Min 638 +eval/Rewards Mean 0.0015674 +eval/Rewards Std 0.0395593 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.073134 +eval/Actions Std 0.700941 +eval/Actions Max 0.99904 +eval/Actions Min -0.999249 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.50386e-06 +time/evaluation sampling (s) 2.74767 +time/exploration sampling (s) 3.73449 +time/logging (s) 0.00836638 +time/saving (s) 0.0139775 +time/training (s) 13.0917 +time/epoch (s) 19.5962 +time/total (s) 5073.08 +Epoch -763 +------------------------------ ---------------- +2022-05-15 19:27:20.008583 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -762 finished +------------------------------ ---------------- +epoch -762 +replay_buffer/size 999047 +trainer/num train calls 239000 +trainer/QF1 Loss 0.545514 +trainer/QF2 Loss 0.628499 +trainer/Policy Loss 10.9238 +trainer/Q1 Predictions Mean -71.9052 +trainer/Q1 Predictions Std 19.1654 +trainer/Q1 Predictions Max -3.06901 +trainer/Q1 Predictions Min -86.2982 +trainer/Q2 Predictions Mean -71.9197 +trainer/Q2 Predictions Std 19.1864 +trainer/Q2 Predictions Max -2.27034 +trainer/Q2 Predictions Min -86.5549 +trainer/Q Targets Mean -71.5393 +trainer/Q Targets Std 19.2361 +trainer/Q Targets Max -3.29767 +trainer/Q Targets Min -86.4866 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000989973 +trainer/policy/mean Std 0.720043 +trainer/policy/mean Max 0.998231 +trainer/policy/mean Min -0.993849 +trainer/policy/std Mean 0.427651 +trainer/policy/std Std 0.0210051 +trainer/policy/std Max 0.450037 +trainer/policy/std Min 0.391651 +trainer/Advantage Weights Mean 2.69826 +trainer/Advantage Weights Std 12.053 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.69513e-16 +trainer/Advantage Score Mean -0.385314 +trainer/Advantage Score Std 0.582433 +trainer/Advantage Score Max 0.55616 +trainer/Advantage Score Min -3.58499 +trainer/V1 Predictions Mean -71.2406 +trainer/V1 Predictions Std 19.3934 +trainer/V1 Predictions Max -1.10524 +trainer/V1 Predictions Min -86.2911 +trainer/VF Loss 0.0556872 +expl/num steps total 239000 +expl/num paths total 258 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0342317 +expl/Actions Std 0.839082 +expl/Actions Max 2.41639 +expl/Actions Min -2.21279 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 232936 +eval/num paths total 239 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.031682 +eval/Actions Std 0.743348 +eval/Actions Max 0.999427 +eval/Actions Min -0.99797 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.06127e-06 +time/evaluation sampling (s) 3.08608 +time/exploration sampling (s) 3.97366 +time/logging (s) 0.00731283 +time/saving (s) 0.0111619 +time/training (s) 13.6966 +time/epoch (s) 20.7748 +time/total (s) 5093.85 +Epoch -762 +------------------------------ ---------------- +2022-05-15 19:27:41.194719 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -761 finished +------------------------------ ---------------- +epoch -761 +replay_buffer/size 999047 +trainer/num train calls 240000 +trainer/QF1 Loss 0.49537 +trainer/QF2 Loss 0.444836 +trainer/Policy Loss 10.3082 +trainer/Q1 Predictions Mean -72.6934 +trainer/Q1 Predictions Std 18.0093 +trainer/Q1 Predictions Max -0.687463 +trainer/Q1 Predictions Min -86.3946 +trainer/Q2 Predictions Mean -72.7348 +trainer/Q2 Predictions Std 17.9251 +trainer/Q2 Predictions Max -0.343413 +trainer/Q2 Predictions Min -86.6478 +trainer/Q Targets Mean -72.5951 +trainer/Q Targets Std 18.0254 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9151 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00358147 +trainer/policy/mean Std 0.703143 +trainer/policy/mean Max 0.998887 +trainer/policy/mean Min -0.998036 +trainer/policy/std Mean 0.429073 +trainer/policy/std Std 0.0222758 +trainer/policy/std Max 0.4531 +trainer/policy/std Min 0.392395 +trainer/Advantage Weights Mean 2.94118 +trainer/Advantage Weights Std 13.9768 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.91539e-14 +trainer/Advantage Score Mean -0.407838 +trainer/Advantage Score Std 0.585667 +trainer/Advantage Score Max 1.5012 +trainer/Advantage Score Min -3.08713 +trainer/V1 Predictions Mean -72.3406 +trainer/V1 Predictions Std 18.0262 +trainer/V1 Predictions Max 0.25381 +trainer/V1 Predictions Min -85.9774 +trainer/VF Loss 0.0644361 +expl/num steps total 240000 +expl/num paths total 260 +expl/path length Mean 500 +expl/path length Std 399 +expl/path length Max 899 +expl/path length Min 101 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0408812 +expl/Actions Std 0.825568 +expl/Actions Max 2.59326 +expl/Actions Min -2.35462 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 233936 +eval/num paths total 240 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.14669 +eval/Actions Std 0.655821 +eval/Actions Max 0.99819 +eval/Actions Min -0.997679 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.63005e-06 +time/evaluation sampling (s) 3.37309 +time/exploration sampling (s) 3.87054 +time/logging (s) 0.00705948 +time/saving (s) 0.00951244 +time/training (s) 13.9183 +time/epoch (s) 21.1785 +time/total (s) 5115.04 +Epoch -761 +------------------------------ ---------------- +2022-05-15 19:28:01.821771 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -760 finished +------------------------------ ---------------- +epoch -760 +replay_buffer/size 999047 +trainer/num train calls 241000 +trainer/QF1 Loss 0.808025 +trainer/QF2 Loss 0.942298 +trainer/Policy Loss 15.5416 +trainer/Q1 Predictions Mean -71.5925 +trainer/Q1 Predictions Std 19.2393 +trainer/Q1 Predictions Max -0.565456 +trainer/Q1 Predictions Min -86.4271 +trainer/Q2 Predictions Mean -71.6745 +trainer/Q2 Predictions Std 19.2197 +trainer/Q2 Predictions Max -0.866865 +trainer/Q2 Predictions Min -86.8199 +trainer/Q Targets Mean -71.197 +trainer/Q Targets Std 19.3576 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7346 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0115904 +trainer/policy/mean Std 0.725877 +trainer/policy/mean Max 0.998803 +trainer/policy/mean Min -0.997984 +trainer/policy/std Mean 0.430028 +trainer/policy/std Std 0.0223942 +trainer/policy/std Max 0.453182 +trainer/policy/std Min 0.392547 +trainer/Advantage Weights Mean 4.27338 +trainer/Advantage Weights Std 16.4496 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.26781e-13 +trainer/Advantage Score Mean -0.295987 +trainer/Advantage Score Std 0.471332 +trainer/Advantage Score Max 1.4261 +trainer/Advantage Score Min -2.96963 +trainer/V1 Predictions Mean -70.9495 +trainer/V1 Predictions Std 19.377 +trainer/V1 Predictions Max -0.179268 +trainer/V1 Predictions Min -86.6396 +trainer/VF Loss 0.0484385 +expl/num steps total 241000 +expl/num paths total 261 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0267894 +expl/Actions Std 0.863744 +expl/Actions Max 2.54273 +expl/Actions Min -2.3927 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 234936 +eval/num paths total 241 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.142262 +eval/Actions Std 0.718883 +eval/Actions Max 0.99933 +eval/Actions Min -0.999845 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.60491e-06 +time/evaluation sampling (s) 3.03943 +time/exploration sampling (s) 3.61293 +time/logging (s) 0.00868437 +time/saving (s) 0.0132699 +time/training (s) 13.948 +time/epoch (s) 20.6223 +time/total (s) 5135.66 +Epoch -760 +------------------------------ ---------------- +2022-05-15 19:28:21.535671 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -759 finished +------------------------------ ---------------- +epoch -759 +replay_buffer/size 999047 +trainer/num train calls 242000 +trainer/QF1 Loss 0.436517 +trainer/QF2 Loss 0.42963 +trainer/Policy Loss 17.9147 +trainer/Q1 Predictions Mean -70.6945 +trainer/Q1 Predictions Std 18.5095 +trainer/Q1 Predictions Max -0.270586 +trainer/Q1 Predictions Min -86.1955 +trainer/Q2 Predictions Mean -70.6795 +trainer/Q2 Predictions Std 18.455 +trainer/Q2 Predictions Max -0.407216 +trainer/Q2 Predictions Min -86.0066 +trainer/Q Targets Mean -70.7448 +trainer/Q Targets Std 18.4205 +trainer/Q Targets Max 0.600213 +trainer/Q Targets Min -86.2811 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00552626 +trainer/policy/mean Std 0.703435 +trainer/policy/mean Max 0.998979 +trainer/policy/mean Min -0.996569 +trainer/policy/std Mean 0.429169 +trainer/policy/std Std 0.0228957 +trainer/policy/std Max 0.452117 +trainer/policy/std Min 0.391088 +trainer/Advantage Weights Mean 4.34796 +trainer/Advantage Weights Std 17.0788 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.16481e-16 +trainer/Advantage Score Mean -0.326633 +trainer/Advantage Score Std 0.552469 +trainer/Advantage Score Max 1.62707 +trainer/Advantage Score Min -3.6069 +trainer/V1 Predictions Mean -70.4292 +trainer/V1 Predictions Std 18.6803 +trainer/V1 Predictions Max 2.31492 +trainer/V1 Predictions Min -86.09 +trainer/VF Loss 0.0602173 +expl/num steps total 242000 +expl/num paths total 263 +expl/path length Mean 500 +expl/path length Std 454 +expl/path length Max 954 +expl/path length Min 46 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0583501 +expl/Actions Std 0.842941 +expl/Actions Max 2.26725 +expl/Actions Min -2.41472 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 235936 +eval/num paths total 242 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.471057 +eval/Actions Std 0.618236 +eval/Actions Max 0.998069 +eval/Actions Min -0.994777 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.03402e-06 +time/evaluation sampling (s) 3.06998 +time/exploration sampling (s) 3.12209 +time/logging (s) 0.00764028 +time/saving (s) 0.0117084 +time/training (s) 13.4928 +time/epoch (s) 19.7042 +time/total (s) 5155.37 +Epoch -759 +------------------------------ ---------------- +2022-05-15 19:28:41.069199 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -758 finished +------------------------------ ---------------- +epoch -758 +replay_buffer/size 999047 +trainer/num train calls 243000 +trainer/QF1 Loss 0.879191 +trainer/QF2 Loss 0.932365 +trainer/Policy Loss 8.84782 +trainer/Q1 Predictions Mean -71.4941 +trainer/Q1 Predictions Std 18.8804 +trainer/Q1 Predictions Max -0.98223 +trainer/Q1 Predictions Min -86.3378 +trainer/Q2 Predictions Mean -71.5266 +trainer/Q2 Predictions Std 18.8537 +trainer/Q2 Predictions Max -0.823405 +trainer/Q2 Predictions Min -86.2967 +trainer/Q Targets Mean -71.0934 +trainer/Q Targets Std 19.2514 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9047 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0185708 +trainer/policy/mean Std 0.701091 +trainer/policy/mean Max 0.998745 +trainer/policy/mean Min -0.993936 +trainer/policy/std Mean 0.429272 +trainer/policy/std Std 0.0215498 +trainer/policy/std Max 0.450387 +trainer/policy/std Min 0.391915 +trainer/Advantage Weights Mean 1.40302 +trainer/Advantage Weights Std 9.2251 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.26677e-16 +trainer/Advantage Score Mean -0.6881 +trainer/Advantage Score Std 0.735381 +trainer/Advantage Score Max 0.949485 +trainer/Advantage Score Min -3.6023 +trainer/V1 Predictions Mean -70.8202 +trainer/V1 Predictions Std 19.4068 +trainer/V1 Predictions Max 1.25411 +trainer/V1 Predictions Min -86.1505 +trainer/VF Loss 0.106907 +expl/num steps total 243000 +expl/num paths total 265 +expl/path length Mean 500 +expl/path length Std 451 +expl/path length Max 951 +expl/path length Min 49 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0480697 +expl/Actions Std 0.814157 +expl/Actions Max 2.3984 +expl/Actions Min -2.39462 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 236583 +eval/num paths total 243 +eval/path length Mean 647 +eval/path length Std 0 +eval/path length Max 647 +eval/path length Min 647 +eval/Rewards Mean 0.0015456 +eval/Rewards Std 0.0392837 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0223778 +eval/Actions Std 0.703684 +eval/Actions Max 0.998431 +eval/Actions Min -0.997596 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.11015e-06 +time/evaluation sampling (s) 2.69508 +time/exploration sampling (s) 3.57011 +time/logging (s) 0.00600821 +time/saving (s) 0.00933685 +time/training (s) 13.2438 +time/epoch (s) 19.5244 +time/total (s) 5174.9 +Epoch -758 +------------------------------ ---------------- +2022-05-15 19:29:01.580544 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -757 finished +------------------------------ ---------------- +epoch -757 +replay_buffer/size 999047 +trainer/num train calls 244000 +trainer/QF1 Loss 0.370446 +trainer/QF2 Loss 0.46453 +trainer/Policy Loss 8.9927 +trainer/Q1 Predictions Mean -71.9432 +trainer/Q1 Predictions Std 15.5359 +trainer/Q1 Predictions Max -0.919427 +trainer/Q1 Predictions Min -85.5452 +trainer/Q2 Predictions Mean -71.9793 +trainer/Q2 Predictions Std 15.5201 +trainer/Q2 Predictions Max -0.735292 +trainer/Q2 Predictions Min -85.5446 +trainer/Q Targets Mean -71.7061 +trainer/Q Targets Std 15.628 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.3837 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00850583 +trainer/policy/mean Std 0.708698 +trainer/policy/mean Max 0.998785 +trainer/policy/mean Min -0.997661 +trainer/policy/std Mean 0.428333 +trainer/policy/std Std 0.020792 +trainer/policy/std Max 0.448269 +trainer/policy/std Min 0.39175 +trainer/Advantage Weights Mean 2.13299 +trainer/Advantage Weights Std 13.8471 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.50323e-10 +trainer/Advantage Score Mean -0.589759 +trainer/Advantage Score Std 0.49286 +trainer/Advantage Score Max 0.932304 +trainer/Advantage Score Min -2.21083 +trainer/V1 Predictions Mean -71.443 +trainer/V1 Predictions Std 15.7235 +trainer/V1 Predictions Max 0.421888 +trainer/V1 Predictions Min -85.2242 +trainer/VF Loss 0.0675973 +expl/num steps total 244000 +expl/num paths total 266 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0498824 +expl/Actions Std 0.851386 +expl/Actions Max 2.45936 +expl/Actions Min -2.62355 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 237583 +eval/num paths total 244 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0838197 +eval/Actions Std 0.66172 +eval/Actions Max 0.999425 +eval/Actions Min -0.9979 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.61236e-06 +time/evaluation sampling (s) 2.80103 +time/exploration sampling (s) 3.95986 +time/logging (s) 0.0130148 +time/saving (s) 0.0157949 +time/training (s) 13.7226 +time/epoch (s) 20.5123 +time/total (s) 5195.42 +Epoch -757 +------------------------------ ---------------- +2022-05-15 19:29:22.846532 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -756 finished +------------------------------ ---------------- +epoch -756 +replay_buffer/size 999047 +trainer/num train calls 245000 +trainer/QF1 Loss 1.03042 +trainer/QF2 Loss 0.962196 +trainer/Policy Loss 34.7935 +trainer/Q1 Predictions Mean -70.3362 +trainer/Q1 Predictions Std 19.0854 +trainer/Q1 Predictions Max -0.771101 +trainer/Q1 Predictions Min -85.8238 +trainer/Q2 Predictions Mean -70.3695 +trainer/Q2 Predictions Std 19.0718 +trainer/Q2 Predictions Max -0.794104 +trainer/Q2 Predictions Min -85.9996 +trainer/Q Targets Mean -70.5197 +trainer/Q Targets Std 19.133 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.6944 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0083457 +trainer/policy/mean Std 0.704567 +trainer/policy/mean Max 0.998612 +trainer/policy/mean Min -0.997943 +trainer/policy/std Mean 0.428041 +trainer/policy/std Std 0.0227422 +trainer/policy/std Max 0.45131 +trainer/policy/std Min 0.389616 +trainer/Advantage Weights Mean 6.88739 +trainer/Advantage Weights Std 21.0062 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.27341e-12 +trainer/Advantage Score Mean -0.2786 +trainer/Advantage Score Std 0.535412 +trainer/Advantage Score Max 1.34142 +trainer/Advantage Score Min -2.68097 +trainer/V1 Predictions Mean -70.2818 +trainer/V1 Predictions Std 19.1359 +trainer/V1 Predictions Max 0.562969 +trainer/V1 Predictions Min -85.6293 +trainer/VF Loss 0.0725627 +expl/num steps total 245000 +expl/num paths total 267 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0745739 +expl/Actions Std 0.815619 +expl/Actions Max 2.44799 +expl/Actions Min -2.16205 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 238583 +eval/num paths total 245 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.44018 +eval/Actions Std 0.579125 +eval/Actions Max 0.99825 +eval/Actions Min -0.998257 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.64123e-06 +time/evaluation sampling (s) 3.36488 +time/exploration sampling (s) 3.99325 +time/logging (s) 0.0070243 +time/saving (s) 0.0141258 +time/training (s) 13.8689 +time/epoch (s) 21.2482 +time/total (s) 5216.67 +Epoch -756 +------------------------------ ---------------- +2022-05-15 19:29:43.811998 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -755 finished +------------------------------ ---------------- +epoch -755 +replay_buffer/size 999047 +trainer/num train calls 246000 +trainer/QF1 Loss 0.787685 +trainer/QF2 Loss 0.679962 +trainer/Policy Loss 19.2238 +trainer/Q1 Predictions Mean -71.6004 +trainer/Q1 Predictions Std 17.3738 +trainer/Q1 Predictions Max -0.820557 +trainer/Q1 Predictions Min -85.8234 +trainer/Q2 Predictions Mean -71.5992 +trainer/Q2 Predictions Std 17.4242 +trainer/Q2 Predictions Max -1.43172 +trainer/Q2 Predictions Min -86.0366 +trainer/Q Targets Mean -71.8122 +trainer/Q Targets Std 17.4366 +trainer/Q Targets Max -1.0459 +trainer/Q Targets Min -86.2571 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0205856 +trainer/policy/mean Std 0.724033 +trainer/policy/mean Max 0.999014 +trainer/policy/mean Min -0.999601 +trainer/policy/std Mean 0.427472 +trainer/policy/std Std 0.0213016 +trainer/policy/std Max 0.453316 +trainer/policy/std Min 0.390679 +trainer/Advantage Weights Mean 4.31363 +trainer/Advantage Weights Std 17.6253 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.18985e-11 +trainer/Advantage Score Mean -0.342751 +trainer/Advantage Score Std 0.540513 +trainer/Advantage Score Max 1.21027 +trainer/Advantage Score Min -2.45446 +trainer/V1 Predictions Mean -71.5488 +trainer/V1 Predictions Std 17.4682 +trainer/V1 Predictions Max -0.816764 +trainer/V1 Predictions Min -86.1702 +trainer/VF Loss 0.064387 +expl/num steps total 246000 +expl/num paths total 268 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.045326 +expl/Actions Std 0.834721 +expl/Actions Max 2.30269 +expl/Actions Min -2.34422 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 239583 +eval/num paths total 246 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0311737 +eval/Actions Std 0.740231 +eval/Actions Max 0.999399 +eval/Actions Min -0.998843 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.63099e-06 +time/evaluation sampling (s) 3.15155 +time/exploration sampling (s) 3.77613 +time/logging (s) 0.00732358 +time/saving (s) 0.0104073 +time/training (s) 14.014 +time/epoch (s) 20.9594 +time/total (s) 5237.64 +Epoch -755 +------------------------------ ---------------- +2022-05-15 19:30:03.872048 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -754 finished +------------------------------ ---------------- +epoch -754 +replay_buffer/size 999047 +trainer/num train calls 247000 +trainer/QF1 Loss 0.860769 +trainer/QF2 Loss 0.94913 +trainer/Policy Loss 36.2136 +trainer/Q1 Predictions Mean -71.0983 +trainer/Q1 Predictions Std 18.1461 +trainer/Q1 Predictions Max -0.399717 +trainer/Q1 Predictions Min -86.1662 +trainer/Q2 Predictions Mean -71.0319 +trainer/Q2 Predictions Std 18.0788 +trainer/Q2 Predictions Max -0.554497 +trainer/Q2 Predictions Min -86.3253 +trainer/Q Targets Mean -71.5617 +trainer/Q Targets Std 18.3111 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2111 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0251561 +trainer/policy/mean Std 0.702163 +trainer/policy/mean Max 0.998852 +trainer/policy/mean Min -0.995974 +trainer/policy/std Mean 0.428415 +trainer/policy/std Std 0.0222825 +trainer/policy/std Max 0.450624 +trainer/policy/std Min 0.388837 +trainer/Advantage Weights Mean 8.53226 +trainer/Advantage Weights Std 23.903 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.25369e-10 +trainer/Advantage Score Mean -0.166812 +trainer/Advantage Score Std 0.459188 +trainer/Advantage Score Max 1.57854 +trainer/Advantage Score Min -2.27998 +trainer/V1 Predictions Mean -71.3237 +trainer/V1 Predictions Std 18.3196 +trainer/V1 Predictions Max -0.676155 +trainer/V1 Predictions Min -86.1266 +trainer/VF Loss 0.0638512 +expl/num steps total 247000 +expl/num paths total 269 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00357689 +expl/Actions Std 0.837216 +expl/Actions Max 2.32772 +expl/Actions Min -2.41506 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 240583 +eval/num paths total 247 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.071556 +eval/Actions Std 0.732522 +eval/Actions Max 0.999735 +eval/Actions Min -0.998214 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.6077e-06 +time/evaluation sampling (s) 3.17104 +time/exploration sampling (s) 3.04373 +time/logging (s) 0.00765348 +time/saving (s) 0.0120784 +time/training (s) 13.8192 +time/epoch (s) 20.0537 +time/total (s) 5257.69 +Epoch -754 +------------------------------ ---------------- +2022-05-15 19:30:23.858515 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -753 finished +------------------------------ ---------------- +epoch -753 +replay_buffer/size 999047 +trainer/num train calls 248000 +trainer/QF1 Loss 0.965679 +trainer/QF2 Loss 0.990236 +trainer/Policy Loss 8.80589 +trainer/Q1 Predictions Mean -70.1569 +trainer/Q1 Predictions Std 19.2667 +trainer/Q1 Predictions Max -0.399826 +trainer/Q1 Predictions Min -85.6269 +trainer/Q2 Predictions Mean -70.1661 +trainer/Q2 Predictions Std 19.2714 +trainer/Q2 Predictions Max -0.333502 +trainer/Q2 Predictions Min -85.6026 +trainer/Q Targets Mean -70.2143 +trainer/Q Targets Std 19.6958 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.5909 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0208225 +trainer/policy/mean Std 0.707874 +trainer/policy/mean Max 0.997901 +trainer/policy/mean Min -0.997731 +trainer/policy/std Mean 0.427862 +trainer/policy/std Std 0.0211992 +trainer/policy/std Max 0.451757 +trainer/policy/std Min 0.389746 +trainer/Advantage Weights Mean 2.63924 +trainer/Advantage Weights Std 11.8099 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.49046e-16 +trainer/Advantage Score Mean -0.420931 +trainer/Advantage Score Std 0.686776 +trainer/Advantage Score Max 0.867534 +trainer/Advantage Score Min -3.64423 +trainer/V1 Predictions Mean -69.9264 +trainer/V1 Predictions Std 19.7571 +trainer/V1 Predictions Max 0.640377 +trainer/V1 Predictions Min -85.6546 +trainer/VF Loss 0.073468 +expl/num steps total 248000 +expl/num paths total 270 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.223526 +expl/Actions Std 0.813966 +expl/Actions Max 2.58591 +expl/Actions Min -2.02892 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 241583 +eval/num paths total 248 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.2211 +eval/Actions Std 0.628711 +eval/Actions Max 0.998427 +eval/Actions Min -0.998686 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.80282e-06 +time/evaluation sampling (s) 2.768 +time/exploration sampling (s) 3.417 +time/logging (s) 0.00736186 +time/saving (s) 0.0102756 +time/training (s) 13.7756 +time/epoch (s) 19.9782 +time/total (s) 5277.68 +Epoch -753 +------------------------------ ---------------- +2022-05-15 19:30:44.019475 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -752 finished +------------------------------ ---------------- +epoch -752 +replay_buffer/size 999047 +trainer/num train calls 249000 +trainer/QF1 Loss 0.464371 +trainer/QF2 Loss 0.513815 +trainer/Policy Loss 5.36672 +trainer/Q1 Predictions Mean -71.6246 +trainer/Q1 Predictions Std 17.6332 +trainer/Q1 Predictions Max -0.789211 +trainer/Q1 Predictions Min -86.2236 +trainer/Q2 Predictions Mean -71.6746 +trainer/Q2 Predictions Std 17.6577 +trainer/Q2 Predictions Max -0.703714 +trainer/Q2 Predictions Min -86.3228 +trainer/Q Targets Mean -71.4802 +trainer/Q Targets Std 17.7287 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7883 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00787333 +trainer/policy/mean Std 0.719814 +trainer/policy/mean Max 0.998239 +trainer/policy/mean Min -0.999231 +trainer/policy/std Mean 0.428382 +trainer/policy/std Std 0.0211503 +trainer/policy/std Max 0.452283 +trainer/policy/std Min 0.390178 +trainer/Advantage Weights Mean 1.46703 +trainer/Advantage Weights Std 10.3281 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.27937e-13 +trainer/Advantage Score Mean -0.56141 +trainer/Advantage Score Std 0.488736 +trainer/Advantage Score Max 0.895432 +trainer/Advantage Score Min -2.91097 +trainer/V1 Predictions Mean -71.158 +trainer/V1 Predictions Std 17.8999 +trainer/V1 Predictions Max -0.13739 +trainer/V1 Predictions Min -85.6465 +trainer/VF Loss 0.060467 +expl/num steps total 249000 +expl/num paths total 271 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0294517 +expl/Actions Std 0.81376 +expl/Actions Max 2.45394 +expl/Actions Min -2.19878 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 242583 +eval/num paths total 249 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0455551 +eval/Actions Std 0.668384 +eval/Actions Max 0.997623 +eval/Actions Min -0.996859 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.29595e-06 +time/evaluation sampling (s) 2.91203 +time/exploration sampling (s) 3.47478 +time/logging (s) 0.00768305 +time/saving (s) 0.012098 +time/training (s) 13.7482 +time/epoch (s) 20.1548 +time/total (s) 5297.84 +Epoch -752 +------------------------------ ---------------- +2022-05-15 19:31:04.199768 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -751 finished +------------------------------ ---------------- +epoch -751 +replay_buffer/size 999047 +trainer/num train calls 250000 +trainer/QF1 Loss 0.55234 +trainer/QF2 Loss 0.633104 +trainer/Policy Loss 8.40628 +trainer/Q1 Predictions Mean -71.4168 +trainer/Q1 Predictions Std 17.803 +trainer/Q1 Predictions Max -0.789328 +trainer/Q1 Predictions Min -86.704 +trainer/Q2 Predictions Mean -71.5482 +trainer/Q2 Predictions Std 17.8369 +trainer/Q2 Predictions Max -0.51451 +trainer/Q2 Predictions Min -86.8072 +trainer/Q Targets Mean -71.3824 +trainer/Q Targets Std 18.0473 +trainer/Q Targets Max -0.109817 +trainer/Q Targets Min -86.4292 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00780202 +trainer/policy/mean Std 0.712461 +trainer/policy/mean Max 0.999272 +trainer/policy/mean Min -0.998776 +trainer/policy/std Mean 0.429145 +trainer/policy/std Std 0.0204913 +trainer/policy/std Max 0.453009 +trainer/policy/std Min 0.393984 +trainer/Advantage Weights Mean 1.64737 +trainer/Advantage Weights Std 10.1092 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.21584e-15 +trainer/Advantage Score Mean -0.561116 +trainer/Advantage Score Std 0.660431 +trainer/Advantage Score Max 1.03642 +trainer/Advantage Score Min -3.43433 +trainer/V1 Predictions Mean -71.1154 +trainer/V1 Predictions Std 18.1392 +trainer/V1 Predictions Max 0.15652 +trainer/V1 Predictions Min -86.2883 +trainer/VF Loss 0.0818437 +expl/num steps total 250000 +expl/num paths total 272 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.046921 +expl/Actions Std 0.8332 +expl/Actions Max 2.56034 +expl/Actions Min -2.55999 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 243505 +eval/num paths total 250 +eval/path length Mean 922 +eval/path length Std 0 +eval/path length Max 922 +eval/path length Min 922 +eval/Rewards Mean 0.0010846 +eval/Rewards Std 0.0329154 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00666377 +eval/Actions Std 0.726986 +eval/Actions Max 0.999872 +eval/Actions Min -0.999661 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.18629e-06 +time/evaluation sampling (s) 3.16284 +time/exploration sampling (s) 3.61803 +time/logging (s) 0.0107338 +time/saving (s) 0.0153475 +time/training (s) 13.3685 +time/epoch (s) 20.1755 +time/total (s) 5318.02 +Epoch -751 +------------------------------ ---------------- +2022-05-15 19:31:24.701253 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -750 finished +------------------------------ ---------------- +epoch -750 +replay_buffer/size 999047 +trainer/num train calls 251000 +trainer/QF1 Loss 0.564447 +trainer/QF2 Loss 0.61635 +trainer/Policy Loss 8.28955 +trainer/Q1 Predictions Mean -73.6659 +trainer/Q1 Predictions Std 16.7062 +trainer/Q1 Predictions Max -2.52608 +trainer/Q1 Predictions Min -86.5754 +trainer/Q2 Predictions Mean -73.678 +trainer/Q2 Predictions Std 16.6992 +trainer/Q2 Predictions Max -2.32312 +trainer/Q2 Predictions Min -86.5009 +trainer/Q Targets Mean -73.3004 +trainer/Q Targets Std 16.6497 +trainer/Q Targets Max -3.82067 +trainer/Q Targets Min -86.3288 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0206123 +trainer/policy/mean Std 0.709181 +trainer/policy/mean Max 0.998848 +trainer/policy/mean Min -0.998699 +trainer/policy/std Mean 0.427492 +trainer/policy/std Std 0.0226567 +trainer/policy/std Max 0.455396 +trainer/policy/std Min 0.387286 +trainer/Advantage Weights Mean 1.92027 +trainer/Advantage Weights Std 11.6791 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.94297e-14 +trainer/Advantage Score Mean -0.469534 +trainer/Advantage Score Std 0.500859 +trainer/Advantage Score Max 1.18556 +trainer/Advantage Score Min -3.02985 +trainer/V1 Predictions Mean -73.0269 +trainer/V1 Predictions Std 16.7664 +trainer/V1 Predictions Max -3.12836 +trainer/V1 Predictions Min -86.2322 +trainer/VF Loss 0.0559706 +expl/num steps total 251000 +expl/num paths total 273 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.059762 +expl/Actions Std 0.832541 +expl/Actions Max 2.20409 +expl/Actions Min -2.24716 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 244505 +eval/num paths total 251 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.06683 +eval/Actions Std 0.751815 +eval/Actions Max 0.999607 +eval/Actions Min -0.999176 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.10829e-06 +time/evaluation sampling (s) 3.15194 +time/exploration sampling (s) 3.55076 +time/logging (s) 0.00809231 +time/saving (s) 0.0221957 +time/training (s) 13.7582 +time/epoch (s) 20.4912 +time/total (s) 5338.51 +Epoch -750 +------------------------------ ---------------- +2022-05-15 19:31:44.535885 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -749 finished +------------------------------ --------------- +epoch -749 +replay_buffer/size 999047 +trainer/num train calls 252000 +trainer/QF1 Loss 0.578408 +trainer/QF2 Loss 0.562941 +trainer/Policy Loss 9.6344 +trainer/Q1 Predictions Mean -72.2593 +trainer/Q1 Predictions Std 16.799 +trainer/Q1 Predictions Max -0.44809 +trainer/Q1 Predictions Min -86.1849 +trainer/Q2 Predictions Mean -72.2926 +trainer/Q2 Predictions Std 16.7763 +trainer/Q2 Predictions Max -0.400971 +trainer/Q2 Predictions Min -86.2436 +trainer/Q Targets Mean -72.2212 +trainer/Q Targets Std 16.8167 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.6197 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00835061 +trainer/policy/mean Std 0.709327 +trainer/policy/mean Max 0.999126 +trainer/policy/mean Min -0.997005 +trainer/policy/std Mean 0.428503 +trainer/policy/std Std 0.0208838 +trainer/policy/std Max 0.455641 +trainer/policy/std Min 0.392545 +trainer/Advantage Weights Mean 1.98359 +trainer/Advantage Weights Std 11.504 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.3735e-14 +trainer/Advantage Score Mean -0.485369 +trainer/Advantage Score Std 0.533697 +trainer/Advantage Score Max 0.822742 +trainer/Advantage Score Min -3.10202 +trainer/V1 Predictions Mean -71.9157 +trainer/V1 Predictions Std 17.039 +trainer/V1 Predictions Max 1.20824 +trainer/V1 Predictions Min -85.5994 +trainer/VF Loss 0.0587794 +expl/num steps total 252000 +expl/num paths total 275 +expl/path length Mean 500 +expl/path length Std 465 +expl/path length Max 965 +expl/path length Min 35 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0525669 +expl/Actions Std 0.818306 +expl/Actions Max 2.56415 +expl/Actions Min -2.24606 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 245044 +eval/num paths total 252 +eval/path length Mean 539 +eval/path length Std 0 +eval/path length Max 539 +eval/path length Min 539 +eval/Rewards Mean 0.00185529 +eval/Rewards Std 0.0430331 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0389457 +eval/Actions Std 0.721655 +eval/Actions Max 0.999772 +eval/Actions Min -0.999404 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.0417e-06 +time/evaluation sampling (s) 3.10627 +time/exploration sampling (s) 2.90267 +time/logging (s) 0.00676191 +time/saving (s) 0.014215 +time/training (s) 13.7962 +time/epoch (s) 19.8261 +time/total (s) 5358.34 +Epoch -749 +------------------------------ --------------- +2022-05-15 19:32:04.626808 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -748 finished +------------------------------ ---------------- +epoch -748 +replay_buffer/size 999047 +trainer/num train calls 253000 +trainer/QF1 Loss 0.93639 +trainer/QF2 Loss 0.910986 +trainer/Policy Loss 24.6889 +trainer/Q1 Predictions Mean -73.1817 +trainer/Q1 Predictions Std 17.3066 +trainer/Q1 Predictions Max -2.74318 +trainer/Q1 Predictions Min -86.2697 +trainer/Q2 Predictions Mean -73.1881 +trainer/Q2 Predictions Std 17.3349 +trainer/Q2 Predictions Max -3.22304 +trainer/Q2 Predictions Min -86.8373 +trainer/Q Targets Mean -72.9103 +trainer/Q Targets Std 17.4989 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4689 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000650706 +trainer/policy/mean Std 0.704209 +trainer/policy/mean Max 0.998208 +trainer/policy/mean Min -0.997792 +trainer/policy/std Mean 0.42796 +trainer/policy/std Std 0.0222428 +trainer/policy/std Max 0.452905 +trainer/policy/std Min 0.388865 +trainer/Advantage Weights Mean 4.88786 +trainer/Advantage Weights Std 17.2097 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.14254e-17 +trainer/Advantage Score Mean -0.325386 +trainer/Advantage Score Std 0.589442 +trainer/Advantage Score Max 1.0961 +trainer/Advantage Score Min -3.70468 +trainer/V1 Predictions Mean -72.6932 +trainer/V1 Predictions Std 17.51 +trainer/V1 Predictions Max -2.93681 +trainer/V1 Predictions Min -86.3233 +trainer/VF Loss 0.0620071 +expl/num steps total 253000 +expl/num paths total 277 +expl/path length Mean 500 +expl/path length Std 358 +expl/path length Max 858 +expl/path length Min 142 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0503728 +expl/Actions Std 0.844989 +expl/Actions Max 2.12631 +expl/Actions Min -2.27503 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 246044 +eval/num paths total 253 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.294491 +eval/Actions Std 0.662825 +eval/Actions Max 0.999523 +eval/Actions Min -0.998727 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.22564e-06 +time/evaluation sampling (s) 2.72124 +time/exploration sampling (s) 3.15595 +time/logging (s) 0.00864701 +time/saving (s) 0.0111068 +time/training (s) 14.1897 +time/epoch (s) 20.0866 +time/total (s) 5378.43 +Epoch -748 +------------------------------ ---------------- +2022-05-15 19:32:24.998356 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -747 finished +------------------------------ ---------------- +epoch -747 +replay_buffer/size 999047 +trainer/num train calls 254000 +trainer/QF1 Loss 0.899358 +trainer/QF2 Loss 0.915714 +trainer/Policy Loss 6.41091 +trainer/Q1 Predictions Mean -71.8249 +trainer/Q1 Predictions Std 18.1235 +trainer/Q1 Predictions Max -0.333358 +trainer/Q1 Predictions Min -85.9619 +trainer/Q2 Predictions Mean -71.809 +trainer/Q2 Predictions Std 18.1608 +trainer/Q2 Predictions Max -0.638041 +trainer/Q2 Predictions Min -85.9546 +trainer/Q Targets Mean -71.4807 +trainer/Q Targets Std 18.0383 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.6457 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0102391 +trainer/policy/mean Std 0.711095 +trainer/policy/mean Max 0.996971 +trainer/policy/mean Min -0.99821 +trainer/policy/std Mean 0.428138 +trainer/policy/std Std 0.0229275 +trainer/policy/std Max 0.451592 +trainer/policy/std Min 0.385547 +trainer/Advantage Weights Mean 1.42023 +trainer/Advantage Weights Std 10.8757 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.2651e-17 +trainer/Advantage Score Mean -0.596987 +trainer/Advantage Score Std 0.513822 +trainer/Advantage Score Max 0.724536 +trainer/Advantage Score Min -3.73089 +trainer/V1 Predictions Mean -71.1921 +trainer/V1 Predictions Std 18.2252 +trainer/V1 Predictions Max 0.00945413 +trainer/V1 Predictions Min -85.5217 +trainer/VF Loss 0.0664882 +expl/num steps total 254000 +expl/num paths total 278 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.163146 +expl/Actions Std 0.795613 +expl/Actions Max 2.35849 +expl/Actions Min -2.36373 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 246453 +eval/num paths total 254 +eval/path length Mean 409 +eval/path length Std 0 +eval/path length Max 409 +eval/path length Min 409 +eval/Rewards Mean 0.00244499 +eval/Rewards Std 0.0493863 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0336526 +eval/Actions Std 0.719107 +eval/Actions Max 0.998543 +eval/Actions Min -0.997078 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.01283e-06 +time/evaluation sampling (s) 3.09293 +time/exploration sampling (s) 3.53491 +time/logging (s) 0.00584142 +time/saving (s) 0.0102333 +time/training (s) 13.7177 +time/epoch (s) 20.3616 +time/total (s) 5398.8 +Epoch -747 +------------------------------ ---------------- +2022-05-15 19:32:45.734389 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -746 finished +------------------------------ ---------------- +epoch -746 +replay_buffer/size 999047 +trainer/num train calls 255000 +trainer/QF1 Loss 1.0111 +trainer/QF2 Loss 1.16949 +trainer/Policy Loss 6.08628 +trainer/Q1 Predictions Mean -73.1185 +trainer/Q1 Predictions Std 15.7186 +trainer/Q1 Predictions Max -2.15064 +trainer/Q1 Predictions Min -86.6679 +trainer/Q2 Predictions Mean -73.2029 +trainer/Q2 Predictions Std 15.7428 +trainer/Q2 Predictions Max -1.72327 +trainer/Q2 Predictions Min -87.5739 +trainer/Q Targets Mean -72.4086 +trainer/Q Targets Std 15.9134 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.3272 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0422496 +trainer/policy/mean Std 0.731471 +trainer/policy/mean Max 0.999228 +trainer/policy/mean Min -0.999393 +trainer/policy/std Mean 0.42959 +trainer/policy/std Std 0.0216678 +trainer/policy/std Max 0.452126 +trainer/policy/std Min 0.391059 +trainer/Advantage Weights Mean 0.885564 +trainer/Advantage Weights Std 8.1836 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.32947e-14 +trainer/Advantage Score Mean -0.719339 +trainer/Advantage Score Std 0.495554 +trainer/Advantage Score Max 0.620823 +trainer/Advantage Score Min -3.19514 +trainer/V1 Predictions Mean -72.2233 +trainer/V1 Predictions Std 15.8994 +trainer/V1 Predictions Max -0.559464 +trainer/V1 Predictions Min -85.2482 +trainer/VF Loss 0.0785382 +expl/num steps total 255000 +expl/num paths total 279 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0242487 +expl/Actions Std 0.892721 +expl/Actions Max 2.18923 +expl/Actions Min -2.54651 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 247453 +eval/num paths total 255 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0362035 +eval/Actions Std 0.74107 +eval/Actions Max 0.999808 +eval/Actions Min -0.999081 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.44984e-06 +time/evaluation sampling (s) 3.43418 +time/exploration sampling (s) 3.63795 +time/logging (s) 0.00689418 +time/saving (s) 0.0108586 +time/training (s) 13.6408 +time/epoch (s) 20.7307 +time/total (s) 5419.53 +Epoch -746 +------------------------------ ---------------- +2022-05-15 19:33:06.726180 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -745 finished +------------------------------ ---------------- +epoch -745 +replay_buffer/size 999047 +trainer/num train calls 256000 +trainer/QF1 Loss 0.694193 +trainer/QF2 Loss 0.746147 +trainer/Policy Loss 25.8222 +trainer/Q1 Predictions Mean -71.2874 +trainer/Q1 Predictions Std 19.7852 +trainer/Q1 Predictions Max -0.507868 +trainer/Q1 Predictions Min -85.9413 +trainer/Q2 Predictions Mean -71.2708 +trainer/Q2 Predictions Std 19.7837 +trainer/Q2 Predictions Max -0.410343 +trainer/Q2 Predictions Min -85.8679 +trainer/Q Targets Mean -71.302 +trainer/Q Targets Std 20.184 +trainer/Q Targets Max 0.44021 +trainer/Q Targets Min -85.7052 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000767897 +trainer/policy/mean Std 0.717029 +trainer/policy/mean Max 0.998757 +trainer/policy/mean Min -0.997914 +trainer/policy/std Mean 0.431631 +trainer/policy/std Std 0.0212281 +trainer/policy/std Max 0.4555 +trainer/policy/std Min 0.393085 +trainer/Advantage Weights Mean 3.89357 +trainer/Advantage Weights Std 15.6712 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.14257e-11 +trainer/Advantage Score Mean -0.330135 +trainer/Advantage Score Std 0.534739 +trainer/Advantage Score Max 1.58502 +trainer/Advantage Score Min -2.51952 +trainer/V1 Predictions Mean -71.0113 +trainer/V1 Predictions Std 20.3571 +trainer/V1 Predictions Max 1.8187 +trainer/V1 Predictions Min -85.7104 +trainer/VF Loss 0.0572762 +expl/num steps total 256000 +expl/num paths total 280 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0842494 +expl/Actions Std 0.848194 +expl/Actions Max 2.26005 +expl/Actions Min -2.38661 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 248453 +eval/num paths total 256 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0618296 +eval/Actions Std 0.737064 +eval/Actions Max 0.999166 +eval/Actions Min -0.997732 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.92202e-06 +time/evaluation sampling (s) 3.15065 +time/exploration sampling (s) 3.54231 +time/logging (s) 0.00776653 +time/saving (s) 0.0119737 +time/training (s) 14.2736 +time/epoch (s) 20.9863 +time/total (s) 5440.52 +Epoch -745 +------------------------------ ---------------- +2022-05-15 19:33:26.459022 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -744 finished +------------------------------ ---------------- +epoch -744 +replay_buffer/size 999047 +trainer/num train calls 257000 +trainer/QF1 Loss 0.595356 +trainer/QF2 Loss 0.580575 +trainer/Policy Loss 11.1701 +trainer/Q1 Predictions Mean -72.3585 +trainer/Q1 Predictions Std 17.5204 +trainer/Q1 Predictions Max -1.54821 +trainer/Q1 Predictions Min -86.3534 +trainer/Q2 Predictions Mean -72.3201 +trainer/Q2 Predictions Std 17.5782 +trainer/Q2 Predictions Max -0.821284 +trainer/Q2 Predictions Min -86.4655 +trainer/Q Targets Mean -72.2823 +trainer/Q Targets Std 17.5951 +trainer/Q Targets Max 0.869995 +trainer/Q Targets Min -86.5026 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00555041 +trainer/policy/mean Std 0.714606 +trainer/policy/mean Max 0.998694 +trainer/policy/mean Min -0.998822 +trainer/policy/std Mean 0.429225 +trainer/policy/std Std 0.0219518 +trainer/policy/std Max 0.452827 +trainer/policy/std Min 0.390922 +trainer/Advantage Weights Mean 2.94015 +trainer/Advantage Weights Std 14.3958 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.2756e-18 +trainer/Advantage Score Mean -0.450693 +trainer/Advantage Score Std 0.570459 +trainer/Advantage Score Max 0.684692 +trainer/Advantage Score Min -3.97834 +trainer/V1 Predictions Mean -72.0242 +trainer/V1 Predictions Std 17.7047 +trainer/V1 Predictions Max 1.95043 +trainer/V1 Predictions Min -86.2268 +trainer/VF Loss 0.0607569 +expl/num steps total 257000 +expl/num paths total 281 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.214042 +expl/Actions Std 0.84395 +expl/Actions Max 2.47914 +expl/Actions Min -2.19464 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 248923 +eval/num paths total 257 +eval/path length Mean 470 +eval/path length Std 0 +eval/path length Max 470 +eval/path length Min 470 +eval/Rewards Mean 0.00212766 +eval/Rewards Std 0.0460775 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0428529 +eval/Actions Std 0.716187 +eval/Actions Max 0.999011 +eval/Actions Min -0.998388 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.56626e-06 +time/evaluation sampling (s) 2.95219 +time/exploration sampling (s) 2.88727 +time/logging (s) 0.00619834 +time/saving (s) 0.0127776 +time/training (s) 13.8648 +time/epoch (s) 19.7232 +time/total (s) 5460.25 +Epoch -744 +------------------------------ ---------------- +2022-05-15 19:33:46.194589 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -743 finished +------------------------------ ---------------- +epoch -743 +replay_buffer/size 999047 +trainer/num train calls 258000 +trainer/QF1 Loss 0.839828 +trainer/QF2 Loss 0.812106 +trainer/Policy Loss 43.8556 +trainer/Q1 Predictions Mean -72.1625 +trainer/Q1 Predictions Std 17.846 +trainer/Q1 Predictions Max -0.362751 +trainer/Q1 Predictions Min -85.6717 +trainer/Q2 Predictions Mean -72.1981 +trainer/Q2 Predictions Std 17.9109 +trainer/Q2 Predictions Max -0.409314 +trainer/Q2 Predictions Min -85.6565 +trainer/Q Targets Mean -72.5753 +trainer/Q Targets Std 17.5282 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7416 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00169453 +trainer/policy/mean Std 0.720515 +trainer/policy/mean Max 0.999018 +trainer/policy/mean Min -0.998177 +trainer/policy/std Mean 0.427872 +trainer/policy/std Std 0.0223251 +trainer/policy/std Max 0.452758 +trainer/policy/std Min 0.389637 +trainer/Advantage Weights Mean 9.81784 +trainer/Advantage Weights Std 25.115 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.1818e-11 +trainer/Advantage Score Mean -0.127855 +trainer/Advantage Score Std 0.46449 +trainer/Advantage Score Max 1.64139 +trainer/Advantage Score Min -2.51614 +trainer/V1 Predictions Mean -72.2993 +trainer/V1 Predictions Std 17.776 +trainer/V1 Predictions Max -0.480586 +trainer/V1 Predictions Min -85.6629 +trainer/VF Loss 0.07338 +expl/num steps total 258000 +expl/num paths total 282 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.276008 +expl/Actions Std 0.784136 +expl/Actions Max 2.26266 +expl/Actions Min -2.28824 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 249923 +eval/num paths total 258 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.125768 +eval/Actions Std 0.751285 +eval/Actions Max 0.999328 +eval/Actions Min -0.998542 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.82004e-06 +time/evaluation sampling (s) 2.53977 +time/exploration sampling (s) 3.37647 +time/logging (s) 0.00941485 +time/saving (s) 0.0122424 +time/training (s) 13.7925 +time/epoch (s) 19.7304 +time/total (s) 5479.99 +Epoch -743 +------------------------------ ---------------- +2022-05-15 19:34:06.367614 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -742 finished +------------------------------ ---------------- +epoch -742 +replay_buffer/size 999047 +trainer/num train calls 259000 +trainer/QF1 Loss 2.02859 +trainer/QF2 Loss 1.94804 +trainer/Policy Loss 39.1358 +trainer/Q1 Predictions Mean -71.5542 +trainer/Q1 Predictions Std 18.6042 +trainer/Q1 Predictions Max -0.433526 +trainer/Q1 Predictions Min -85.6396 +trainer/Q2 Predictions Mean -71.6018 +trainer/Q2 Predictions Std 18.636 +trainer/Q2 Predictions Max -0.367168 +trainer/Q2 Predictions Min -85.9264 +trainer/Q Targets Mean -71.993 +trainer/Q Targets Std 18.8899 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4882 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0170889 +trainer/policy/mean Std 0.721337 +trainer/policy/mean Max 0.999383 +trainer/policy/mean Min -0.99735 +trainer/policy/std Mean 0.427542 +trainer/policy/std Std 0.0211462 +trainer/policy/std Max 0.44977 +trainer/policy/std Min 0.392459 +trainer/Advantage Weights Mean 9.55735 +trainer/Advantage Weights Std 22.6424 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.26685e-15 +trainer/Advantage Score Mean -0.219344 +trainer/Advantage Score Std 0.629058 +trainer/Advantage Score Max 1.04639 +trainer/Advantage Score Min -3.33549 +trainer/V1 Predictions Mean -71.6809 +trainer/V1 Predictions Std 19.1399 +trainer/V1 Predictions Max 0.770953 +trainer/V1 Predictions Min -86.5423 +trainer/VF Loss 0.0734296 +expl/num steps total 259000 +expl/num paths total 283 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0427973 +expl/Actions Std 0.81642 +expl/Actions Max 2.72354 +expl/Actions Min -2.33055 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 250923 +eval/num paths total 259 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.029073 +eval/Actions Std 0.733006 +eval/Actions Max 0.999765 +eval/Actions Min -0.999281 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.80887e-06 +time/evaluation sampling (s) 2.95062 +time/exploration sampling (s) 3.63969 +time/logging (s) 0.0106977 +time/saving (s) 0.0139592 +time/training (s) 13.5515 +time/epoch (s) 20.1665 +time/total (s) 5500.16 +Epoch -742 +------------------------------ ---------------- +2022-05-15 19:34:26.707641 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -741 finished +------------------------------ ---------------- +epoch -741 +replay_buffer/size 999047 +trainer/num train calls 260000 +trainer/QF1 Loss 0.353545 +trainer/QF2 Loss 0.421528 +trainer/Policy Loss 17.9228 +trainer/Q1 Predictions Mean -71.5272 +trainer/Q1 Predictions Std 18.1509 +trainer/Q1 Predictions Max -0.567129 +trainer/Q1 Predictions Min -86.4216 +trainer/Q2 Predictions Mean -71.5172 +trainer/Q2 Predictions Std 18.0878 +trainer/Q2 Predictions Max -0.437258 +trainer/Q2 Predictions Min -86.0254 +trainer/Q Targets Mean -71.4329 +trainer/Q Targets Std 18.1882 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5327 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00804038 +trainer/policy/mean Std 0.702194 +trainer/policy/mean Max 0.999543 +trainer/policy/mean Min -0.998852 +trainer/policy/std Mean 0.427072 +trainer/policy/std Std 0.0210458 +trainer/policy/std Max 0.450011 +trainer/policy/std Min 0.393068 +trainer/Advantage Weights Mean 4.23826 +trainer/Advantage Weights Std 17.0951 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.96664e-12 +trainer/Advantage Score Mean -0.367208 +trainer/Advantage Score Std 0.523716 +trainer/Advantage Score Max 2.08759 +trainer/Advantage Score Min -2.56899 +trainer/V1 Predictions Mean -71.1423 +trainer/V1 Predictions Std 18.2096 +trainer/V1 Predictions Max 0.542278 +trainer/V1 Predictions Min -86.5108 +trainer/VF Loss 0.0678564 +expl/num steps total 260000 +expl/num paths total 284 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.162223 +expl/Actions Std 0.864573 +expl/Actions Max 2.58938 +expl/Actions Min -2.26192 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 251875 +eval/num paths total 260 +eval/path length Mean 952 +eval/path length Std 0 +eval/path length Max 952 +eval/path length Min 952 +eval/Rewards Mean 0.00105042 +eval/Rewards Std 0.0323932 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00980499 +eval/Actions Std 0.733729 +eval/Actions Max 0.999496 +eval/Actions Min -0.99858 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.18909e-06 +time/evaluation sampling (s) 3.08691 +time/exploration sampling (s) 3.70329 +time/logging (s) 0.0072087 +time/saving (s) 0.0127596 +time/training (s) 13.5174 +time/epoch (s) 20.3276 +time/total (s) 5520.49 +Epoch -741 +------------------------------ ---------------- +2022-05-15 19:34:47.009070 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -740 finished +------------------------------ ---------------- +epoch -740 +replay_buffer/size 999047 +trainer/num train calls 261000 +trainer/QF1 Loss 0.98649 +trainer/QF2 Loss 0.994012 +trainer/Policy Loss 18.6538 +trainer/Q1 Predictions Mean -70.8807 +trainer/Q1 Predictions Std 20.0037 +trainer/Q1 Predictions Max -0.356428 +trainer/Q1 Predictions Min -86.1257 +trainer/Q2 Predictions Mean -70.9109 +trainer/Q2 Predictions Std 19.9586 +trainer/Q2 Predictions Max -0.378575 +trainer/Q2 Predictions Min -86.213 +trainer/Q Targets Mean -71.0488 +trainer/Q Targets Std 19.667 +trainer/Q Targets Max 1.37268 +trainer/Q Targets Min -85.8832 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00294934 +trainer/policy/mean Std 0.717834 +trainer/policy/mean Max 0.999532 +trainer/policy/mean Min -0.998274 +trainer/policy/std Mean 0.428506 +trainer/policy/std Std 0.0221461 +trainer/policy/std Max 0.452305 +trainer/policy/std Min 0.393728 +trainer/Advantage Weights Mean 3.03068 +trainer/Advantage Weights Std 13.6763 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.87511e-12 +trainer/Advantage Score Mean -0.367605 +trainer/Advantage Score Std 0.476136 +trainer/Advantage Score Max 0.725052 +trainer/Advantage Score Min -2.70024 +trainer/V1 Predictions Mean -70.8355 +trainer/V1 Predictions Std 19.792 +trainer/V1 Predictions Max 2.31918 +trainer/V1 Predictions Min -85.7414 +trainer/VF Loss 0.0447465 +expl/num steps total 261000 +expl/num paths total 285 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0491802 +expl/Actions Std 0.83082 +expl/Actions Max 2.42827 +expl/Actions Min -2.35119 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 252875 +eval/num paths total 261 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0231228 +eval/Actions Std 0.765186 +eval/Actions Max 0.999507 +eval/Actions Min -0.997887 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.73483e-06 +time/evaluation sampling (s) 3.15139 +time/exploration sampling (s) 3.54517 +time/logging (s) 0.00715412 +time/saving (s) 0.00996123 +time/training (s) 13.5796 +time/epoch (s) 20.2933 +time/total (s) 5540.79 +Epoch -740 +------------------------------ ---------------- +2022-05-15 19:35:07.430179 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -739 finished +------------------------------ ---------------- +epoch -739 +replay_buffer/size 999047 +trainer/num train calls 262000 +trainer/QF1 Loss 0.698873 +trainer/QF2 Loss 0.722548 +trainer/Policy Loss 20.055 +trainer/Q1 Predictions Mean -73.543 +trainer/Q1 Predictions Std 15.6244 +trainer/Q1 Predictions Max -1.22033 +trainer/Q1 Predictions Min -85.6881 +trainer/Q2 Predictions Mean -73.5503 +trainer/Q2 Predictions Std 15.5764 +trainer/Q2 Predictions Max -1.44374 +trainer/Q2 Predictions Min -85.5787 +trainer/Q Targets Mean -73.8873 +trainer/Q Targets Std 15.7065 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0166 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00251024 +trainer/policy/mean Std 0.719331 +trainer/policy/mean Max 0.999481 +trainer/policy/mean Min -0.997803 +trainer/policy/std Mean 0.429093 +trainer/policy/std Std 0.0224497 +trainer/policy/std Max 0.454487 +trainer/policy/std Min 0.394131 +trainer/Advantage Weights Mean 3.7329 +trainer/Advantage Weights Std 14.8652 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.73611e-13 +trainer/Advantage Score Mean -0.237437 +trainer/Advantage Score Std 0.413881 +trainer/Advantage Score Max 1.16272 +trainer/Advantage Score Min -2.80261 +trainer/V1 Predictions Mean -73.6764 +trainer/V1 Predictions Std 15.6975 +trainer/V1 Predictions Max -2.35319 +trainer/V1 Predictions Min -85.844 +trainer/VF Loss 0.037193 +expl/num steps total 262000 +expl/num paths total 287 +expl/path length Mean 500 +expl/path length Std 27 +expl/path length Max 527 +expl/path length Min 473 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0821826 +expl/Actions Std 0.830779 +expl/Actions Max 2.62159 +expl/Actions Min -2.35608 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 253875 +eval/num paths total 262 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0821955 +eval/Actions Std 0.749929 +eval/Actions Max 0.999731 +eval/Actions Min -0.998472 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.95602e-06 +time/evaluation sampling (s) 3.24864 +time/exploration sampling (s) 3.26006 +time/logging (s) 0.00738606 +time/saving (s) 0.010253 +time/training (s) 13.8884 +time/epoch (s) 20.4148 +time/total (s) 5561.21 +Epoch -739 +------------------------------ ---------------- +2022-05-15 19:35:26.714059 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -738 finished +------------------------------ ---------------- +epoch -738 +replay_buffer/size 999047 +trainer/num train calls 263000 +trainer/QF1 Loss 0.602272 +trainer/QF2 Loss 0.52049 +trainer/Policy Loss 17.8146 +trainer/Q1 Predictions Mean -72.3733 +trainer/Q1 Predictions Std 17.2362 +trainer/Q1 Predictions Max -0.467765 +trainer/Q1 Predictions Min -85.9058 +trainer/Q2 Predictions Mean -72.3954 +trainer/Q2 Predictions Std 17.178 +trainer/Q2 Predictions Max -0.433459 +trainer/Q2 Predictions Min -85.7621 +trainer/Q Targets Mean -72.7121 +trainer/Q Targets Std 17.037 +trainer/Q Targets Max -0.231068 +trainer/Q Targets Min -85.8873 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0124821 +trainer/policy/mean Std 0.717443 +trainer/policy/mean Max 0.998522 +trainer/policy/mean Min -0.997615 +trainer/policy/std Mean 0.428287 +trainer/policy/std Std 0.0226667 +trainer/policy/std Max 0.454369 +trainer/policy/std Min 0.392013 +trainer/Advantage Weights Mean 4.35404 +trainer/Advantage Weights Std 18.8648 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.96558e-12 +trainer/Advantage Score Mean -0.459583 +trainer/Advantage Score Std 0.489639 +trainer/Advantage Score Max 1.3058 +trainer/Advantage Score Min -2.60285 +trainer/V1 Predictions Mean -72.42 +trainer/V1 Predictions Std 17.3682 +trainer/V1 Predictions Max 1.20505 +trainer/V1 Predictions Min -85.8092 +trainer/VF Loss 0.0654025 +expl/num steps total 263000 +expl/num paths total 288 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0485013 +expl/Actions Std 0.82818 +expl/Actions Max 2.23045 +expl/Actions Min -2.60892 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 254875 +eval/num paths total 263 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.239813 +eval/Actions Std 0.743785 +eval/Actions Max 0.999759 +eval/Actions Min -0.999532 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.62214e-06 +time/evaluation sampling (s) 2.8072 +time/exploration sampling (s) 3.19789 +time/logging (s) 0.00934611 +time/saving (s) 0.0140299 +time/training (s) 13.2507 +time/epoch (s) 19.2792 +time/total (s) 5580.49 +Epoch -738 +------------------------------ ---------------- +2022-05-15 19:35:46.607190 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -737 finished +------------------------------ ---------------- +epoch -737 +replay_buffer/size 999047 +trainer/num train calls 264000 +trainer/QF1 Loss 0.769425 +trainer/QF2 Loss 0.792794 +trainer/Policy Loss 11.368 +trainer/Q1 Predictions Mean -72.7781 +trainer/Q1 Predictions Std 15.5515 +trainer/Q1 Predictions Max -0.698057 +trainer/Q1 Predictions Min -86.2456 +trainer/Q2 Predictions Mean -72.7429 +trainer/Q2 Predictions Std 15.5737 +trainer/Q2 Predictions Max -0.770282 +trainer/Q2 Predictions Min -86.4363 +trainer/Q Targets Mean -72.9401 +trainer/Q Targets Std 15.5989 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6568 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00106825 +trainer/policy/mean Std 0.704755 +trainer/policy/mean Max 0.999713 +trainer/policy/mean Min -0.999267 +trainer/policy/std Mean 0.426199 +trainer/policy/std Std 0.0224173 +trainer/policy/std Max 0.452387 +trainer/policy/std Min 0.390828 +trainer/Advantage Weights Mean 3.26841 +trainer/Advantage Weights Std 15.427 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.41697e-15 +trainer/Advantage Score Mean -0.392985 +trainer/Advantage Score Std 0.508606 +trainer/Advantage Score Max 1.03419 +trainer/Advantage Score Min -3.30533 +trainer/V1 Predictions Mean -72.705 +trainer/V1 Predictions Std 15.6068 +trainer/V1 Predictions Max 0.499095 +trainer/V1 Predictions Min -86.7319 +trainer/VF Loss 0.0543283 +expl/num steps total 264000 +expl/num paths total 289 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0684077 +expl/Actions Std 0.812558 +expl/Actions Max 2.47218 +expl/Actions Min -2.40366 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 255875 +eval/num paths total 264 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.245837 +eval/Actions Std 0.812755 +eval/Actions Max 0.999317 +eval/Actions Min -0.998979 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.63611e-06 +time/evaluation sampling (s) 2.60454 +time/exploration sampling (s) 3.58331 +time/logging (s) 0.00700598 +time/saving (s) 0.00960846 +time/training (s) 13.6776 +time/epoch (s) 19.882 +time/total (s) 5600.38 +Epoch -737 +------------------------------ ---------------- +2022-05-15 19:36:07.448999 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -736 finished +------------------------------ ---------------- +epoch -736 +replay_buffer/size 999047 +trainer/num train calls 265000 +trainer/QF1 Loss 0.941284 +trainer/QF2 Loss 0.904277 +trainer/Policy Loss 11.0241 +trainer/Q1 Predictions Mean -69.8477 +trainer/Q1 Predictions Std 19.7315 +trainer/Q1 Predictions Max -1.05948 +trainer/Q1 Predictions Min -86.5026 +trainer/Q2 Predictions Mean -69.9007 +trainer/Q2 Predictions Std 19.7108 +trainer/Q2 Predictions Max -0.835811 +trainer/Q2 Predictions Min -86.5544 +trainer/Q Targets Mean -69.4777 +trainer/Q Targets Std 19.858 +trainer/Q Targets Max 1.14438 +trainer/Q Targets Min -86.1983 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00187903 +trainer/policy/mean Std 0.703897 +trainer/policy/mean Max 0.998289 +trainer/policy/mean Min -0.996964 +trainer/policy/std Mean 0.427516 +trainer/policy/std Std 0.0216405 +trainer/policy/std Max 0.451987 +trainer/policy/std Min 0.391221 +trainer/Advantage Weights Mean 2.50738 +trainer/Advantage Weights Std 13.1445 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.70718e-15 +trainer/Advantage Score Mean -0.656735 +trainer/Advantage Score Std 0.670287 +trainer/Advantage Score Max 0.84658 +trainer/Advantage Score Min -3.24966 +trainer/V1 Predictions Mean -69.1873 +trainer/V1 Predictions Std 19.9943 +trainer/V1 Predictions Max 0.48482 +trainer/V1 Predictions Min -86.0054 +trainer/VF Loss 0.0965555 +expl/num steps total 265000 +expl/num paths total 290 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.232247 +expl/Actions Std 0.828705 +expl/Actions Max 2.42345 +expl/Actions Min -2.12444 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 256875 +eval/num paths total 265 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0699776 +eval/Actions Std 0.721203 +eval/Actions Max 0.998702 +eval/Actions Min -0.999374 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.56905e-06 +time/evaluation sampling (s) 3.02985 +time/exploration sampling (s) 3.78299 +time/logging (s) 0.00850705 +time/saving (s) 0.0122564 +time/training (s) 14.0033 +time/epoch (s) 20.8369 +time/total (s) 5621.22 +Epoch -736 +------------------------------ ---------------- +2022-05-15 19:36:27.317020 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -735 finished +------------------------------ ---------------- +epoch -735 +replay_buffer/size 999047 +trainer/num train calls 266000 +trainer/QF1 Loss 0.534934 +trainer/QF2 Loss 0.487838 +trainer/Policy Loss 2.28153 +trainer/Q1 Predictions Mean -70.7078 +trainer/Q1 Predictions Std 19.8951 +trainer/Q1 Predictions Max -1.37671 +trainer/Q1 Predictions Min -86.4369 +trainer/Q2 Predictions Mean -70.6776 +trainer/Q2 Predictions Std 19.9237 +trainer/Q2 Predictions Max -1.09961 +trainer/Q2 Predictions Min -86.3566 +trainer/Q Targets Mean -70.4757 +trainer/Q Targets Std 19.9799 +trainer/Q Targets Max -2.00508 +trainer/Q Targets Min -86.0775 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00751367 +trainer/policy/mean Std 0.709966 +trainer/policy/mean Max 0.999284 +trainer/policy/mean Min -0.998615 +trainer/policy/std Mean 0.428463 +trainer/policy/std Std 0.0203842 +trainer/policy/std Max 0.449697 +trainer/policy/std Min 0.393863 +trainer/Advantage Weights Mean 0.754231 +trainer/Advantage Weights Std 6.471 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.10666e-15 +trainer/Advantage Score Mean -0.595325 +trainer/Advantage Score Std 0.658942 +trainer/Advantage Score Max 0.573303 +trainer/Advantage Score Min -3.44374 +trainer/V1 Predictions Mean -70.1958 +trainer/V1 Predictions Std 20.1096 +trainer/V1 Predictions Max -0.769925 +trainer/V1 Predictions Min -85.8925 +trainer/VF Loss 0.0810319 +expl/num steps total 266000 +expl/num paths total 291 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0979462 +expl/Actions Std 0.825384 +expl/Actions Max 2.44287 +expl/Actions Min -2.12196 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 257543 +eval/num paths total 266 +eval/path length Mean 668 +eval/path length Std 0 +eval/path length Max 668 +eval/path length Min 668 +eval/Rewards Mean 0.00149701 +eval/Rewards Std 0.0386622 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00998584 +eval/Actions Std 0.729619 +eval/Actions Max 0.999355 +eval/Actions Min -0.999571 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.22914e-06 +time/evaluation sampling (s) 3.06335 +time/exploration sampling (s) 3.65054 +time/logging (s) 0.00602329 +time/saving (s) 0.00940769 +time/training (s) 13.1293 +time/epoch (s) 19.8586 +time/total (s) 5641.08 +Epoch -735 +------------------------------ ---------------- +2022-05-15 19:36:47.243306 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -734 finished +------------------------------ ---------------- +epoch -734 +replay_buffer/size 999047 +trainer/num train calls 267000 +trainer/QF1 Loss 0.750616 +trainer/QF2 Loss 0.905391 +trainer/Policy Loss 11.7096 +trainer/Q1 Predictions Mean -70.7829 +trainer/Q1 Predictions Std 18.6813 +trainer/Q1 Predictions Max -0.734793 +trainer/Q1 Predictions Min -86.3063 +trainer/Q2 Predictions Mean -70.8686 +trainer/Q2 Predictions Std 18.5378 +trainer/Q2 Predictions Max -1.27165 +trainer/Q2 Predictions Min -86.6938 +trainer/Q Targets Mean -70.8869 +trainer/Q Targets Std 18.908 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4778 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0146435 +trainer/policy/mean Std 0.714497 +trainer/policy/mean Max 0.998803 +trainer/policy/mean Min -0.999235 +trainer/policy/std Mean 0.429053 +trainer/policy/std Std 0.0212077 +trainer/policy/std Max 0.451276 +trainer/policy/std Min 0.396091 +trainer/Advantage Weights Mean 3.01854 +trainer/Advantage Weights Std 15.3545 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.38544e-15 +trainer/Advantage Score Mean -0.546894 +trainer/Advantage Score Std 0.57438 +trainer/Advantage Score Max 1.67595 +trainer/Advantage Score Min -3.42128 +trainer/V1 Predictions Mean -70.6794 +trainer/V1 Predictions Std 18.726 +trainer/V1 Predictions Max -1.83921 +trainer/V1 Predictions Min -86.7063 +trainer/VF Loss 0.091223 +expl/num steps total 267000 +expl/num paths total 293 +expl/path length Mean 500 +expl/path length Std 418 +expl/path length Max 918 +expl/path length Min 82 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0761931 +expl/Actions Std 0.810712 +expl/Actions Max 2.59462 +expl/Actions Min -2.34462 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 258115 +eval/num paths total 267 +eval/path length Mean 572 +eval/path length Std 0 +eval/path length Max 572 +eval/path length Min 572 +eval/Rewards Mean 0.00174825 +eval/Rewards Std 0.0417755 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0325132 +eval/Actions Std 0.717588 +eval/Actions Max 0.999148 +eval/Actions Min -0.999126 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.80002e-06 +time/evaluation sampling (s) 3.10017 +time/exploration sampling (s) 3.45996 +time/logging (s) 0.00653833 +time/saving (s) 0.0132658 +time/training (s) 13.3408 +time/epoch (s) 19.9207 +time/total (s) 5661 +Epoch -734 +------------------------------ ---------------- +2022-05-15 19:37:07.410420 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -733 finished +------------------------------ ---------------- +epoch -733 +replay_buffer/size 999047 +trainer/num train calls 268000 +trainer/QF1 Loss 4.19932 +trainer/QF2 Loss 4.0083 +trainer/Policy Loss 46.9857 +trainer/Q1 Predictions Mean -72.3024 +trainer/Q1 Predictions Std 17.0071 +trainer/Q1 Predictions Max -0.737555 +trainer/Q1 Predictions Min -86.8262 +trainer/Q2 Predictions Mean -72.3718 +trainer/Q2 Predictions Std 16.9148 +trainer/Q2 Predictions Max -1.02522 +trainer/Q2 Predictions Min -86.8602 +trainer/Q Targets Mean -73.1372 +trainer/Q Targets Std 15.8531 +trainer/Q Targets Max -2.8095 +trainer/Q Targets Min -85.7608 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00864375 +trainer/policy/mean Std 0.714537 +trainer/policy/mean Max 0.998567 +trainer/policy/mean Min -0.997266 +trainer/policy/std Mean 0.430063 +trainer/policy/std Std 0.0209659 +trainer/policy/std Max 0.452172 +trainer/policy/std Min 0.397102 +trainer/Advantage Weights Mean 7.91138 +trainer/Advantage Weights Std 24.541 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.53489e-09 +trainer/Advantage Score Mean -0.311841 +trainer/Advantage Score Std 0.493305 +trainer/Advantage Score Max 1.48356 +trainer/Advantage Score Min -1.92115 +trainer/V1 Predictions Mean -72.8576 +trainer/V1 Predictions Std 15.9258 +trainer/V1 Predictions Max -1.95832 +trainer/V1 Predictions Min -85.5846 +trainer/VF Loss 0.0751824 +expl/num steps total 268000 +expl/num paths total 294 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.232606 +expl/Actions Std 0.849664 +expl/Actions Max 2.35395 +expl/Actions Min -2.25606 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 259115 +eval/num paths total 268 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0710357 +eval/Actions Std 0.708513 +eval/Actions Max 0.999644 +eval/Actions Min -0.998098 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.20002e-06 +time/evaluation sampling (s) 3.07775 +time/exploration sampling (s) 3.14074 +time/logging (s) 0.00903936 +time/saving (s) 0.0143411 +time/training (s) 13.9181 +time/epoch (s) 20.16 +time/total (s) 5681.17 +Epoch -733 +------------------------------ ---------------- +2022-05-15 19:37:27.048532 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -732 finished +------------------------------ ---------------- +epoch -732 +replay_buffer/size 999047 +trainer/num train calls 269000 +trainer/QF1 Loss 0.648863 +trainer/QF2 Loss 0.585868 +trainer/Policy Loss 11.6916 +trainer/Q1 Predictions Mean -72.8126 +trainer/Q1 Predictions Std 16.5837 +trainer/Q1 Predictions Max -0.472585 +trainer/Q1 Predictions Min -86.3391 +trainer/Q2 Predictions Mean -72.7857 +trainer/Q2 Predictions Std 16.5193 +trainer/Q2 Predictions Max -0.596932 +trainer/Q2 Predictions Min -86.2055 +trainer/Q Targets Mean -73.081 +trainer/Q Targets Std 16.6217 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5973 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0150404 +trainer/policy/mean Std 0.723906 +trainer/policy/mean Max 0.997535 +trainer/policy/mean Min -0.998795 +trainer/policy/std Mean 0.429688 +trainer/policy/std Std 0.0216331 +trainer/policy/std Max 0.449694 +trainer/policy/std Min 0.391564 +trainer/Advantage Weights Mean 3.06184 +trainer/Advantage Weights Std 13.6652 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.42329e-18 +trainer/Advantage Score Mean -0.359929 +trainer/Advantage Score Std 0.517031 +trainer/Advantage Score Max 0.952244 +trainer/Advantage Score Min -3.95866 +trainer/V1 Predictions Mean -72.8065 +trainer/V1 Predictions Std 16.6866 +trainer/V1 Predictions Max 0.495475 +trainer/V1 Predictions Min -86.6022 +trainer/VF Loss 0.0494995 +expl/num steps total 269000 +expl/num paths total 295 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0354693 +expl/Actions Std 0.814906 +expl/Actions Max 2.45302 +expl/Actions Min -2.54737 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 260115 +eval/num paths total 269 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0709458 +eval/Actions Std 0.750966 +eval/Actions Max 0.99921 +eval/Actions Min -0.998217 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.9183e-06 +time/evaluation sampling (s) 2.69673 +time/exploration sampling (s) 3.39459 +time/logging (s) 0.00723063 +time/saving (s) 0.0118777 +time/training (s) 13.516 +time/epoch (s) 19.6265 +time/total (s) 5700.8 +Epoch -732 +------------------------------ ---------------- +2022-05-15 19:37:47.237635 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -731 finished +------------------------------ ---------------- +epoch -731 +replay_buffer/size 999047 +trainer/num train calls 270000 +trainer/QF1 Loss 1.07603 +trainer/QF2 Loss 1.25442 +trainer/Policy Loss 13.1019 +trainer/Q1 Predictions Mean -70.3598 +trainer/Q1 Predictions Std 19.7963 +trainer/Q1 Predictions Max -0.525097 +trainer/Q1 Predictions Min -85.9522 +trainer/Q2 Predictions Mean -70.254 +trainer/Q2 Predictions Std 19.8025 +trainer/Q2 Predictions Max -0.382332 +trainer/Q2 Predictions Min -85.8912 +trainer/Q Targets Mean -70.8499 +trainer/Q Targets Std 19.5703 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7293 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0265636 +trainer/policy/mean Std 0.720396 +trainer/policy/mean Max 0.998492 +trainer/policy/mean Min -0.997797 +trainer/policy/std Mean 0.429562 +trainer/policy/std Std 0.0211743 +trainer/policy/std Max 0.451 +trainer/policy/std Min 0.393172 +trainer/Advantage Weights Mean 2.60697 +trainer/Advantage Weights Std 14.2503 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.17329e-21 +trainer/Advantage Score Mean -0.626571 +trainer/Advantage Score Std 0.782145 +trainer/Advantage Score Max 1.52881 +trainer/Advantage Score Min -4.67108 +trainer/V1 Predictions Mean -70.4704 +trainer/V1 Predictions Std 19.7815 +trainer/V1 Predictions Max 1.96169 +trainer/V1 Predictions Min -85.756 +trainer/VF Loss 0.119445 +expl/num steps total 270000 +expl/num paths total 296 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0307391 +expl/Actions Std 0.856275 +expl/Actions Max 2.55047 +expl/Actions Min -2.54841 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 261115 +eval/num paths total 270 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0438535 +eval/Actions Std 0.739752 +eval/Actions Max 0.999567 +eval/Actions Min -0.999396 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.87406e-06 +time/evaluation sampling (s) 2.73781 +time/exploration sampling (s) 3.78165 +time/logging (s) 0.00833128 +time/saving (s) 0.0120004 +time/training (s) 13.642 +time/epoch (s) 20.1818 +time/total (s) 5720.99 +Epoch -731 +------------------------------ ---------------- +2022-05-15 19:38:06.994284 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -730 finished +------------------------------ ---------------- +epoch -730 +replay_buffer/size 999047 +trainer/num train calls 271000 +trainer/QF1 Loss 1.4588 +trainer/QF2 Loss 1.49851 +trainer/Policy Loss 45.9059 +trainer/Q1 Predictions Mean -67.5102 +trainer/Q1 Predictions Std 22.2173 +trainer/Q1 Predictions Max -0.286524 +trainer/Q1 Predictions Min -85.458 +trainer/Q2 Predictions Mean -67.4935 +trainer/Q2 Predictions Std 22.2214 +trainer/Q2 Predictions Max -0.365878 +trainer/Q2 Predictions Min -85.5356 +trainer/Q Targets Mean -67.8619 +trainer/Q Targets Std 21.991 +trainer/Q Targets Max 0.200245 +trainer/Q Targets Min -85.9305 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00496881 +trainer/policy/mean Std 0.712091 +trainer/policy/mean Max 0.999678 +trainer/policy/mean Min -0.997681 +trainer/policy/std Mean 0.428268 +trainer/policy/std Std 0.0212552 +trainer/policy/std Max 0.450721 +trainer/policy/std Min 0.395069 +trainer/Advantage Weights Mean 10.7272 +trainer/Advantage Weights Std 27.0017 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.20165e-12 +trainer/Advantage Score Mean -0.134062 +trainer/Advantage Score Std 0.55706 +trainer/Advantage Score Max 2.10439 +trainer/Advantage Score Min -2.68418 +trainer/V1 Predictions Mean -67.651 +trainer/V1 Predictions Std 21.8872 +trainer/V1 Predictions Max 0.565909 +trainer/V1 Predictions Min -85.6615 +trainer/VF Loss 0.106612 +expl/num steps total 271000 +expl/num paths total 297 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0265531 +expl/Actions Std 0.802032 +expl/Actions Max 2.67413 +expl/Actions Min -2.27397 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 262115 +eval/num paths total 271 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.417729 +eval/Actions Std 0.543863 +eval/Actions Max 0.999509 +eval/Actions Min -0.998576 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.82237e-06 +time/evaluation sampling (s) 3.18825 +time/exploration sampling (s) 3.61994 +time/logging (s) 0.00693284 +time/saving (s) 0.00943081 +time/training (s) 12.9229 +time/epoch (s) 19.7474 +time/total (s) 5740.74 +Epoch -730 +------------------------------ ---------------- +2022-05-15 19:38:26.862496 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -729 finished +------------------------------ ---------------- +epoch -729 +replay_buffer/size 999047 +trainer/num train calls 272000 +trainer/QF1 Loss 0.871887 +trainer/QF2 Loss 0.850975 +trainer/Policy Loss 36.2864 +trainer/Q1 Predictions Mean -73.0023 +trainer/Q1 Predictions Std 15.7947 +trainer/Q1 Predictions Max -0.664525 +trainer/Q1 Predictions Min -86.0361 +trainer/Q2 Predictions Mean -73.0899 +trainer/Q2 Predictions Std 15.7887 +trainer/Q2 Predictions Max -0.77682 +trainer/Q2 Predictions Min -85.8123 +trainer/Q Targets Mean -73.3033 +trainer/Q Targets Std 15.9953 +trainer/Q Targets Max -0.0808101 +trainer/Q Targets Min -86.0606 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00204802 +trainer/policy/mean Std 0.718151 +trainer/policy/mean Max 0.999142 +trainer/policy/mean Min -0.998538 +trainer/policy/std Mean 0.42673 +trainer/policy/std Std 0.019964 +trainer/policy/std Max 0.449636 +trainer/policy/std Min 0.395486 +trainer/Advantage Weights Mean 8.20461 +trainer/Advantage Weights Std 21.7919 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.72936e-14 +trainer/Advantage Score Mean -0.223402 +trainer/Advantage Score Std 0.569534 +trainer/Advantage Score Max 1.04806 +trainer/Advantage Score Min -3.06824 +trainer/V1 Predictions Mean -72.999 +trainer/V1 Predictions Std 16.12 +trainer/V1 Predictions Max 0.342865 +trainer/V1 Predictions Min -85.985 +trainer/VF Loss 0.0641543 +expl/num steps total 272000 +expl/num paths total 298 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0273446 +expl/Actions Std 0.844605 +expl/Actions Max 2.18122 +expl/Actions Min -2.27864 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 262712 +eval/num paths total 272 +eval/path length Mean 597 +eval/path length Std 0 +eval/path length Max 597 +eval/path length Min 597 +eval/Rewards Mean 0.00167504 +eval/Rewards Std 0.040893 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0339843 +eval/Actions Std 0.729315 +eval/Actions Max 0.999507 +eval/Actions Min -0.998579 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.71061e-06 +time/evaluation sampling (s) 3.12746 +time/exploration sampling (s) 3.57079 +time/logging (s) 0.00773201 +time/saving (s) 0.012719 +time/training (s) 13.1438 +time/epoch (s) 19.8625 +time/total (s) 5760.61 +Epoch -729 +------------------------------ ---------------- +2022-05-15 19:38:46.963096 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -728 finished +------------------------------ ---------------- +epoch -728 +replay_buffer/size 999047 +trainer/num train calls 273000 +trainer/QF1 Loss 0.607373 +trainer/QF2 Loss 0.585984 +trainer/Policy Loss 5.28545 +trainer/Q1 Predictions Mean -70.8394 +trainer/Q1 Predictions Std 17.6759 +trainer/Q1 Predictions Max -0.576939 +trainer/Q1 Predictions Min -86.1194 +trainer/Q2 Predictions Mean -70.8768 +trainer/Q2 Predictions Std 17.6879 +trainer/Q2 Predictions Max -0.634527 +trainer/Q2 Predictions Min -86.374 +trainer/Q Targets Mean -70.6393 +trainer/Q Targets Std 17.7644 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4144 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0168702 +trainer/policy/mean Std 0.707126 +trainer/policy/mean Max 0.997899 +trainer/policy/mean Min -0.998594 +trainer/policy/std Mean 0.426583 +trainer/policy/std Std 0.0198262 +trainer/policy/std Max 0.448649 +trainer/policy/std Min 0.393534 +trainer/Advantage Weights Mean 0.744819 +trainer/Advantage Weights Std 6.7198 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.29493e-19 +trainer/Advantage Score Mean -0.614227 +trainer/Advantage Score Std 0.598977 +trainer/Advantage Score Max 0.526115 +trainer/Advantage Score Min -4.16335 +trainer/V1 Predictions Mean -70.3436 +trainer/V1 Predictions Std 18.044 +trainer/V1 Predictions Max -0.165324 +trainer/V1 Predictions Min -86.1534 +trainer/VF Loss 0.075304 +expl/num steps total 273000 +expl/num paths total 300 +expl/path length Mean 500 +expl/path length Std 179 +expl/path length Max 679 +expl/path length Min 321 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0391684 +expl/Actions Std 0.816082 +expl/Actions Max 2.27098 +expl/Actions Min -2.17294 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 263712 +eval/num paths total 273 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0471441 +eval/Actions Std 0.739576 +eval/Actions Max 0.999639 +eval/Actions Min -0.999751 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.31504e-06 +time/evaluation sampling (s) 2.99233 +time/exploration sampling (s) 3.36212 +time/logging (s) 0.00749371 +time/saving (s) 0.0117183 +time/training (s) 13.7198 +time/epoch (s) 20.0935 +time/total (s) 5780.71 +Epoch -728 +------------------------------ ---------------- +2022-05-15 19:39:07.204795 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -727 finished +------------------------------ ---------------- +epoch -727 +replay_buffer/size 999047 +trainer/num train calls 274000 +trainer/QF1 Loss 0.681779 +trainer/QF2 Loss 0.662984 +trainer/Policy Loss 13.2334 +trainer/Q1 Predictions Mean -69.9734 +trainer/Q1 Predictions Std 20.4743 +trainer/Q1 Predictions Max -0.43207 +trainer/Q1 Predictions Min -85.7279 +trainer/Q2 Predictions Mean -70.0027 +trainer/Q2 Predictions Std 20.5279 +trainer/Q2 Predictions Max -0.37822 +trainer/Q2 Predictions Min -86.0545 +trainer/Q Targets Mean -69.817 +trainer/Q Targets Std 20.6455 +trainer/Q Targets Max 0.927572 +trainer/Q Targets Min -85.7247 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00417737 +trainer/policy/mean Std 0.702943 +trainer/policy/mean Max 0.998514 +trainer/policy/mean Min -0.997695 +trainer/policy/std Mean 0.425629 +trainer/policy/std Std 0.0194494 +trainer/policy/std Max 0.446667 +trainer/policy/std Min 0.392713 +trainer/Advantage Weights Mean 2.90582 +trainer/Advantage Weights Std 14.4964 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.4235e-17 +trainer/Advantage Score Mean -0.49337 +trainer/Advantage Score Std 0.565629 +trainer/Advantage Score Max 1.13008 +trainer/Advantage Score Min -3.79133 +trainer/V1 Predictions Mean -69.5059 +trainer/V1 Predictions Std 20.7912 +trainer/V1 Predictions Max 0.64772 +trainer/V1 Predictions Min -85.8477 +trainer/VF Loss 0.069331 +expl/num steps total 274000 +expl/num paths total 302 +expl/path length Mean 500 +expl/path length Std 335 +expl/path length Max 835 +expl/path length Min 165 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0464593 +expl/Actions Std 0.81497 +expl/Actions Max 2.24043 +expl/Actions Min -2.51887 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 264712 +eval/num paths total 274 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.201817 +eval/Actions Std 0.84894 +eval/Actions Max 0.999422 +eval/Actions Min -0.996984 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.75997e-06 +time/evaluation sampling (s) 3.00312 +time/exploration sampling (s) 3.12233 +time/logging (s) 0.0105892 +time/saving (s) 0.0147636 +time/training (s) 14.0862 +time/epoch (s) 20.237 +time/total (s) 5800.95 +Epoch -727 +------------------------------ ---------------- +2022-05-15 19:39:27.660190 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -726 finished +------------------------------ ---------------- +epoch -726 +replay_buffer/size 999047 +trainer/num train calls 275000 +trainer/QF1 Loss 1.41123 +trainer/QF2 Loss 1.57844 +trainer/Policy Loss 27.8432 +trainer/Q1 Predictions Mean -72.342 +trainer/Q1 Predictions Std 16.4151 +trainer/Q1 Predictions Max -1.01265 +trainer/Q1 Predictions Min -85.9057 +trainer/Q2 Predictions Mean -72.2897 +trainer/Q2 Predictions Std 16.5216 +trainer/Q2 Predictions Max -0.801892 +trainer/Q2 Predictions Min -86.0082 +trainer/Q Targets Mean -72.5373 +trainer/Q Targets Std 16.322 +trainer/Q Targets Max -0.956112 +trainer/Q Targets Min -86.0719 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.021157 +trainer/policy/mean Std 0.725657 +trainer/policy/mean Max 0.998863 +trainer/policy/mean Min -0.9979 +trainer/policy/std Mean 0.426956 +trainer/policy/std Std 0.0203929 +trainer/policy/std Max 0.446191 +trainer/policy/std Min 0.391395 +trainer/Advantage Weights Mean 5.1532 +trainer/Advantage Weights Std 16.3206 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.64693e-16 +trainer/Advantage Score Mean -0.25759 +trainer/Advantage Score Std 0.522127 +trainer/Advantage Score Max 1.32125 +trainer/Advantage Score Min -3.51103 +trainer/V1 Predictions Mean -72.3263 +trainer/V1 Predictions Std 16.4767 +trainer/V1 Predictions Max -1.11562 +trainer/V1 Predictions Min -86.5079 +trainer/VF Loss 0.0540865 +expl/num steps total 275000 +expl/num paths total 303 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.240262 +expl/Actions Std 0.776408 +expl/Actions Max 2.15189 +expl/Actions Min -2.38294 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 265712 +eval/num paths total 275 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0876707 +eval/Actions Std 0.739706 +eval/Actions Max 0.998975 +eval/Actions Min -0.997354 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.15392e-06 +time/evaluation sampling (s) 2.7563 +time/exploration sampling (s) 3.92082 +time/logging (s) 0.00991086 +time/saving (s) 0.0145847 +time/training (s) 13.7446 +time/epoch (s) 20.4462 +time/total (s) 5821.4 +Epoch -726 +------------------------------ ---------------- +2022-05-15 19:39:47.794519 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -725 finished +------------------------------ ---------------- +epoch -725 +replay_buffer/size 999047 +trainer/num train calls 276000 +trainer/QF1 Loss 1.1655 +trainer/QF2 Loss 1.3291 +trainer/Policy Loss 49.9313 +trainer/Q1 Predictions Mean -70.8109 +trainer/Q1 Predictions Std 18.7316 +trainer/Q1 Predictions Max -0.684076 +trainer/Q1 Predictions Min -86.1899 +trainer/Q2 Predictions Mean -70.8518 +trainer/Q2 Predictions Std 18.7768 +trainer/Q2 Predictions Max -0.307298 +trainer/Q2 Predictions Min -86.4497 +trainer/Q Targets Mean -71.3732 +trainer/Q Targets Std 18.6372 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6655 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00453567 +trainer/policy/mean Std 0.715431 +trainer/policy/mean Max 0.998119 +trainer/policy/mean Min -0.998833 +trainer/policy/std Mean 0.427414 +trainer/policy/std Std 0.0205716 +trainer/policy/std Max 0.450306 +trainer/policy/std Min 0.388473 +trainer/Advantage Weights Mean 11.2075 +trainer/Advantage Weights Std 26.0443 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.16608e-12 +trainer/Advantage Score Mean -0.093855 +trainer/Advantage Score Std 0.445872 +trainer/Advantage Score Max 1.03377 +trainer/Advantage Score Min -2.68581 +trainer/V1 Predictions Mean -71.0851 +trainer/V1 Predictions Std 18.7177 +trainer/V1 Predictions Max -0.364977 +trainer/V1 Predictions Min -86.5234 +trainer/VF Loss 0.0591551 +expl/num steps total 276000 +expl/num paths total 305 +expl/path length Mean 500 +expl/path length Std 486 +expl/path length Max 986 +expl/path length Min 14 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0477922 +expl/Actions Std 0.839178 +expl/Actions Max 2.65504 +expl/Actions Min -2.33762 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 266383 +eval/num paths total 276 +eval/path length Mean 671 +eval/path length Std 0 +eval/path length Max 671 +eval/path length Min 671 +eval/Rewards Mean 0.00149031 +eval/Rewards Std 0.0385758 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0367443 +eval/Actions Std 0.733763 +eval/Actions Max 0.999377 +eval/Actions Min -0.999629 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.07709e-06 +time/evaluation sampling (s) 3.11362 +time/exploration sampling (s) 3.75029 +time/logging (s) 0.00589043 +time/saving (s) 0.0093804 +time/training (s) 13.2413 +time/epoch (s) 20.1204 +time/total (s) 5841.52 +Epoch -725 +------------------------------ ---------------- +2022-05-15 19:40:07.310660 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -724 finished +------------------------------ ---------------- +epoch -724 +replay_buffer/size 999047 +trainer/num train calls 277000 +trainer/QF1 Loss 0.804959 +trainer/QF2 Loss 0.885808 +trainer/Policy Loss 35.3574 +trainer/Q1 Predictions Mean -71.9875 +trainer/Q1 Predictions Std 18.5797 +trainer/Q1 Predictions Max -0.538714 +trainer/Q1 Predictions Min -86.0522 +trainer/Q2 Predictions Mean -71.9783 +trainer/Q2 Predictions Std 18.6163 +trainer/Q2 Predictions Max -0.357945 +trainer/Q2 Predictions Min -86.0662 +trainer/Q Targets Mean -72.1359 +trainer/Q Targets Std 18.6497 +trainer/Q Targets Max 0.49881 +trainer/Q Targets Min -86.2085 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000928977 +trainer/policy/mean Std 0.709179 +trainer/policy/mean Max 0.998916 +trainer/policy/mean Min -0.99872 +trainer/policy/std Mean 0.428219 +trainer/policy/std Std 0.0204375 +trainer/policy/std Max 0.451347 +trainer/policy/std Min 0.39034 +trainer/Advantage Weights Mean 8.5749 +trainer/Advantage Weights Std 22.9783 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.94369e-13 +trainer/Advantage Score Mean -0.172792 +trainer/Advantage Score Std 0.512391 +trainer/Advantage Score Max 1.71198 +trainer/Advantage Score Min -2.83355 +trainer/V1 Predictions Mean -71.9015 +trainer/V1 Predictions Std 18.6409 +trainer/V1 Predictions Max 2.26723 +trainer/V1 Predictions Min -86.0698 +trainer/VF Loss 0.0737899 +expl/num steps total 277000 +expl/num paths total 307 +expl/path length Mean 500 +expl/path length Std 483 +expl/path length Max 983 +expl/path length Min 17 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0442797 +expl/Actions Std 0.816268 +expl/Actions Max 2.32824 +expl/Actions Min -2.37219 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 267383 +eval/num paths total 277 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0734247 +eval/Actions Std 0.740091 +eval/Actions Max 0.999344 +eval/Actions Min -0.997319 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.63704e-06 +time/evaluation sampling (s) 3.08585 +time/exploration sampling (s) 3.82633 +time/logging (s) 0.00689402 +time/saving (s) 0.00926649 +time/training (s) 12.5827 +time/epoch (s) 19.5111 +time/total (s) 5861.04 +Epoch -724 +------------------------------ ---------------- +2022-05-15 19:40:27.937897 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -723 finished +------------------------------ ---------------- +epoch -723 +replay_buffer/size 999047 +trainer/num train calls 278000 +trainer/QF1 Loss 0.543565 +trainer/QF2 Loss 0.629321 +trainer/Policy Loss 24.476 +trainer/Q1 Predictions Mean -74.4161 +trainer/Q1 Predictions Std 14.2734 +trainer/Q1 Predictions Max -10.0473 +trainer/Q1 Predictions Min -85.8666 +trainer/Q2 Predictions Mean -74.3799 +trainer/Q2 Predictions Std 14.208 +trainer/Q2 Predictions Max -10.0969 +trainer/Q2 Predictions Min -85.9739 +trainer/Q Targets Mean -74.6547 +trainer/Q Targets Std 14.373 +trainer/Q Targets Max -7.19614 +trainer/Q Targets Min -85.725 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0254709 +trainer/policy/mean Std 0.718132 +trainer/policy/mean Max 0.999718 +trainer/policy/mean Min -0.99993 +trainer/policy/std Mean 0.427226 +trainer/policy/std Std 0.0194318 +trainer/policy/std Max 0.447671 +trainer/policy/std Min 0.39303 +trainer/Advantage Weights Mean 4.99615 +trainer/Advantage Weights Std 17.8998 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.58997e-13 +trainer/Advantage Score Mean -0.293798 +trainer/Advantage Score Std 0.517916 +trainer/Advantage Score Max 0.90699 +trainer/Advantage Score Min -2.80481 +trainer/V1 Predictions Mean -74.3581 +trainer/V1 Predictions Std 14.5555 +trainer/V1 Predictions Max -7.06406 +trainer/V1 Predictions Min -85.5929 +trainer/VF Loss 0.0523835 +expl/num steps total 278000 +expl/num paths total 308 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.141633 +expl/Actions Std 0.828609 +expl/Actions Max 2.47789 +expl/Actions Min -2.49114 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 268383 +eval/num paths total 278 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.143507 +eval/Actions Std 0.673434 +eval/Actions Max 0.999263 +eval/Actions Min -0.999143 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.41026e-06 +time/evaluation sampling (s) 3.02668 +time/exploration sampling (s) 3.63054 +time/logging (s) 0.00677293 +time/saving (s) 0.00937034 +time/training (s) 13.9475 +time/epoch (s) 20.6209 +time/total (s) 5881.66 +Epoch -723 +------------------------------ ---------------- +2022-05-15 19:40:47.314405 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -722 finished +------------------------------ ---------------- +epoch -722 +replay_buffer/size 999047 +trainer/num train calls 279000 +trainer/QF1 Loss 0.371292 +trainer/QF2 Loss 0.382138 +trainer/Policy Loss 9.5572 +trainer/Q1 Predictions Mean -72.314 +trainer/Q1 Predictions Std 18.2482 +trainer/Q1 Predictions Max -0.869927 +trainer/Q1 Predictions Min -86.5375 +trainer/Q2 Predictions Mean -72.2855 +trainer/Q2 Predictions Std 18.2572 +trainer/Q2 Predictions Max -0.667091 +trainer/Q2 Predictions Min -86.4788 +trainer/Q Targets Mean -72.3067 +trainer/Q Targets Std 18.258 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.232 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00837982 +trainer/policy/mean Std 0.723991 +trainer/policy/mean Max 0.997326 +trainer/policy/mean Min -0.998677 +trainer/policy/std Mean 0.424979 +trainer/policy/std Std 0.0207085 +trainer/policy/std Max 0.449905 +trainer/policy/std Min 0.38937 +trainer/Advantage Weights Mean 2.2975 +trainer/Advantage Weights Std 12.5823 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.59018e-14 +trainer/Advantage Score Mean -0.43132 +trainer/Advantage Score Std 0.518583 +trainer/Advantage Score Max 0.663063 +trainer/Advantage Score Min -3.12845 +trainer/V1 Predictions Mean -72.1042 +trainer/V1 Predictions Std 18.3932 +trainer/V1 Predictions Max 0.652569 +trainer/V1 Predictions Min -86.097 +trainer/VF Loss 0.0525887 +expl/num steps total 279000 +expl/num paths total 309 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.104693 +expl/Actions Std 0.819276 +expl/Actions Max 2.35418 +expl/Actions Min -2.19137 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 269056 +eval/num paths total 279 +eval/path length Mean 673 +eval/path length Std 0 +eval/path length Max 673 +eval/path length Min 673 +eval/Rewards Mean 0.00148588 +eval/Rewards Std 0.0385185 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0591597 +eval/Actions Std 0.707483 +eval/Actions Max 0.998295 +eval/Actions Min -0.997527 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 7.60891e-06 +time/evaluation sampling (s) 2.91668 +time/exploration sampling (s) 3.02817 +time/logging (s) 0.00699853 +time/saving (s) 0.0151856 +time/training (s) 13.403 +time/epoch (s) 19.37 +time/total (s) 5901.04 +Epoch -722 +------------------------------ ---------------- +2022-05-15 19:41:06.644531 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -721 finished +------------------------------ ---------------- +epoch -721 +replay_buffer/size 999047 +trainer/num train calls 280000 +trainer/QF1 Loss 0.87284 +trainer/QF2 Loss 0.743382 +trainer/Policy Loss 13.6201 +trainer/Q1 Predictions Mean -72.33 +trainer/Q1 Predictions Std 17.6926 +trainer/Q1 Predictions Max -2.81392 +trainer/Q1 Predictions Min -86.6521 +trainer/Q2 Predictions Mean -72.3355 +trainer/Q2 Predictions Std 17.6823 +trainer/Q2 Predictions Max -2.80415 +trainer/Q2 Predictions Min -86.9199 +trainer/Q Targets Mean -72.4954 +trainer/Q Targets Std 17.8382 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1501 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0275112 +trainer/policy/mean Std 0.718589 +trainer/policy/mean Max 0.997143 +trainer/policy/mean Min -0.995887 +trainer/policy/std Mean 0.426919 +trainer/policy/std Std 0.0201507 +trainer/policy/std Max 0.447765 +trainer/policy/std Min 0.390913 +trainer/Advantage Weights Mean 2.71765 +trainer/Advantage Weights Std 11.8806 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.7092e-14 +trainer/Advantage Score Mean -0.391357 +trainer/Advantage Score Std 0.613917 +trainer/Advantage Score Max 0.963827 +trainer/Advantage Score Min -3.06867 +trainer/V1 Predictions Mean -72.1559 +trainer/V1 Predictions Std 18.0329 +trainer/V1 Predictions Max -1.68217 +trainer/V1 Predictions Min -87.1249 +trainer/VF Loss 0.0624828 +expl/num steps total 280000 +expl/num paths total 310 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0742098 +expl/Actions Std 0.838313 +expl/Actions Max 2.40017 +expl/Actions Min -2.44014 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 270056 +eval/num paths total 280 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.249934 +eval/Actions Std 0.773416 +eval/Actions Max 0.997679 +eval/Actions Min -0.995319 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.60817e-06 +time/evaluation sampling (s) 2.59513 +time/exploration sampling (s) 3.56463 +time/logging (s) 0.00750289 +time/saving (s) 0.0120053 +time/training (s) 13.1416 +time/epoch (s) 19.3208 +time/total (s) 5920.36 +Epoch -721 +------------------------------ ---------------- +2022-05-15 19:41:26.342062 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -720 finished +------------------------------ ---------------- +epoch -720 +replay_buffer/size 999047 +trainer/num train calls 281000 +trainer/QF1 Loss 19.0259 +trainer/QF2 Loss 19.0722 +trainer/Policy Loss 3.24371 +trainer/Q1 Predictions Mean -70.7995 +trainer/Q1 Predictions Std 18.3215 +trainer/Q1 Predictions Max -0.322564 +trainer/Q1 Predictions Min -86.1095 +trainer/Q2 Predictions Mean -70.8331 +trainer/Q2 Predictions Std 18.2243 +trainer/Q2 Predictions Max -0.50655 +trainer/Q2 Predictions Min -86.0292 +trainer/Q Targets Mean -70.1612 +trainer/Q Targets Std 18.358 +trainer/Q Targets Max -1.98687 +trainer/Q Targets Min -85.9116 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00863383 +trainer/policy/mean Std 0.717886 +trainer/policy/mean Max 0.99513 +trainer/policy/mean Min -0.997903 +trainer/policy/std Mean 0.42664 +trainer/policy/std Std 0.0205727 +trainer/policy/std Max 0.447928 +trainer/policy/std Min 0.392762 +trainer/Advantage Weights Mean 1.46969 +trainer/Advantage Weights Std 10.8988 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.27658e-18 +trainer/Advantage Score Mean -0.709728 +trainer/Advantage Score Std 0.614847 +trainer/Advantage Score Max 0.94302 +trainer/Advantage Score Min -4.06239 +trainer/V1 Predictions Mean -70.112 +trainer/V1 Predictions Std 18.3156 +trainer/V1 Predictions Max -0.12284 +trainer/V1 Predictions Min -85.808 +trainer/VF Loss 0.0952735 +expl/num steps total 281000 +expl/num paths total 311 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0603971 +expl/Actions Std 0.842585 +expl/Actions Max 2.24045 +expl/Actions Min -2.50914 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 271056 +eval/num paths total 281 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.266426 +eval/Actions Std 0.656063 +eval/Actions Max 0.999667 +eval/Actions Min -0.997904 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.60631e-06 +time/evaluation sampling (s) 2.74641 +time/exploration sampling (s) 3.82147 +time/logging (s) 0.00696751 +time/saving (s) 0.00948563 +time/training (s) 13.1047 +time/epoch (s) 19.6891 +time/total (s) 5940.06 +Epoch -720 +------------------------------ ---------------- +2022-05-15 19:41:46.542288 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -719 finished +------------------------------ ---------------- +epoch -719 +replay_buffer/size 999047 +trainer/num train calls 282000 +trainer/QF1 Loss 0.653513 +trainer/QF2 Loss 0.696332 +trainer/Policy Loss 10.627 +trainer/Q1 Predictions Mean -71.8267 +trainer/Q1 Predictions Std 18.8745 +trainer/Q1 Predictions Max -0.470142 +trainer/Q1 Predictions Min -86.6081 +trainer/Q2 Predictions Mean -71.8559 +trainer/Q2 Predictions Std 18.8479 +trainer/Q2 Predictions Max -0.793718 +trainer/Q2 Predictions Min -86.3305 +trainer/Q Targets Mean -71.4973 +trainer/Q Targets Std 18.9258 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.8004 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0221863 +trainer/policy/mean Std 0.721434 +trainer/policy/mean Max 0.998362 +trainer/policy/mean Min -0.995449 +trainer/policy/std Mean 0.428372 +trainer/policy/std Std 0.0207448 +trainer/policy/std Max 0.451291 +trainer/policy/std Min 0.392739 +trainer/Advantage Weights Mean 1.22043 +trainer/Advantage Weights Std 8.10593 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.08649e-13 +trainer/Advantage Score Mean -0.499697 +trainer/Advantage Score Std 0.496914 +trainer/Advantage Score Max 1.13473 +trainer/Advantage Score Min -2.77268 +trainer/V1 Predictions Mean -71.2335 +trainer/V1 Predictions Std 19.117 +trainer/V1 Predictions Max 2.22338 +trainer/V1 Predictions Min -85.6943 +trainer/VF Loss 0.0558515 +expl/num steps total 282000 +expl/num paths total 312 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0865008 +expl/Actions Std 0.81608 +expl/Actions Max 2.35599 +expl/Actions Min -2.13109 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 271561 +eval/num paths total 282 +eval/path length Mean 505 +eval/path length Std 0 +eval/path length Max 505 +eval/path length Min 505 +eval/Rewards Mean 0.0019802 +eval/Rewards Std 0.0444553 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.051732 +eval/Actions Std 0.731674 +eval/Actions Max 0.999 +eval/Actions Min -0.998673 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.19723e-06 +time/evaluation sampling (s) 2.87685 +time/exploration sampling (s) 3.80603 +time/logging (s) 0.00604135 +time/saving (s) 0.00987712 +time/training (s) 13.494 +time/epoch (s) 20.1928 +time/total (s) 5960.25 +Epoch -719 +------------------------------ ---------------- +2022-05-15 19:42:06.502962 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -718 finished +------------------------------ ---------------- +epoch -718 +replay_buffer/size 999047 +trainer/num train calls 283000 +trainer/QF1 Loss 0.992011 +trainer/QF2 Loss 0.98046 +trainer/Policy Loss 9.40598 +trainer/Q1 Predictions Mean -72.1472 +trainer/Q1 Predictions Std 17.9096 +trainer/Q1 Predictions Max -0.456043 +trainer/Q1 Predictions Min -86.2699 +trainer/Q2 Predictions Mean -72.1445 +trainer/Q2 Predictions Std 17.8997 +trainer/Q2 Predictions Max -0.373441 +trainer/Q2 Predictions Min -86.2647 +trainer/Q Targets Mean -71.6624 +trainer/Q Targets Std 17.8404 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.8259 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0104802 +trainer/policy/mean Std 0.709492 +trainer/policy/mean Max 0.99848 +trainer/policy/mean Min -0.999061 +trainer/policy/std Mean 0.427733 +trainer/policy/std Std 0.0206616 +trainer/policy/std Max 0.452042 +trainer/policy/std Min 0.392943 +trainer/Advantage Weights Mean 3.09618 +trainer/Advantage Weights Std 15.8114 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.26011e-15 +trainer/Advantage Score Mean -0.432271 +trainer/Advantage Score Std 0.503638 +trainer/Advantage Score Max 1.37787 +trainer/Advantage Score Min -3.23131 +trainer/V1 Predictions Mean -71.4249 +trainer/V1 Predictions Std 17.858 +trainer/V1 Predictions Max 0.529276 +trainer/V1 Predictions Min -85.6997 +trainer/VF Loss 0.0620984 +expl/num steps total 283000 +expl/num paths total 314 +expl/path length Mean 500 +expl/path length Std 271 +expl/path length Max 771 +expl/path length Min 229 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0463974 +expl/Actions Std 0.823421 +expl/Actions Max 2.46892 +expl/Actions Min -2.45167 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 272561 +eval/num paths total 283 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0284945 +eval/Actions Std 0.717062 +eval/Actions Max 0.998903 +eval/Actions Min -0.999435 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.19211e-06 +time/evaluation sampling (s) 2.89637 +time/exploration sampling (s) 3.72292 +time/logging (s) 0.00887603 +time/saving (s) 0.00966026 +time/training (s) 13.3193 +time/epoch (s) 19.9572 +time/total (s) 5980.22 +Epoch -718 +------------------------------ ---------------- +2022-05-15 19:42:26.316458 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -717 finished +------------------------------ ---------------- +epoch -717 +replay_buffer/size 999047 +trainer/num train calls 284000 +trainer/QF1 Loss 0.778728 +trainer/QF2 Loss 0.736214 +trainer/Policy Loss 7.23356 +trainer/Q1 Predictions Mean -73.9686 +trainer/Q1 Predictions Std 14.8287 +trainer/Q1 Predictions Max -4.07344 +trainer/Q1 Predictions Min -85.7527 +trainer/Q2 Predictions Mean -73.966 +trainer/Q2 Predictions Std 14.8727 +trainer/Q2 Predictions Max -4.15384 +trainer/Q2 Predictions Min -85.698 +trainer/Q Targets Mean -74.1642 +trainer/Q Targets Std 14.5408 +trainer/Q Targets Max -2.86686 +trainer/Q Targets Min -85.857 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0302232 +trainer/policy/mean Std 0.718055 +trainer/policy/mean Max 0.999163 +trainer/policy/mean Min -0.998917 +trainer/policy/std Mean 0.427129 +trainer/policy/std Std 0.0203559 +trainer/policy/std Max 0.448965 +trainer/policy/std Min 0.394454 +trainer/Advantage Weights Mean 2.03404 +trainer/Advantage Weights Std 11.1337 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.84388e-10 +trainer/Advantage Score Mean -0.366655 +trainer/Advantage Score Std 0.401314 +trainer/Advantage Score Max 1.01935 +trainer/Advantage Score Min -2.14481 +trainer/V1 Predictions Mean -73.8973 +trainer/V1 Predictions Std 14.7994 +trainer/V1 Predictions Max -3.51446 +trainer/V1 Predictions Min -85.6722 +trainer/VF Loss 0.0375667 +expl/num steps total 284000 +expl/num paths total 315 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0069388 +expl/Actions Std 0.820029 +expl/Actions Max 2.40656 +expl/Actions Min -2.26516 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 273561 +eval/num paths total 284 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00489498 +eval/Actions Std 0.692219 +eval/Actions Max 0.99914 +eval/Actions Min -0.998906 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.96906e-06 +time/evaluation sampling (s) 2.9347 +time/exploration sampling (s) 3.00222 +time/logging (s) 0.00866174 +time/saving (s) 0.0120536 +time/training (s) 13.8419 +time/epoch (s) 19.7995 +time/total (s) 6000.02 +Epoch -717 +------------------------------ ---------------- +2022-05-15 19:42:46.000350 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -716 finished +------------------------------ ---------------- +epoch -716 +replay_buffer/size 999047 +trainer/num train calls 285000 +trainer/QF1 Loss 0.505975 +trainer/QF2 Loss 0.619798 +trainer/Policy Loss 23.0435 +trainer/Q1 Predictions Mean -72.0469 +trainer/Q1 Predictions Std 18.0031 +trainer/Q1 Predictions Max -0.617655 +trainer/Q1 Predictions Min -86.2731 +trainer/Q2 Predictions Mean -72.0744 +trainer/Q2 Predictions Std 17.9424 +trainer/Q2 Predictions Max -0.461484 +trainer/Q2 Predictions Min -86.2995 +trainer/Q Targets Mean -72.0106 +trainer/Q Targets Std 18.2446 +trainer/Q Targets Max 0.72611 +trainer/Q Targets Min -86.4431 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0034098 +trainer/policy/mean Std 0.707767 +trainer/policy/mean Max 0.999521 +trainer/policy/mean Min -0.998451 +trainer/policy/std Mean 0.42843 +trainer/policy/std Std 0.0200089 +trainer/policy/std Max 0.452199 +trainer/policy/std Min 0.394488 +trainer/Advantage Weights Mean 5.10196 +trainer/Advantage Weights Std 17.2231 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.24992e-15 +trainer/Advantage Score Mean -0.232819 +trainer/Advantage Score Std 0.531719 +trainer/Advantage Score Max 0.77285 +trainer/Advantage Score Min -3.43157 +trainer/V1 Predictions Mean -71.7653 +trainer/V1 Predictions Std 18.2105 +trainer/V1 Predictions Max -0.136229 +trainer/V1 Predictions Min -86.2795 +trainer/VF Loss 0.0499503 +expl/num steps total 285000 +expl/num paths total 316 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0699919 +expl/Actions Std 0.821197 +expl/Actions Max 2.51169 +expl/Actions Min -2.27593 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 274561 +eval/num paths total 285 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.462472 +eval/Actions Std 0.68816 +eval/Actions Max 0.998724 +eval/Actions Min -0.998355 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.65799e-06 +time/evaluation sampling (s) 2.54659 +time/exploration sampling (s) 3.60978 +time/logging (s) 0.00684227 +time/saving (s) 0.00931086 +time/training (s) 13.5019 +time/epoch (s) 19.6744 +time/total (s) 6019.7 +Epoch -716 +------------------------------ ---------------- +2022-05-15 19:43:05.853367 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -715 finished +------------------------------ ---------------- +epoch -715 +replay_buffer/size 999047 +trainer/num train calls 286000 +trainer/QF1 Loss 0.800013 +trainer/QF2 Loss 0.751133 +trainer/Policy Loss 10.134 +trainer/Q1 Predictions Mean -73.84 +trainer/Q1 Predictions Std 15.8144 +trainer/Q1 Predictions Max -0.286598 +trainer/Q1 Predictions Min -87.1869 +trainer/Q2 Predictions Mean -73.9019 +trainer/Q2 Predictions Std 15.7647 +trainer/Q2 Predictions Max -0.346443 +trainer/Q2 Predictions Min -87.1786 +trainer/Q Targets Mean -73.7943 +trainer/Q Targets Std 15.3747 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6613 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00692944 +trainer/policy/mean Std 0.723422 +trainer/policy/mean Max 0.999461 +trainer/policy/mean Min -0.998718 +trainer/policy/std Mean 0.427479 +trainer/policy/std Std 0.0206717 +trainer/policy/std Max 0.449646 +trainer/policy/std Min 0.39093 +trainer/Advantage Weights Mean 2.26229 +trainer/Advantage Weights Std 11.8059 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.59638e-13 +trainer/Advantage Score Mean -0.419923 +trainer/Advantage Score Std 0.455968 +trainer/Advantage Score Max 1.14497 +trainer/Advantage Score Min -2.86537 +trainer/V1 Predictions Mean -73.4758 +trainer/V1 Predictions Std 15.6534 +trainer/V1 Predictions Max -0.357891 +trainer/V1 Predictions Min -86.3293 +trainer/VF Loss 0.0472676 +expl/num steps total 286000 +expl/num paths total 317 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.150773 +expl/Actions Std 0.832419 +expl/Actions Max 2.34644 +expl/Actions Min -2.7774 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 275492 +eval/num paths total 286 +eval/path length Mean 931 +eval/path length Std 0 +eval/path length Max 931 +eval/path length Min 931 +eval/Rewards Mean 0.00107411 +eval/Rewards Std 0.0327561 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0642625 +eval/Actions Std 0.730732 +eval/Actions Max 0.999601 +eval/Actions Min -0.999301 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.47826e-06 +time/evaluation sampling (s) 2.96032 +time/exploration sampling (s) 3.83582 +time/logging (s) 0.00679626 +time/saving (s) 0.00959107 +time/training (s) 13.0339 +time/epoch (s) 19.8464 +time/total (s) 6039.55 +Epoch -715 +------------------------------ ---------------- +2022-05-15 19:43:26.281292 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -714 finished +------------------------------ ---------------- +epoch -714 +replay_buffer/size 999047 +trainer/num train calls 287000 +trainer/QF1 Loss 0.614379 +trainer/QF2 Loss 0.657939 +trainer/Policy Loss 23.7441 +trainer/Q1 Predictions Mean -72.4811 +trainer/Q1 Predictions Std 17.4788 +trainer/Q1 Predictions Max -0.261068 +trainer/Q1 Predictions Min -86.1939 +trainer/Q2 Predictions Mean -72.4281 +trainer/Q2 Predictions Std 17.454 +trainer/Q2 Predictions Max -0.390414 +trainer/Q2 Predictions Min -86.2813 +trainer/Q Targets Mean -72.6606 +trainer/Q Targets Std 17.3406 +trainer/Q Targets Max 2.41064 +trainer/Q Targets Min -85.98 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0255875 +trainer/policy/mean Std 0.700458 +trainer/policy/mean Max 0.998898 +trainer/policy/mean Min -0.998149 +trainer/policy/std Mean 0.425732 +trainer/policy/std Std 0.0211766 +trainer/policy/std Max 0.448955 +trainer/policy/std Min 0.38851 +trainer/Advantage Weights Mean 6.30454 +trainer/Advantage Weights Std 21.1378 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.72118e-15 +trainer/Advantage Score Mean -0.256348 +trainer/Advantage Score Std 0.516532 +trainer/Advantage Score Max 1.21454 +trainer/Advantage Score Min -3.29867 +trainer/V1 Predictions Mean -72.3615 +trainer/V1 Predictions Std 17.6096 +trainer/V1 Predictions Max 1.10024 +trainer/V1 Predictions Min -85.8628 +trainer/VF Loss 0.0572308 +expl/num steps total 287000 +expl/num paths total 318 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.08539 +expl/Actions Std 0.809018 +expl/Actions Max 2.46039 +expl/Actions Min -2.33687 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 276492 +eval/num paths total 287 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0805524 +eval/Actions Std 0.705654 +eval/Actions Max 0.999699 +eval/Actions Min -0.999102 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.77301e-06 +time/evaluation sampling (s) 2.94174 +time/exploration sampling (s) 3.86696 +time/logging (s) 0.00754688 +time/saving (s) 0.0116768 +time/training (s) 13.5943 +time/epoch (s) 20.4222 +time/total (s) 6059.98 +Epoch -714 +------------------------------ ---------------- +2022-05-15 19:43:45.696956 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -713 finished +------------------------------ ---------------- +epoch -713 +replay_buffer/size 999047 +trainer/num train calls 288000 +trainer/QF1 Loss 0.894545 +trainer/QF2 Loss 0.84931 +trainer/Policy Loss 8.31056 +trainer/Q1 Predictions Mean -72.5456 +trainer/Q1 Predictions Std 16.8602 +trainer/Q1 Predictions Max -0.943373 +trainer/Q1 Predictions Min -86.1975 +trainer/Q2 Predictions Mean -72.5669 +trainer/Q2 Predictions Std 16.887 +trainer/Q2 Predictions Max -1.01304 +trainer/Q2 Predictions Min -86.0099 +trainer/Q Targets Mean -72.4306 +trainer/Q Targets Std 16.9671 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0834 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00951781 +trainer/policy/mean Std 0.708529 +trainer/policy/mean Max 0.999443 +trainer/policy/mean Min -0.997657 +trainer/policy/std Mean 0.425743 +trainer/policy/std Std 0.0197364 +trainer/policy/std Max 0.446948 +trainer/policy/std Min 0.393022 +trainer/Advantage Weights Mean 1.60773 +trainer/Advantage Weights Std 9.01771 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.47514e-16 +trainer/Advantage Score Mean -0.395582 +trainer/Advantage Score Std 0.574959 +trainer/Advantage Score Max 0.661186 +trainer/Advantage Score Min -3.51411 +trainer/V1 Predictions Mean -72.0796 +trainer/V1 Predictions Std 17.2464 +trainer/V1 Predictions Max -0.624081 +trainer/V1 Predictions Min -86.0239 +trainer/VF Loss 0.0533049 +expl/num steps total 288000 +expl/num paths total 320 +expl/path length Mean 500 +expl/path length Std 241 +expl/path length Max 741 +expl/path length Min 259 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0551028 +expl/Actions Std 0.829285 +expl/Actions Max 2.38336 +expl/Actions Min -2.4145 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 276893 +eval/num paths total 288 +eval/path length Mean 401 +eval/path length Std 0 +eval/path length Max 401 +eval/path length Min 401 +eval/Rewards Mean 0.00249377 +eval/Rewards Std 0.0498753 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0363599 +eval/Actions Std 0.740927 +eval/Actions Max 0.9981 +eval/Actions Min -0.998573 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.77488e-06 +time/evaluation sampling (s) 2.90263 +time/exploration sampling (s) 3.47526 +time/logging (s) 0.00538315 +time/saving (s) 0.00923449 +time/training (s) 13.0129 +time/epoch (s) 19.4055 +time/total (s) 6079.39 +Epoch -713 +------------------------------ ---------------- +2022-05-15 19:44:05.040331 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -712 finished +------------------------------ ---------------- +epoch -712 +replay_buffer/size 999047 +trainer/num train calls 289000 +trainer/QF1 Loss 1.93304 +trainer/QF2 Loss 2.11645 +trainer/Policy Loss 3.21117 +trainer/Q1 Predictions Mean -70.5782 +trainer/Q1 Predictions Std 19.7652 +trainer/Q1 Predictions Max -0.657794 +trainer/Q1 Predictions Min -87.2822 +trainer/Q2 Predictions Mean -70.6491 +trainer/Q2 Predictions Std 19.8139 +trainer/Q2 Predictions Max -0.739472 +trainer/Q2 Predictions Min -86.7982 +trainer/Q Targets Mean -69.8055 +trainer/Q Targets Std 19.9202 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4804 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0300224 +trainer/policy/mean Std 0.705453 +trainer/policy/mean Max 0.996886 +trainer/policy/mean Min -0.998358 +trainer/policy/std Mean 0.42585 +trainer/policy/std Std 0.020733 +trainer/policy/std Max 0.447182 +trainer/policy/std Min 0.390642 +trainer/Advantage Weights Mean 1.25894 +trainer/Advantage Weights Std 10.004 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.74797e-22 +trainer/Advantage Score Mean -0.691842 +trainer/Advantage Score Std 0.62586 +trainer/Advantage Score Max 0.893059 +trainer/Advantage Score Min -5.00984 +trainer/V1 Predictions Mean -69.5511 +trainer/V1 Predictions Std 20.0071 +trainer/V1 Predictions Max 0.550604 +trainer/V1 Predictions Min -86.2937 +trainer/VF Loss 0.0915341 +expl/num steps total 289000 +expl/num paths total 321 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0461264 +expl/Actions Std 0.810655 +expl/Actions Max 2.5107 +expl/Actions Min -2.47159 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 277893 +eval/num paths total 289 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0920229 +eval/Actions Std 0.72059 +eval/Actions Max 0.999487 +eval/Actions Min -0.999452 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.18373e-06 +time/evaluation sampling (s) 2.88411 +time/exploration sampling (s) 2.98405 +time/logging (s) 0.00732589 +time/saving (s) 0.0102509 +time/training (s) 13.4535 +time/epoch (s) 19.3392 +time/total (s) 6098.73 +Epoch -712 +------------------------------ ---------------- +2022-05-15 19:44:24.169811 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -711 finished +------------------------------ ---------------- +epoch -711 +replay_buffer/size 999047 +trainer/num train calls 290000 +trainer/QF1 Loss 0.926354 +trainer/QF2 Loss 0.951264 +trainer/Policy Loss 13.0548 +trainer/Q1 Predictions Mean -72.8475 +trainer/Q1 Predictions Std 15.9986 +trainer/Q1 Predictions Max -0.878909 +trainer/Q1 Predictions Min -85.5562 +trainer/Q2 Predictions Mean -72.8197 +trainer/Q2 Predictions Std 16.0313 +trainer/Q2 Predictions Max -0.829245 +trainer/Q2 Predictions Min -85.5686 +trainer/Q Targets Mean -73.2659 +trainer/Q Targets Std 15.6533 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9567 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0124971 +trainer/policy/mean Std 0.718075 +trainer/policy/mean Max 0.997711 +trainer/policy/mean Min -0.997413 +trainer/policy/std Mean 0.426571 +trainer/policy/std Std 0.0208526 +trainer/policy/std Max 0.448584 +trainer/policy/std Min 0.39246 +trainer/Advantage Weights Mean 4.17664 +trainer/Advantage Weights Std 16.0396 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.56492e-11 +trainer/Advantage Score Mean -0.249786 +trainer/Advantage Score Std 0.440081 +trainer/Advantage Score Max 1.61818 +trainer/Advantage Score Min -2.48806 +trainer/V1 Predictions Mean -72.9761 +trainer/V1 Predictions Std 15.9058 +trainer/V1 Predictions Max -0.592533 +trainer/V1 Predictions Min -85.7532 +trainer/VF Loss 0.0479582 +expl/num steps total 290000 +expl/num paths total 322 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0773242 +expl/Actions Std 0.80135 +expl/Actions Max 2.58308 +expl/Actions Min -2.75208 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 278727 +eval/num paths total 290 +eval/path length Mean 834 +eval/path length Std 0 +eval/path length Max 834 +eval/path length Min 834 +eval/Rewards Mean 0.00119904 +eval/Rewards Std 0.0346064 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0531289 +eval/Actions Std 0.723502 +eval/Actions Max 0.999142 +eval/Actions Min -0.998195 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.94996e-06 +time/evaluation sampling (s) 2.42939 +time/exploration sampling (s) 3.47034 +time/logging (s) 0.00631096 +time/saving (s) 0.00945636 +time/training (s) 13.2062 +time/epoch (s) 19.1217 +time/total (s) 6117.85 +Epoch -711 +------------------------------ ---------------- +2022-05-15 19:44:44.071740 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -710 finished +------------------------------ ---------------- +epoch -710 +replay_buffer/size 999047 +trainer/num train calls 291000 +trainer/QF1 Loss 0.965482 +trainer/QF2 Loss 0.94095 +trainer/Policy Loss 26.1108 +trainer/Q1 Predictions Mean -70.6886 +trainer/Q1 Predictions Std 19.2107 +trainer/Q1 Predictions Max -2.02804 +trainer/Q1 Predictions Min -87.6262 +trainer/Q2 Predictions Mean -70.7214 +trainer/Q2 Predictions Std 19.1802 +trainer/Q2 Predictions Max -1.76176 +trainer/Q2 Predictions Min -86.972 +trainer/Q Targets Mean -70.5402 +trainer/Q Targets Std 19.0982 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0646 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0209285 +trainer/policy/mean Std 0.714638 +trainer/policy/mean Max 0.998569 +trainer/policy/mean Min -0.998299 +trainer/policy/std Mean 0.427407 +trainer/policy/std Std 0.0206642 +trainer/policy/std Max 0.449018 +trainer/policy/std Min 0.392768 +trainer/Advantage Weights Mean 6.44051 +trainer/Advantage Weights Std 22.7865 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.36361e-17 +trainer/Advantage Score Mean -0.293237 +trainer/Advantage Score Std 0.588179 +trainer/Advantage Score Max 2.37007 +trainer/Advantage Score Min -3.72933 +trainer/V1 Predictions Mean -70.2997 +trainer/V1 Predictions Std 19.0308 +trainer/V1 Predictions Max -2.97671 +trainer/V1 Predictions Min -85.8978 +trainer/VF Loss 0.114233 +expl/num steps total 291000 +expl/num paths total 323 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.119879 +expl/Actions Std 0.851962 +expl/Actions Max 2.36765 +expl/Actions Min -2.36369 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 279727 +eval/num paths total 291 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0565506 +eval/Actions Std 0.756255 +eval/Actions Max 0.999445 +eval/Actions Min -0.998971 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.85264e-06 +time/evaluation sampling (s) 2.82159 +time/exploration sampling (s) 3.76265 +time/logging (s) 0.00706206 +time/saving (s) 0.00989252 +time/training (s) 13.295 +time/epoch (s) 19.8962 +time/total (s) 6137.75 +Epoch -710 +------------------------------ ---------------- +2022-05-15 19:45:04.388689 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -709 finished +------------------------------ ---------------- +epoch -709 +replay_buffer/size 999047 +trainer/num train calls 292000 +trainer/QF1 Loss 0.781844 +trainer/QF2 Loss 0.884571 +trainer/Policy Loss 18.8046 +trainer/Q1 Predictions Mean -71.4319 +trainer/Q1 Predictions Std 17.2078 +trainer/Q1 Predictions Max -0.493357 +trainer/Q1 Predictions Min -86.225 +trainer/Q2 Predictions Mean -71.3612 +trainer/Q2 Predictions Std 17.1999 +trainer/Q2 Predictions Max -0.39609 +trainer/Q2 Predictions Min -86.0068 +trainer/Q Targets Mean -71.7678 +trainer/Q Targets Std 17.3688 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9925 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000907265 +trainer/policy/mean Std 0.718649 +trainer/policy/mean Max 0.999612 +trainer/policy/mean Min -0.997365 +trainer/policy/std Mean 0.424879 +trainer/policy/std Std 0.0217902 +trainer/policy/std Max 0.446365 +trainer/policy/std Min 0.388002 +trainer/Advantage Weights Mean 3.20737 +trainer/Advantage Weights Std 15.5029 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.08832e-18 +trainer/Advantage Score Mean -0.510085 +trainer/Advantage Score Std 0.666534 +trainer/Advantage Score Max 1.74251 +trainer/Advantage Score Min -3.98196 +trainer/V1 Predictions Mean -71.556 +trainer/V1 Predictions Std 17.4444 +trainer/V1 Predictions Max 1.69247 +trainer/V1 Predictions Min -86.7313 +trainer/VF Loss 0.0892519 +expl/num steps total 292000 +expl/num paths total 324 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0355476 +expl/Actions Std 0.838952 +expl/Actions Max 2.53717 +expl/Actions Min -2.2179 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 280727 +eval/num paths total 292 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.249313 +eval/Actions Std 0.708831 +eval/Actions Max 0.998301 +eval/Actions Min -0.998819 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.77022e-06 +time/evaluation sampling (s) 3.11586 +time/exploration sampling (s) 3.73513 +time/logging (s) 0.0080766 +time/saving (s) 0.0125068 +time/training (s) 13.439 +time/epoch (s) 20.3106 +time/total (s) 6158.07 +Epoch -709 +------------------------------ ---------------- +2022-05-15 19:45:24.334561 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -708 finished +------------------------------ ---------------- +epoch -708 +replay_buffer/size 999047 +trainer/num train calls 293000 +trainer/QF1 Loss 0.708471 +trainer/QF2 Loss 0.693317 +trainer/Policy Loss 22.6924 +trainer/Q1 Predictions Mean -70.9227 +trainer/Q1 Predictions Std 19.7691 +trainer/Q1 Predictions Max -0.834557 +trainer/Q1 Predictions Min -86.15 +trainer/Q2 Predictions Mean -70.9606 +trainer/Q2 Predictions Std 19.735 +trainer/Q2 Predictions Max -0.827597 +trainer/Q2 Predictions Min -86.0501 +trainer/Q Targets Mean -71.099 +trainer/Q Targets Std 19.7295 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8334 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00431803 +trainer/policy/mean Std 0.718844 +trainer/policy/mean Max 0.998255 +trainer/policy/mean Min -0.998412 +trainer/policy/std Mean 0.426089 +trainer/policy/std Std 0.0221587 +trainer/policy/std Max 0.451755 +trainer/policy/std Min 0.388465 +trainer/Advantage Weights Mean 5.27328 +trainer/Advantage Weights Std 19.5061 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.64653e-13 +trainer/Advantage Score Mean -0.264501 +trainer/Advantage Score Std 0.512717 +trainer/Advantage Score Max 2.49191 +trainer/Advantage Score Min -2.7667 +trainer/V1 Predictions Mean -70.822 +trainer/V1 Predictions Std 19.8253 +trainer/V1 Predictions Max 0.464045 +trainer/V1 Predictions Min -86.4369 +trainer/VF Loss 0.0788657 +expl/num steps total 293000 +expl/num paths total 325 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0516481 +expl/Actions Std 0.794247 +expl/Actions Max 2.38623 +expl/Actions Min -2.2286 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 281727 +eval/num paths total 293 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0364775 +eval/Actions Std 0.669641 +eval/Actions Max 0.998803 +eval/Actions Min -0.999473 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.5453e-06 +time/evaluation sampling (s) 3.06699 +time/exploration sampling (s) 3.56974 +time/logging (s) 0.00691882 +time/saving (s) 0.00981132 +time/training (s) 13.2828 +time/epoch (s) 19.9363 +time/total (s) 6178.01 +Epoch -708 +------------------------------ ---------------- +2022-05-15 19:45:44.339813 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -707 finished +------------------------------ ---------------- +epoch -707 +replay_buffer/size 999047 +trainer/num train calls 294000 +trainer/QF1 Loss 0.508236 +trainer/QF2 Loss 0.515158 +trainer/Policy Loss 19.968 +trainer/Q1 Predictions Mean -72.5472 +trainer/Q1 Predictions Std 17.4952 +trainer/Q1 Predictions Max -0.646378 +trainer/Q1 Predictions Min -86.1008 +trainer/Q2 Predictions Mean -72.5484 +trainer/Q2 Predictions Std 17.4801 +trainer/Q2 Predictions Max -0.565605 +trainer/Q2 Predictions Min -86.2395 +trainer/Q Targets Mean -72.7598 +trainer/Q Targets Std 17.4898 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.8161 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00324686 +trainer/policy/mean Std 0.721283 +trainer/policy/mean Max 0.999387 +trainer/policy/mean Min -0.999239 +trainer/policy/std Mean 0.42717 +trainer/policy/std Std 0.0218559 +trainer/policy/std Max 0.449608 +trainer/policy/std Min 0.390176 +trainer/Advantage Weights Mean 3.49097 +trainer/Advantage Weights Std 15.7687 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.14531e-12 +trainer/Advantage Score Mean -0.392153 +trainer/Advantage Score Std 0.496896 +trainer/Advantage Score Max 1.07459 +trainer/Advantage Score Min -2.74953 +trainer/V1 Predictions Mean -72.464 +trainer/V1 Predictions Std 17.6108 +trainer/V1 Predictions Max -0.439728 +trainer/V1 Predictions Min -85.76 +trainer/VF Loss 0.0527871 +expl/num steps total 294000 +expl/num paths total 326 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.143599 +expl/Actions Std 0.909397 +expl/Actions Max 2.32457 +expl/Actions Min -2.635 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 282727 +eval/num paths total 294 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0795796 +eval/Actions Std 0.723881 +eval/Actions Max 0.999122 +eval/Actions Min -0.998937 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.74787e-06 +time/evaluation sampling (s) 3.1242 +time/exploration sampling (s) 3.09786 +time/logging (s) 0.00916007 +time/saving (s) 0.0136931 +time/training (s) 13.7555 +time/epoch (s) 20.0004 +time/total (s) 6198.02 +Epoch -707 +------------------------------ ---------------- +2022-05-15 19:46:03.872296 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -706 finished +------------------------------ ---------------- +epoch -706 +replay_buffer/size 999047 +trainer/num train calls 295000 +trainer/QF1 Loss 0.622053 +trainer/QF2 Loss 0.600624 +trainer/Policy Loss 27.7043 +trainer/Q1 Predictions Mean -74.0197 +trainer/Q1 Predictions Std 14.2691 +trainer/Q1 Predictions Max -0.904487 +trainer/Q1 Predictions Min -86.6518 +trainer/Q2 Predictions Mean -74.0584 +trainer/Q2 Predictions Std 14.2333 +trainer/Q2 Predictions Max -0.590066 +trainer/Q2 Predictions Min -86.6171 +trainer/Q Targets Mean -74.267 +trainer/Q Targets Std 14.069 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4029 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0427471 +trainer/policy/mean Std 0.730993 +trainer/policy/mean Max 0.999301 +trainer/policy/mean Min -0.998359 +trainer/policy/std Mean 0.425946 +trainer/policy/std Std 0.0206026 +trainer/policy/std Max 0.448499 +trainer/policy/std Min 0.391541 +trainer/Advantage Weights Mean 6.12362 +trainer/Advantage Weights Std 20.3134 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.76174e-13 +trainer/Advantage Score Mean -0.268404 +trainer/Advantage Score Std 0.469164 +trainer/Advantage Score Max 0.931538 +trainer/Advantage Score Min -2.8373 +trainer/V1 Predictions Mean -73.936 +trainer/V1 Predictions Std 14.2254 +trainer/V1 Predictions Max -0.799957 +trainer/V1 Predictions Min -86.4042 +trainer/VF Loss 0.048071 +expl/num steps total 295000 +expl/num paths total 327 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.048395 +expl/Actions Std 0.834753 +expl/Actions Max 2.5288 +expl/Actions Min -2.53982 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 283727 +eval/num paths total 295 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.154686 +eval/Actions Std 0.765094 +eval/Actions Max 0.99873 +eval/Actions Min -0.997336 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.55555e-06 +time/evaluation sampling (s) 2.63009 +time/exploration sampling (s) 3.27858 +time/logging (s) 0.00677189 +time/saving (s) 0.0098751 +time/training (s) 13.5955 +time/epoch (s) 19.5208 +time/total (s) 6217.54 +Epoch -706 +------------------------------ ---------------- +2022-05-15 19:46:23.600748 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -705 finished +------------------------------ ---------------- +epoch -705 +replay_buffer/size 999047 +trainer/num train calls 296000 +trainer/QF1 Loss 0.568409 +trainer/QF2 Loss 0.660404 +trainer/Policy Loss 6.39289 +trainer/Q1 Predictions Mean -71.6796 +trainer/Q1 Predictions Std 19.4236 +trainer/Q1 Predictions Max -0.613319 +trainer/Q1 Predictions Min -85.313 +trainer/Q2 Predictions Mean -71.6489 +trainer/Q2 Predictions Std 19.3986 +trainer/Q2 Predictions Max -0.615856 +trainer/Q2 Predictions Min -85.4899 +trainer/Q Targets Mean -71.4921 +trainer/Q Targets Std 19.567 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.6466 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00103575 +trainer/policy/mean Std 0.717578 +trainer/policy/mean Max 0.9979 +trainer/policy/mean Min -0.998598 +trainer/policy/std Mean 0.424238 +trainer/policy/std Std 0.0213565 +trainer/policy/std Max 0.446965 +trainer/policy/std Min 0.38529 +trainer/Advantage Weights Mean 1.43333 +trainer/Advantage Weights Std 9.52752 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.11995e-13 +trainer/Advantage Score Mean -0.550236 +trainer/Advantage Score Std 0.559638 +trainer/Advantage Score Max 1.13631 +trainer/Advantage Score Min -2.87958 +trainer/V1 Predictions Mean -71.2816 +trainer/V1 Predictions Std 19.5864 +trainer/V1 Predictions Max 0.2491 +trainer/V1 Predictions Min -85.4288 +trainer/VF Loss 0.0680932 +expl/num steps total 296000 +expl/num paths total 328 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.12269 +expl/Actions Std 0.816934 +expl/Actions Max 2.57444 +expl/Actions Min -2.47013 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 284727 +eval/num paths total 296 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.353494 +eval/Actions Std 0.627306 +eval/Actions Max 0.999398 +eval/Actions Min -0.998638 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.1379e-06 +time/evaluation sampling (s) 2.79211 +time/exploration sampling (s) 3.33372 +time/logging (s) 0.00699628 +time/saving (s) 0.00998859 +time/training (s) 13.5794 +time/epoch (s) 19.7222 +time/total (s) 6237.27 +Epoch -705 +------------------------------ ---------------- +2022-05-15 19:46:43.613892 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -704 finished +------------------------------ ---------------- +epoch -704 +replay_buffer/size 999047 +trainer/num train calls 297000 +trainer/QF1 Loss 0.4522 +trainer/QF2 Loss 0.452883 +trainer/Policy Loss 10.4309 +trainer/Q1 Predictions Mean -73.0474 +trainer/Q1 Predictions Std 16.0394 +trainer/Q1 Predictions Max -1.18049 +trainer/Q1 Predictions Min -86.5819 +trainer/Q2 Predictions Mean -73.0056 +trainer/Q2 Predictions Std 16.0921 +trainer/Q2 Predictions Max -1.28974 +trainer/Q2 Predictions Min -86.8522 +trainer/Q Targets Mean -73.0089 +trainer/Q Targets Std 16.0956 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6935 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00489327 +trainer/policy/mean Std 0.715918 +trainer/policy/mean Max 0.999689 +trainer/policy/mean Min -0.998336 +trainer/policy/std Mean 0.425594 +trainer/policy/std Std 0.0205659 +trainer/policy/std Max 0.449566 +trainer/policy/std Min 0.388942 +trainer/Advantage Weights Mean 3.96334 +trainer/Advantage Weights Std 16.2002 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.48002e-12 +trainer/Advantage Score Mean -0.299206 +trainer/Advantage Score Std 0.421556 +trainer/Advantage Score Max 1.01579 +trainer/Advantage Score Min -2.67228 +trainer/V1 Predictions Mean -72.759 +trainer/V1 Predictions Std 16.1365 +trainer/V1 Predictions Max -0.366464 +trainer/V1 Predictions Min -86.689 +trainer/VF Loss 0.0404452 +expl/num steps total 297000 +expl/num paths total 329 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0595804 +expl/Actions Std 0.817814 +expl/Actions Max 2.29906 +expl/Actions Min -2.24954 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 285727 +eval/num paths total 297 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0626773 +eval/Actions Std 0.753625 +eval/Actions Max 0.999616 +eval/Actions Min -0.998878 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.03891e-06 +time/evaluation sampling (s) 3.0496 +time/exploration sampling (s) 3.64507 +time/logging (s) 0.00737688 +time/saving (s) 0.0117224 +time/training (s) 13.2932 +time/epoch (s) 20.0069 +time/total (s) 6257.28 +Epoch -704 +------------------------------ ---------------- +2022-05-15 19:47:03.867135 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -703 finished +------------------------------ ---------------- +epoch -703 +replay_buffer/size 999047 +trainer/num train calls 298000 +trainer/QF1 Loss 0.962513 +trainer/QF2 Loss 0.968798 +trainer/Policy Loss 37.4126 +trainer/Q1 Predictions Mean -70.648 +trainer/Q1 Predictions Std 20.4632 +trainer/Q1 Predictions Max -0.423565 +trainer/Q1 Predictions Min -86.6231 +trainer/Q2 Predictions Mean -70.6168 +trainer/Q2 Predictions Std 20.4593 +trainer/Q2 Predictions Max -0.391791 +trainer/Q2 Predictions Min -86.8174 +trainer/Q Targets Mean -70.4584 +trainer/Q Targets Std 20.1734 +trainer/Q Targets Max 0.146079 +trainer/Q Targets Min -86.4523 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0048545 +trainer/policy/mean Std 0.714788 +trainer/policy/mean Max 0.998236 +trainer/policy/mean Min -0.999563 +trainer/policy/std Mean 0.425238 +trainer/policy/std Std 0.0199766 +trainer/policy/std Max 0.4479 +trainer/policy/std Min 0.392087 +trainer/Advantage Weights Mean 5.88505 +trainer/Advantage Weights Std 21.02 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.25092e-15 +trainer/Advantage Score Mean -0.350667 +trainer/Advantage Score Std 0.566753 +trainer/Advantage Score Max 2.1214 +trainer/Advantage Score Min -3.30916 +trainer/V1 Predictions Mean -70.2235 +trainer/V1 Predictions Std 20.2632 +trainer/V1 Predictions Max 1.30386 +trainer/V1 Predictions Min -86.2281 +trainer/VF Loss 0.0991346 +expl/num steps total 298000 +expl/num paths total 330 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0507581 +expl/Actions Std 0.832051 +expl/Actions Max 2.38365 +expl/Actions Min -2.47876 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 286727 +eval/num paths total 298 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.145284 +eval/Actions Std 0.688212 +eval/Actions Max 0.99962 +eval/Actions Min -0.998016 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.57697e-06 +time/evaluation sampling (s) 3.24225 +time/exploration sampling (s) 3.69677 +time/logging (s) 0.00730436 +time/saving (s) 0.0104577 +time/training (s) 13.2884 +time/epoch (s) 20.2452 +time/total (s) 6277.53 +Epoch -703 +------------------------------ ---------------- +2022-05-15 19:47:24.248626 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -702 finished +------------------------------ ---------------- +epoch -702 +replay_buffer/size 999047 +trainer/num train calls 299000 +trainer/QF1 Loss 1.65824 +trainer/QF2 Loss 1.7355 +trainer/Policy Loss 30.2449 +trainer/Q1 Predictions Mean -72.2888 +trainer/Q1 Predictions Std 17.8614 +trainer/Q1 Predictions Max -0.273182 +trainer/Q1 Predictions Min -85.6132 +trainer/Q2 Predictions Mean -72.2882 +trainer/Q2 Predictions Std 17.7922 +trainer/Q2 Predictions Max -0.355902 +trainer/Q2 Predictions Min -85.9099 +trainer/Q Targets Mean -72.6418 +trainer/Q Targets Std 18.405 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9156 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.010971 +trainer/policy/mean Std 0.71994 +trainer/policy/mean Max 0.998791 +trainer/policy/mean Min -0.99832 +trainer/policy/std Mean 0.424874 +trainer/policy/std Std 0.0203795 +trainer/policy/std Max 0.446206 +trainer/policy/std Min 0.391325 +trainer/Advantage Weights Mean 8.56515 +trainer/Advantage Weights Std 23.204 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.45045e-12 +trainer/Advantage Score Mean -0.165424 +trainer/Advantage Score Std 0.493891 +trainer/Advantage Score Max 1.66158 +trainer/Advantage Score Min -2.63925 +trainer/V1 Predictions Mean -72.4759 +trainer/V1 Predictions Std 18.1598 +trainer/V1 Predictions Max 0.067671 +trainer/V1 Predictions Min -85.7926 +trainer/VF Loss 0.0658444 +expl/num steps total 299000 +expl/num paths total 331 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0447999 +expl/Actions Std 0.826073 +expl/Actions Max 2.53476 +expl/Actions Min -2.27523 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 287727 +eval/num paths total 299 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.262979 +eval/Actions Std 0.64875 +eval/Actions Max 0.999774 +eval/Actions Min -0.997857 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.73064e-06 +time/evaluation sampling (s) 3.39391 +time/exploration sampling (s) 3.38422 +time/logging (s) 0.0082142 +time/saving (s) 0.0127279 +time/training (s) 13.5764 +time/epoch (s) 20.3755 +time/total (s) 6297.91 +Epoch -702 +------------------------------ ---------------- +2022-05-15 19:47:43.798743 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -701 finished +------------------------------ ---------------- +epoch -701 +replay_buffer/size 999047 +trainer/num train calls 300000 +trainer/QF1 Loss 1.04154 +trainer/QF2 Loss 0.882814 +trainer/Policy Loss 17.5515 +trainer/Q1 Predictions Mean -71.3895 +trainer/Q1 Predictions Std 19.0856 +trainer/Q1 Predictions Max -0.503354 +trainer/Q1 Predictions Min -86.6767 +trainer/Q2 Predictions Mean -71.3087 +trainer/Q2 Predictions Std 19.0068 +trainer/Q2 Predictions Max -0.262974 +trainer/Q2 Predictions Min -86.3346 +trainer/Q Targets Mean -71.2645 +trainer/Q Targets Std 19.0303 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1506 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000630176 +trainer/policy/mean Std 0.706306 +trainer/policy/mean Max 0.998085 +trainer/policy/mean Min -0.996472 +trainer/policy/std Mean 0.425524 +trainer/policy/std Std 0.0203121 +trainer/policy/std Max 0.445866 +trainer/policy/std Min 0.392336 +trainer/Advantage Weights Mean 3.87438 +trainer/Advantage Weights Std 17.4286 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.56591e-15 +trainer/Advantage Score Mean -0.402833 +trainer/Advantage Score Std 0.614217 +trainer/Advantage Score Max 1.59476 +trainer/Advantage Score Min -3.40903 +trainer/V1 Predictions Mean -71.0018 +trainer/V1 Predictions Std 19.1172 +trainer/V1 Predictions Max 0.665409 +trainer/V1 Predictions Min -86.8152 +trainer/VF Loss 0.0861645 +expl/num steps total 300000 +expl/num paths total 332 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.188776 +expl/Actions Std 0.893912 +expl/Actions Max 2.31767 +expl/Actions Min -2.35983 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 288474 +eval/num paths total 300 +eval/path length Mean 747 +eval/path length Std 0 +eval/path length Max 747 +eval/path length Min 747 +eval/Rewards Mean 0.00133869 +eval/Rewards Std 0.0365636 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0441299 +eval/Actions Std 0.730133 +eval/Actions Max 0.999537 +eval/Actions Min -0.999633 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.58302e-06 +time/evaluation sampling (s) 2.64509 +time/exploration sampling (s) 3.41715 +time/logging (s) 0.0067983 +time/saving (s) 0.0122943 +time/training (s) 13.4588 +time/epoch (s) 19.5401 +time/total (s) 6317.45 +Epoch -701 +------------------------------ ---------------- +2022-05-15 19:48:03.746121 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -700 finished +------------------------------ ---------------- +epoch -700 +replay_buffer/size 999047 +trainer/num train calls 301000 +trainer/QF1 Loss 0.904753 +trainer/QF2 Loss 0.89403 +trainer/Policy Loss 12.3816 +trainer/Q1 Predictions Mean -71.4525 +trainer/Q1 Predictions Std 18.8163 +trainer/Q1 Predictions Max -0.257134 +trainer/Q1 Predictions Min -86.052 +trainer/Q2 Predictions Mean -71.4666 +trainer/Q2 Predictions Std 18.8231 +trainer/Q2 Predictions Max -0.231192 +trainer/Q2 Predictions Min -85.9892 +trainer/Q Targets Mean -71.4737 +trainer/Q Targets Std 18.7778 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.5975 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00600086 +trainer/policy/mean Std 0.715454 +trainer/policy/mean Max 0.99886 +trainer/policy/mean Min -0.999855 +trainer/policy/std Mean 0.426779 +trainer/policy/std Std 0.0204079 +trainer/policy/std Max 0.447971 +trainer/policy/std Min 0.394269 +trainer/Advantage Weights Mean 1.97738 +trainer/Advantage Weights Std 11.7055 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.82459e-11 +trainer/Advantage Score Mean -0.489354 +trainer/Advantage Score Std 0.471983 +trainer/Advantage Score Max 1.27889 +trainer/Advantage Score Min -2.37547 +trainer/V1 Predictions Mean -71.1176 +trainer/V1 Predictions Std 18.903 +trainer/V1 Predictions Max 0.515097 +trainer/V1 Predictions Min -85.5296 +trainer/VF Loss 0.0564002 +expl/num steps total 301000 +expl/num paths total 334 +expl/path length Mean 500 +expl/path length Std 148 +expl/path length Max 648 +expl/path length Min 352 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0406765 +expl/Actions Std 0.8397 +expl/Actions Max 2.59976 +expl/Actions Min -2.25042 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 289235 +eval/num paths total 301 +eval/path length Mean 761 +eval/path length Std 0 +eval/path length Max 761 +eval/path length Min 761 +eval/Rewards Mean 0.00131406 +eval/Rewards Std 0.0362261 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0291839 +eval/Actions Std 0.750145 +eval/Actions Max 0.999698 +eval/Actions Min -0.998633 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.80701e-06 +time/evaluation sampling (s) 3.0272 +time/exploration sampling (s) 3.68107 +time/logging (s) 0.0063948 +time/saving (s) 0.0177549 +time/training (s) 13.2066 +time/epoch (s) 19.939 +time/total (s) 6337.4 +Epoch -700 +------------------------------ ---------------- +2022-05-15 19:48:24.243199 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -699 finished +------------------------------ ---------------- +epoch -699 +replay_buffer/size 999047 +trainer/num train calls 302000 +trainer/QF1 Loss 1.33969 +trainer/QF2 Loss 1.39127 +trainer/Policy Loss 45.0788 +trainer/Q1 Predictions Mean -69.7344 +trainer/Q1 Predictions Std 20.5125 +trainer/Q1 Predictions Max -0.328278 +trainer/Q1 Predictions Min -86.1278 +trainer/Q2 Predictions Mean -69.7083 +trainer/Q2 Predictions Std 20.5316 +trainer/Q2 Predictions Max -0.190067 +trainer/Q2 Predictions Min -85.2713 +trainer/Q Targets Mean -70.5352 +trainer/Q Targets Std 20.6215 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8813 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0134515 +trainer/policy/mean Std 0.721034 +trainer/policy/mean Max 0.999273 +trainer/policy/mean Min -0.998329 +trainer/policy/std Mean 0.426825 +trainer/policy/std Std 0.0202588 +trainer/policy/std Max 0.447496 +trainer/policy/std Min 0.392763 +trainer/Advantage Weights Mean 9.86524 +trainer/Advantage Weights Std 25.187 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.41302e-11 +trainer/Advantage Score Mean -0.13885 +trainer/Advantage Score Std 0.477958 +trainer/Advantage Score Max 1.11227 +trainer/Advantage Score Min -2.41008 +trainer/V1 Predictions Mean -70.2142 +trainer/V1 Predictions Std 20.7204 +trainer/V1 Predictions Max -0.391476 +trainer/V1 Predictions Min -86.4751 +trainer/VF Loss 0.0574507 +expl/num steps total 302000 +expl/num paths total 335 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0773941 +expl/Actions Std 0.827048 +expl/Actions Max 2.45709 +expl/Actions Min -2.30614 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 289744 +eval/num paths total 302 +eval/path length Mean 509 +eval/path length Std 0 +eval/path length Max 509 +eval/path length Min 509 +eval/Rewards Mean 0.00196464 +eval/Rewards Std 0.0442807 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0275779 +eval/Actions Std 0.724805 +eval/Actions Max 0.999606 +eval/Actions Min -0.999683 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.76091e-06 +time/evaluation sampling (s) 3.06617 +time/exploration sampling (s) 3.81158 +time/logging (s) 0.00594051 +time/saving (s) 0.00988683 +time/training (s) 13.5961 +time/epoch (s) 20.4897 +time/total (s) 6357.89 +Epoch -699 +------------------------------ ---------------- +2022-05-15 19:48:44.220383 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -698 finished +------------------------------ ---------------- +epoch -698 +replay_buffer/size 999047 +trainer/num train calls 303000 +trainer/QF1 Loss 0.838606 +trainer/QF2 Loss 0.754827 +trainer/Policy Loss 39.2987 +trainer/Q1 Predictions Mean -70.0603 +trainer/Q1 Predictions Std 19.9934 +trainer/Q1 Predictions Max -0.511552 +trainer/Q1 Predictions Min -85.9008 +trainer/Q2 Predictions Mean -70.0524 +trainer/Q2 Predictions Std 19.9806 +trainer/Q2 Predictions Max -0.474152 +trainer/Q2 Predictions Min -85.8006 +trainer/Q Targets Mean -70.1206 +trainer/Q Targets Std 20.3182 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0557 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0166891 +trainer/policy/mean Std 0.702123 +trainer/policy/mean Max 0.995301 +trainer/policy/mean Min -0.99559 +trainer/policy/std Mean 0.426941 +trainer/policy/std Std 0.0215943 +trainer/policy/std Max 0.450155 +trainer/policy/std Min 0.392312 +trainer/Advantage Weights Mean 10.0365 +trainer/Advantage Weights Std 25.7342 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.49986e-14 +trainer/Advantage Score Mean -0.239291 +trainer/Advantage Score Std 0.63455 +trainer/Advantage Score Max 2.00799 +trainer/Advantage Score Min -3.09835 +trainer/V1 Predictions Mean -69.9045 +trainer/V1 Predictions Std 20.247 +trainer/V1 Predictions Max 0.472264 +trainer/V1 Predictions Min -85.9436 +trainer/VF Loss 0.0978961 +expl/num steps total 303000 +expl/num paths total 337 +expl/path length Mean 500 +expl/path length Std 310 +expl/path length Max 810 +expl/path length Min 190 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0345012 +expl/Actions Std 0.83413 +expl/Actions Max 2.21756 +expl/Actions Min -2.45928 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 290744 +eval/num paths total 303 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0374621 +eval/Actions Std 0.722979 +eval/Actions Max 0.99978 +eval/Actions Min -0.998614 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.58909e-06 +time/evaluation sampling (s) 3.14643 +time/exploration sampling (s) 3.54054 +time/logging (s) 0.006865 +time/saving (s) 0.00945593 +time/training (s) 13.2684 +time/epoch (s) 19.9717 +time/total (s) 6377.86 +Epoch -698 +------------------------------ ---------------- +2022-05-15 19:49:04.112774 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -697 finished +------------------------------ ---------------- +epoch -697 +replay_buffer/size 999047 +trainer/num train calls 304000 +trainer/QF1 Loss 4.60001 +trainer/QF2 Loss 4.75391 +trainer/Policy Loss 31.0406 +trainer/Q1 Predictions Mean -72.8512 +trainer/Q1 Predictions Std 16.0949 +trainer/Q1 Predictions Max -0.547053 +trainer/Q1 Predictions Min -87.3851 +trainer/Q2 Predictions Mean -72.8252 +trainer/Q2 Predictions Std 16.0149 +trainer/Q2 Predictions Max -0.457944 +trainer/Q2 Predictions Min -86.5802 +trainer/Q Targets Mean -72.8789 +trainer/Q Targets Std 16.2895 +trainer/Q Targets Max -0.944457 +trainer/Q Targets Min -86.6681 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0453379 +trainer/policy/mean Std 0.723149 +trainer/policy/mean Max 0.99919 +trainer/policy/mean Min -0.99926 +trainer/policy/std Mean 0.426816 +trainer/policy/std Std 0.0215022 +trainer/policy/std Max 0.448636 +trainer/policy/std Min 0.389581 +trainer/Advantage Weights Mean 7.58117 +trainer/Advantage Weights Std 24.3372 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.54206e-12 +trainer/Advantage Score Mean -0.278005 +trainer/Advantage Score Std 0.501845 +trainer/Advantage Score Max 1.0044 +trainer/Advantage Score Min -2.59187 +trainer/V1 Predictions Mean -72.7482 +trainer/V1 Predictions Std 16.2816 +trainer/V1 Predictions Max -0.216803 +trainer/V1 Predictions Min -86.6149 +trainer/VF Loss 0.055536 +expl/num steps total 304000 +expl/num paths total 338 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.260475 +expl/Actions Std 0.8391 +expl/Actions Max 2.45701 +expl/Actions Min -2.52854 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 291744 +eval/num paths total 304 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.154255 +eval/Actions Std 0.699228 +eval/Actions Max 0.999703 +eval/Actions Min -0.99952 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.03984e-06 +time/evaluation sampling (s) 3.04889 +time/exploration sampling (s) 3.21243 +time/logging (s) 0.00680755 +time/saving (s) 0.011962 +time/training (s) 13.6056 +time/epoch (s) 19.8857 +time/total (s) 6397.75 +Epoch -697 +------------------------------ ---------------- +2022-05-15 19:49:22.728136 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -696 finished +------------------------------ ---------------- +epoch -696 +replay_buffer/size 999047 +trainer/num train calls 305000 +trainer/QF1 Loss 1.07903 +trainer/QF2 Loss 1.04182 +trainer/Policy Loss 32.5181 +trainer/Q1 Predictions Mean -72.1995 +trainer/Q1 Predictions Std 17.6086 +trainer/Q1 Predictions Max -1.12866 +trainer/Q1 Predictions Min -86.5223 +trainer/Q2 Predictions Mean -72.2519 +trainer/Q2 Predictions Std 17.5835 +trainer/Q2 Predictions Max -1.46724 +trainer/Q2 Predictions Min -86.115 +trainer/Q Targets Mean -72.1271 +trainer/Q Targets Std 18.1004 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4655 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0177398 +trainer/policy/mean Std 0.711672 +trainer/policy/mean Max 0.997254 +trainer/policy/mean Min -0.998409 +trainer/policy/std Mean 0.425698 +trainer/policy/std Std 0.0218339 +trainer/policy/std Max 0.448634 +trainer/policy/std Min 0.389527 +trainer/Advantage Weights Mean 5.53063 +trainer/Advantage Weights Std 18.3337 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.01551e-15 +trainer/Advantage Score Mean -0.354384 +trainer/Advantage Score Std 0.650981 +trainer/Advantage Score Max 1.0216 +trainer/Advantage Score Min -3.31486 +trainer/V1 Predictions Mean -71.9541 +trainer/V1 Predictions Std 18.0516 +trainer/V1 Predictions Max 0.484429 +trainer/V1 Predictions Min -86.2721 +trainer/VF Loss 0.0747304 +expl/num steps total 305000 +expl/num paths total 340 +expl/path length Mean 500 +expl/path length Std 314 +expl/path length Max 814 +expl/path length Min 186 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0298443 +expl/Actions Std 0.843502 +expl/Actions Max 2.38677 +expl/Actions Min -2.27817 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 292501 +eval/num paths total 305 +eval/path length Mean 757 +eval/path length Std 0 +eval/path length Max 757 +eval/path length Min 757 +eval/Rewards Mean 0.001321 +eval/Rewards Std 0.0363216 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0640284 +eval/Actions Std 0.739225 +eval/Actions Max 0.999676 +eval/Actions Min -0.99951 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.55905e-06 +time/evaluation sampling (s) 2.72835 +time/exploration sampling (s) 3.27697 +time/logging (s) 0.00600528 +time/saving (s) 0.0087786 +time/training (s) 12.5879 +time/epoch (s) 18.608 +time/total (s) 6416.37 +Epoch -696 +------------------------------ ---------------- +2022-05-15 19:49:42.647521 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -695 finished +------------------------------ ---------------- +epoch -695 +replay_buffer/size 999047 +trainer/num train calls 306000 +trainer/QF1 Loss 1.11545 +trainer/QF2 Loss 1.18392 +trainer/Policy Loss 17.2755 +trainer/Q1 Predictions Mean -73.8626 +trainer/Q1 Predictions Std 15.8568 +trainer/Q1 Predictions Max -0.850031 +trainer/Q1 Predictions Min -86.1576 +trainer/Q2 Predictions Mean -73.8564 +trainer/Q2 Predictions Std 15.8284 +trainer/Q2 Predictions Max -0.629588 +trainer/Q2 Predictions Min -86.2374 +trainer/Q Targets Mean -73.9659 +trainer/Q Targets Std 15.9403 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6624 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00342085 +trainer/policy/mean Std 0.71387 +trainer/policy/mean Max 0.997849 +trainer/policy/mean Min -0.998258 +trainer/policy/std Mean 0.426056 +trainer/policy/std Std 0.0206206 +trainer/policy/std Max 0.44835 +trainer/policy/std Min 0.391903 +trainer/Advantage Weights Mean 3.39004 +trainer/Advantage Weights Std 13.6924 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.69433e-15 +trainer/Advantage Score Mean -0.271114 +trainer/Advantage Score Std 0.515719 +trainer/Advantage Score Max 0.718475 +trainer/Advantage Score Min -3.29924 +trainer/V1 Predictions Mean -73.7321 +trainer/V1 Predictions Std 15.9923 +trainer/V1 Predictions Max -0.687138 +trainer/V1 Predictions Min -86.528 +trainer/VF Loss 0.04292 +expl/num steps total 306000 +expl/num paths total 341 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00639289 +expl/Actions Std 0.833674 +expl/Actions Max 2.84367 +expl/Actions Min -2.67677 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 293501 +eval/num paths total 306 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.177575 +eval/Actions Std 0.639751 +eval/Actions Max 0.999115 +eval/Actions Min -0.998589 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.59214e-06 +time/evaluation sampling (s) 2.65019 +time/exploration sampling (s) 3.92007 +time/logging (s) 0.00713784 +time/saving (s) 0.0101509 +time/training (s) 13.3269 +time/epoch (s) 19.9144 +time/total (s) 6436.28 +Epoch -695 +------------------------------ ---------------- +2022-05-15 19:50:02.996971 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -694 finished +------------------------------ ---------------- +epoch -694 +replay_buffer/size 999047 +trainer/num train calls 307000 +trainer/QF1 Loss 1.19788 +trainer/QF2 Loss 1.07014 +trainer/Policy Loss 0.911763 +trainer/Q1 Predictions Mean -71.5934 +trainer/Q1 Predictions Std 18.7215 +trainer/Q1 Predictions Max -1.31198 +trainer/Q1 Predictions Min -85.7163 +trainer/Q2 Predictions Mean -71.5723 +trainer/Q2 Predictions Std 18.7811 +trainer/Q2 Predictions Max -1.07277 +trainer/Q2 Predictions Min -85.9158 +trainer/Q Targets Mean -71.0668 +trainer/Q Targets Std 18.9775 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7968 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0279416 +trainer/policy/mean Std 0.701488 +trainer/policy/mean Max 0.997123 +trainer/policy/mean Min -0.998154 +trainer/policy/std Mean 0.426361 +trainer/policy/std Std 0.0216 +trainer/policy/std Max 0.449585 +trainer/policy/std Min 0.387674 +trainer/Advantage Weights Mean 0.187074 +trainer/Advantage Weights Std 0.886777 +trainer/Advantage Weights Max 9.31545 +trainer/Advantage Weights Min 2.39142e-18 +trainer/Advantage Score Mean -0.694736 +trainer/Advantage Score Std 0.584913 +trainer/Advantage Score Max 0.223167 +trainer/Advantage Score Min -4.05746 +trainer/V1 Predictions Mean -70.7462 +trainer/V1 Predictions Std 19.129 +trainer/V1 Predictions Max -0.092652 +trainer/V1 Predictions Min -85.5723 +trainer/VF Loss 0.0829172 +expl/num steps total 307000 +expl/num paths total 342 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0476012 +expl/Actions Std 0.832721 +expl/Actions Max 2.44229 +expl/Actions Min -2.36362 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 294501 +eval/num paths total 307 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.27639 +eval/Actions Std 0.698386 +eval/Actions Max 0.998827 +eval/Actions Min -0.999267 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.98303e-06 +time/evaluation sampling (s) 3.11114 +time/exploration sampling (s) 3.98351 +time/logging (s) 0.00676749 +time/saving (s) 0.00926391 +time/training (s) 13.2318 +time/epoch (s) 20.3425 +time/total (s) 6456.63 +Epoch -694 +------------------------------ ---------------- +2022-05-15 19:50:23.229474 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -693 finished +------------------------------ ---------------- +epoch -693 +replay_buffer/size 999047 +trainer/num train calls 308000 +trainer/QF1 Loss 0.587212 +trainer/QF2 Loss 0.540857 +trainer/Policy Loss 14.691 +trainer/Q1 Predictions Mean -72.0466 +trainer/Q1 Predictions Std 18.9234 +trainer/Q1 Predictions Max -0.99373 +trainer/Q1 Predictions Min -86.2837 +trainer/Q2 Predictions Mean -72.0698 +trainer/Q2 Predictions Std 18.9106 +trainer/Q2 Predictions Max -0.511274 +trainer/Q2 Predictions Min -86.5651 +trainer/Q Targets Mean -71.8641 +trainer/Q Targets Std 18.709 +trainer/Q Targets Max -1.44926 +trainer/Q Targets Min -86.1058 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0196294 +trainer/policy/mean Std 0.715666 +trainer/policy/mean Max 0.996584 +trainer/policy/mean Min -0.99664 +trainer/policy/std Mean 0.425812 +trainer/policy/std Std 0.0204335 +trainer/policy/std Max 0.448461 +trainer/policy/std Min 0.391486 +trainer/Advantage Weights Mean 3.28225 +trainer/Advantage Weights Std 15.5529 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.69381e-15 +trainer/Advantage Score Mean -0.409486 +trainer/Advantage Score Std 0.592218 +trainer/Advantage Score Max 2.03936 +trainer/Advantage Score Min -3.24984 +trainer/V1 Predictions Mean -71.6132 +trainer/V1 Predictions Std 18.8795 +trainer/V1 Predictions Max -0.415188 +trainer/V1 Predictions Min -85.978 +trainer/VF Loss 0.089218 +expl/num steps total 308000 +expl/num paths total 343 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0431974 +expl/Actions Std 0.870316 +expl/Actions Max 2.32097 +expl/Actions Min -2.41554 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 295501 +eval/num paths total 308 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.251395 +eval/Actions Std 0.761106 +eval/Actions Max 0.998178 +eval/Actions Min -0.998658 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.67662e-06 +time/evaluation sampling (s) 3.17326 +time/exploration sampling (s) 3.8992 +time/logging (s) 0.00894296 +time/saving (s) 0.0149216 +time/training (s) 13.1321 +time/epoch (s) 20.2284 +time/total (s) 6476.86 +Epoch -693 +------------------------------ ---------------- +2022-05-15 19:50:43.149116 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -692 finished +------------------------------ ---------------- +epoch -692 +replay_buffer/size 999047 +trainer/num train calls 309000 +trainer/QF1 Loss 0.960111 +trainer/QF2 Loss 1.10435 +trainer/Policy Loss 19.5165 +trainer/Q1 Predictions Mean -72.7305 +trainer/Q1 Predictions Std 17.4918 +trainer/Q1 Predictions Max -0.507459 +trainer/Q1 Predictions Min -85.6425 +trainer/Q2 Predictions Mean -72.7255 +trainer/Q2 Predictions Std 17.4115 +trainer/Q2 Predictions Max -0.467843 +trainer/Q2 Predictions Min -85.4565 +trainer/Q Targets Mean -72.6421 +trainer/Q Targets Std 17.8158 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.4687 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00147311 +trainer/policy/mean Std 0.71749 +trainer/policy/mean Max 0.997939 +trainer/policy/mean Min -0.996726 +trainer/policy/std Mean 0.426828 +trainer/policy/std Std 0.0207271 +trainer/policy/std Max 0.447747 +trainer/policy/std Min 0.392446 +trainer/Advantage Weights Mean 4.27104 +trainer/Advantage Weights Std 16.2358 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.07762e-14 +trainer/Advantage Score Mean -0.302183 +trainer/Advantage Score Std 0.51823 +trainer/Advantage Score Max 1.10925 +trainer/Advantage Score Min -3.08307 +trainer/V1 Predictions Mean -72.4162 +trainer/V1 Predictions Std 17.8649 +trainer/V1 Predictions Max 0.520357 +trainer/V1 Predictions Min -85.332 +trainer/VF Loss 0.0514057 +expl/num steps total 309000 +expl/num paths total 344 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0200954 +expl/Actions Std 0.819984 +expl/Actions Max 2.47568 +expl/Actions Min -2.33373 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 296501 +eval/num paths total 309 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.147632 +eval/Actions Std 0.719852 +eval/Actions Max 0.999649 +eval/Actions Min -0.998562 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.33134e-06 +time/evaluation sampling (s) 3.24745 +time/exploration sampling (s) 3.50737 +time/logging (s) 0.00739692 +time/saving (s) 0.0101233 +time/training (s) 13.136 +time/epoch (s) 19.9083 +time/total (s) 6496.77 +Epoch -692 +------------------------------ ---------------- +2022-05-15 19:51:02.585347 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -691 finished +------------------------------ ---------------- +epoch -691 +replay_buffer/size 999047 +trainer/num train calls 310000 +trainer/QF1 Loss 0.626077 +trainer/QF2 Loss 0.591877 +trainer/Policy Loss 14.1526 +trainer/Q1 Predictions Mean -74.1987 +trainer/Q1 Predictions Std 14.7371 +trainer/Q1 Predictions Max -1.67131 +trainer/Q1 Predictions Min -86.792 +trainer/Q2 Predictions Mean -74.1832 +trainer/Q2 Predictions Std 14.719 +trainer/Q2 Predictions Max -0.847945 +trainer/Q2 Predictions Min -86.3456 +trainer/Q Targets Mean -73.9416 +trainer/Q Targets Std 14.5757 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9942 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00542196 +trainer/policy/mean Std 0.6993 +trainer/policy/mean Max 0.999738 +trainer/policy/mean Min -0.996907 +trainer/policy/std Mean 0.428501 +trainer/policy/std Std 0.02099 +trainer/policy/std Max 0.450188 +trainer/policy/std Min 0.393666 +trainer/Advantage Weights Mean 2.2384 +trainer/Advantage Weights Std 13.8821 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.79968e-11 +trainer/Advantage Score Mean -0.558472 +trainer/Advantage Score Std 0.464259 +trainer/Advantage Score Max 1.17175 +trainer/Advantage Score Min -2.34116 +trainer/V1 Predictions Mean -73.7035 +trainer/V1 Predictions Std 14.5623 +trainer/V1 Predictions Max -2.00878 +trainer/V1 Predictions Min -85.9486 +trainer/VF Loss 0.06376 +expl/num steps total 310000 +expl/num paths total 345 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.131791 +expl/Actions Std 0.862425 +expl/Actions Max 2.30325 +expl/Actions Min -2.32647 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 297195 +eval/num paths total 310 +eval/path length Mean 694 +eval/path length Std 0 +eval/path length Max 694 +eval/path length Min 694 +eval/Rewards Mean 0.00144092 +eval/Rewards Std 0.0379321 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0486391 +eval/Actions Std 0.732993 +eval/Actions Max 0.999763 +eval/Actions Min -0.99854 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.55182e-06 +time/evaluation sampling (s) 2.87831 +time/exploration sampling (s) 3.29039 +time/logging (s) 0.00578579 +time/saving (s) 0.0090685 +time/training (s) 13.2441 +time/epoch (s) 19.4276 +time/total (s) 6516.21 +Epoch -691 +------------------------------ ---------------- +2022-05-15 19:51:22.202329 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -690 finished +------------------------------ ---------------- +epoch -690 +replay_buffer/size 999047 +trainer/num train calls 311000 +trainer/QF1 Loss 0.635067 +trainer/QF2 Loss 0.590021 +trainer/Policy Loss 13.0082 +trainer/Q1 Predictions Mean -72.8264 +trainer/Q1 Predictions Std 17.056 +trainer/Q1 Predictions Max -0.528348 +trainer/Q1 Predictions Min -86.2469 +trainer/Q2 Predictions Mean -72.8492 +trainer/Q2 Predictions Std 17.0133 +trainer/Q2 Predictions Max -0.752918 +trainer/Q2 Predictions Min -86.0493 +trainer/Q Targets Mean -73.0834 +trainer/Q Targets Std 17.2501 +trainer/Q Targets Max 0.527338 +trainer/Q Targets Min -86.1471 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0271936 +trainer/policy/mean Std 0.712773 +trainer/policy/mean Max 0.998892 +trainer/policy/mean Min -0.995878 +trainer/policy/std Mean 0.429836 +trainer/policy/std Std 0.0204651 +trainer/policy/std Max 0.451029 +trainer/policy/std Min 0.394437 +trainer/Advantage Weights Mean 3.72796 +trainer/Advantage Weights Std 15.6889 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.39986e-11 +trainer/Advantage Score Mean -0.37857 +trainer/Advantage Score Std 0.530324 +trainer/Advantage Score Max 0.59787 +trainer/Advantage Score Min -2.49921 +trainer/V1 Predictions Mean -72.9063 +trainer/V1 Predictions Std 17.1611 +trainer/V1 Predictions Max 1.36814 +trainer/V1 Predictions Min -85.9656 +trainer/VF Loss 0.0514823 +expl/num steps total 311000 +expl/num paths total 347 +expl/path length Mean 500 +expl/path length Std 453 +expl/path length Max 953 +expl/path length Min 47 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0654467 +expl/Actions Std 0.829183 +expl/Actions Max 2.37154 +expl/Actions Min -2.61136 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 298195 +eval/num paths total 311 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0224529 +eval/Actions Std 0.733064 +eval/Actions Max 0.99962 +eval/Actions Min -0.999581 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.57209e-06 +time/evaluation sampling (s) 2.62665 +time/exploration sampling (s) 3.913 +time/logging (s) 0.00708739 +time/saving (s) 0.00924934 +time/training (s) 13.0561 +time/epoch (s) 19.6121 +time/total (s) 6535.82 +Epoch -690 +------------------------------ ---------------- +2022-05-15 19:51:42.423924 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -689 finished +------------------------------ ---------------- +epoch -689 +replay_buffer/size 999047 +trainer/num train calls 312000 +trainer/QF1 Loss 1.29039 +trainer/QF2 Loss 1.46598 +trainer/Policy Loss 11.6179 +trainer/Q1 Predictions Mean -71.4327 +trainer/Q1 Predictions Std 19.5633 +trainer/Q1 Predictions Max -0.990416 +trainer/Q1 Predictions Min -86.4648 +trainer/Q2 Predictions Mean -71.4672 +trainer/Q2 Predictions Std 19.4549 +trainer/Q2 Predictions Max -0.716835 +trainer/Q2 Predictions Min -86.472 +trainer/Q Targets Mean -70.8845 +trainer/Q Targets Std 20.1186 +trainer/Q Targets Max 1.69989 +trainer/Q Targets Min -86.2906 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0175244 +trainer/policy/mean Std 0.718988 +trainer/policy/mean Max 0.998775 +trainer/policy/mean Min -0.999173 +trainer/policy/std Mean 0.42789 +trainer/policy/std Std 0.0204291 +trainer/policy/std Max 0.449594 +trainer/policy/std Min 0.394904 +trainer/Advantage Weights Mean 2.91299 +trainer/Advantage Weights Std 13.2016 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.19195e-16 +trainer/Advantage Score Mean -0.388898 +trainer/Advantage Score Std 0.64433 +trainer/Advantage Score Max 0.815247 +trainer/Advantage Score Min -3.51943 +trainer/V1 Predictions Mean -70.6727 +trainer/V1 Predictions Std 20.094 +trainer/V1 Predictions Max 2.2079 +trainer/V1 Predictions Min -86.1614 +trainer/VF Loss 0.0665597 +expl/num steps total 312000 +expl/num paths total 349 +expl/path length Mean 500 +expl/path length Std 294 +expl/path length Max 794 +expl/path length Min 206 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0447262 +expl/Actions Std 0.828929 +expl/Actions Max 2.39131 +expl/Actions Min -2.53925 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 299195 +eval/num paths total 312 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0236112 +eval/Actions Std 0.674813 +eval/Actions Max 0.999342 +eval/Actions Min -0.996716 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.15299e-06 +time/evaluation sampling (s) 2.97169 +time/exploration sampling (s) 3.94123 +time/logging (s) 0.00683195 +time/saving (s) 0.0135307 +time/training (s) 13.2814 +time/epoch (s) 20.2147 +time/total (s) 6556.04 +Epoch -689 +------------------------------ ---------------- +2022-05-15 19:52:02.707486 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -688 finished +------------------------------ ---------------- +epoch -688 +replay_buffer/size 999047 +trainer/num train calls 313000 +trainer/QF1 Loss 0.820337 +trainer/QF2 Loss 0.828405 +trainer/Policy Loss 8.2046 +trainer/Q1 Predictions Mean -71.4155 +trainer/Q1 Predictions Std 17.1135 +trainer/Q1 Predictions Max -4.1504 +trainer/Q1 Predictions Min -85.8168 +trainer/Q2 Predictions Mean -71.4133 +trainer/Q2 Predictions Std 17.1567 +trainer/Q2 Predictions Max -3.50276 +trainer/Q2 Predictions Min -85.7998 +trainer/Q Targets Mean -71.6057 +trainer/Q Targets Std 17.391 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9918 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0105069 +trainer/policy/mean Std 0.724751 +trainer/policy/mean Max 0.999687 +trainer/policy/mean Min -0.998151 +trainer/policy/std Mean 0.427275 +trainer/policy/std Std 0.0210935 +trainer/policy/std Max 0.451211 +trainer/policy/std Min 0.393426 +trainer/Advantage Weights Mean 1.72867 +trainer/Advantage Weights Std 9.13145 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.13507e-16 +trainer/Advantage Score Mean -0.455611 +trainer/Advantage Score Std 0.655128 +trainer/Advantage Score Max 0.861908 +trainer/Advantage Score Min -3.52053 +trainer/V1 Predictions Mean -71.3546 +trainer/V1 Predictions Std 17.3297 +trainer/V1 Predictions Max -3.74792 +trainer/V1 Predictions Min -85.5577 +trainer/VF Loss 0.0699292 +expl/num steps total 313000 +expl/num paths total 351 +expl/path length Mean 500 +expl/path length Std 337 +expl/path length Max 837 +expl/path length Min 163 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0214818 +expl/Actions Std 0.837557 +expl/Actions Max 2.24113 +expl/Actions Min -2.34026 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 300195 +eval/num paths total 313 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0504471 +eval/Actions Std 0.724762 +eval/Actions Max 0.999415 +eval/Actions Min -0.998679 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.85592e-06 +time/evaluation sampling (s) 3.06451 +time/exploration sampling (s) 3.97477 +time/logging (s) 0.0084473 +time/saving (s) 0.0134447 +time/training (s) 13.2171 +time/epoch (s) 20.2783 +time/total (s) 6576.32 +Epoch -688 +------------------------------ ---------------- +2022-05-15 19:52:22.532768 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -687 finished +------------------------------ ---------------- +epoch -687 +replay_buffer/size 999047 +trainer/num train calls 314000 +trainer/QF1 Loss 0.693661 +trainer/QF2 Loss 0.737029 +trainer/Policy Loss 17.1172 +trainer/Q1 Predictions Mean -70.2124 +trainer/Q1 Predictions Std 19.8704 +trainer/Q1 Predictions Max -0.593454 +trainer/Q1 Predictions Min -86.2061 +trainer/Q2 Predictions Mean -70.1385 +trainer/Q2 Predictions Std 19.9746 +trainer/Q2 Predictions Max -0.405183 +trainer/Q2 Predictions Min -86.1357 +trainer/Q Targets Mean -70.1038 +trainer/Q Targets Std 19.7789 +trainer/Q Targets Max 0.707785 +trainer/Q Targets Min -85.8244 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0160536 +trainer/policy/mean Std 0.727384 +trainer/policy/mean Max 0.999274 +trainer/policy/mean Min -0.998276 +trainer/policy/std Mean 0.425061 +trainer/policy/std Std 0.0215831 +trainer/policy/std Max 0.449912 +trainer/policy/std Min 0.387545 +trainer/Advantage Weights Mean 3.33958 +trainer/Advantage Weights Std 16.4164 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.04856e-15 +trainer/Advantage Score Mean -0.586869 +trainer/Advantage Score Std 0.630109 +trainer/Advantage Score Max 1.55287 +trainer/Advantage Score Min -3.34241 +trainer/V1 Predictions Mean -69.8236 +trainer/V1 Predictions Std 19.9163 +trainer/V1 Predictions Max 0.829045 +trainer/V1 Predictions Min -85.7398 +trainer/VF Loss 0.100621 +expl/num steps total 314000 +expl/num paths total 353 +expl/path length Mean 500 +expl/path length Std 378 +expl/path length Max 878 +expl/path length Min 122 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0430022 +expl/Actions Std 0.809297 +expl/Actions Max 2.29235 +expl/Actions Min -2.43961 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 301195 +eval/num paths total 314 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0945953 +eval/Actions Std 0.642936 +eval/Actions Max 0.999135 +eval/Actions Min -0.998274 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.21493e-06 +time/evaluation sampling (s) 3.21766 +time/exploration sampling (s) 3.59062 +time/logging (s) 0.00679743 +time/saving (s) 0.0091994 +time/training (s) 12.9903 +time/epoch (s) 19.8146 +time/total (s) 6596.14 +Epoch -687 +------------------------------ ---------------- +2022-05-15 19:52:42.215507 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -686 finished +------------------------------ ---------------- +epoch -686 +replay_buffer/size 999047 +trainer/num train calls 315000 +trainer/QF1 Loss 0.700313 +trainer/QF2 Loss 0.72759 +trainer/Policy Loss 21.8496 +trainer/Q1 Predictions Mean -70.0339 +trainer/Q1 Predictions Std 19.38 +trainer/Q1 Predictions Max -0.483231 +trainer/Q1 Predictions Min -87.0953 +trainer/Q2 Predictions Mean -69.9785 +trainer/Q2 Predictions Std 19.4071 +trainer/Q2 Predictions Max -0.467943 +trainer/Q2 Predictions Min -87.0054 +trainer/Q Targets Mean -70.0088 +trainer/Q Targets Std 19.5332 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9661 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0161209 +trainer/policy/mean Std 0.718287 +trainer/policy/mean Max 0.998828 +trainer/policy/mean Min -0.998187 +trainer/policy/std Mean 0.426525 +trainer/policy/std Std 0.021939 +trainer/policy/std Max 0.452218 +trainer/policy/std Min 0.388127 +trainer/Advantage Weights Mean 6.04237 +trainer/Advantage Weights Std 20.317 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.00408e-12 +trainer/Advantage Score Mean -0.233375 +trainer/Advantage Score Std 0.455355 +trainer/Advantage Score Max 1.07641 +trainer/Advantage Score Min -2.56845 +trainer/V1 Predictions Mean -69.7323 +trainer/V1 Predictions Std 19.6181 +trainer/V1 Predictions Max 1.18571 +trainer/V1 Predictions Min -86.8439 +trainer/VF Loss 0.0509784 +expl/num steps total 315000 +expl/num paths total 354 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0275925 +expl/Actions Std 0.84335 +expl/Actions Max 2.56613 +expl/Actions Min -2.34858 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 302195 +eval/num paths total 315 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0033171 +eval/Actions Std 0.705341 +eval/Actions Max 0.999125 +eval/Actions Min -0.998248 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.46918e-06 +time/evaluation sampling (s) 2.93728 +time/exploration sampling (s) 3.21002 +time/logging (s) 0.00693124 +time/saving (s) 0.00950691 +time/training (s) 13.5126 +time/epoch (s) 19.6764 +time/total (s) 6615.82 +Epoch -686 +------------------------------ ---------------- +2022-05-15 19:53:02.364641 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -685 finished +------------------------------ ---------------- +epoch -685 +replay_buffer/size 999047 +trainer/num train calls 316000 +trainer/QF1 Loss 0.637211 +trainer/QF2 Loss 0.663682 +trainer/Policy Loss 15.0333 +trainer/Q1 Predictions Mean -72.1608 +trainer/Q1 Predictions Std 18.3702 +trainer/Q1 Predictions Max -0.575485 +trainer/Q1 Predictions Min -85.8759 +trainer/Q2 Predictions Mean -72.211 +trainer/Q2 Predictions Std 18.3404 +trainer/Q2 Predictions Max -0.553249 +trainer/Q2 Predictions Min -85.6767 +trainer/Q Targets Mean -72.3713 +trainer/Q Targets Std 18.4206 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.1544 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0148886 +trainer/policy/mean Std 0.720762 +trainer/policy/mean Max 0.99908 +trainer/policy/mean Min -0.99835 +trainer/policy/std Mean 0.42587 +trainer/policy/std Std 0.0217687 +trainer/policy/std Max 0.448103 +trainer/policy/std Min 0.387914 +trainer/Advantage Weights Mean 1.72506 +trainer/Advantage Weights Std 9.91186 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.63233e-12 +trainer/Advantage Score Mean -0.425376 +trainer/Advantage Score Std 0.49171 +trainer/Advantage Score Max 1.81914 +trainer/Advantage Score Min -2.66631 +trainer/V1 Predictions Mean -72.1472 +trainer/V1 Predictions Std 18.4926 +trainer/V1 Predictions Max -0.839764 +trainer/V1 Predictions Min -86.0179 +trainer/VF Loss 0.0564137 +expl/num steps total 316000 +expl/num paths total 355 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0165805 +expl/Actions Std 0.827772 +expl/Actions Max 2.3172 +expl/Actions Min -2.53628 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 302965 +eval/num paths total 316 +eval/path length Mean 770 +eval/path length Std 0 +eval/path length Max 770 +eval/path length Min 770 +eval/Rewards Mean 0.0012987 +eval/Rewards Std 0.0360141 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0138757 +eval/Actions Std 0.728548 +eval/Actions Max 0.999062 +eval/Actions Min -0.998742 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.31504e-06 +time/evaluation sampling (s) 2.55612 +time/exploration sampling (s) 3.76419 +time/logging (s) 0.00693917 +time/saving (s) 0.0104171 +time/training (s) 13.8048 +time/epoch (s) 20.1425 +time/total (s) 6635.97 +Epoch -685 +------------------------------ ---------------- +2022-05-15 19:53:22.940688 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -684 finished +------------------------------ ---------------- +epoch -684 +replay_buffer/size 999047 +trainer/num train calls 317000 +trainer/QF1 Loss 1.57272 +trainer/QF2 Loss 1.49721 +trainer/Policy Loss 63.4173 +trainer/Q1 Predictions Mean -70.9321 +trainer/Q1 Predictions Std 18.3285 +trainer/Q1 Predictions Max -0.42281 +trainer/Q1 Predictions Min -85.8185 +trainer/Q2 Predictions Mean -70.951 +trainer/Q2 Predictions Std 18.3377 +trainer/Q2 Predictions Max -0.533965 +trainer/Q2 Predictions Min -85.6141 +trainer/Q Targets Mean -71.8452 +trainer/Q Targets Std 18.5026 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3683 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00906649 +trainer/policy/mean Std 0.724706 +trainer/policy/mean Max 0.998818 +trainer/policy/mean Min -0.99902 +trainer/policy/std Mean 0.426407 +trainer/policy/std Std 0.0206083 +trainer/policy/std Max 0.448 +trainer/policy/std Min 0.391287 +trainer/Advantage Weights Mean 13.238 +trainer/Advantage Weights Std 29.5389 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.25517e-15 +trainer/Advantage Score Mean -0.124993 +trainer/Advantage Score Std 0.580087 +trainer/Advantage Score Max 2.85281 +trainer/Advantage Score Min -3.27054 +trainer/V1 Predictions Mean -71.5672 +trainer/V1 Predictions Std 18.5364 +trainer/V1 Predictions Max 0.896618 +trainer/V1 Predictions Min -86.2993 +trainer/VF Loss 0.116623 +expl/num steps total 317000 +expl/num paths total 357 +expl/path length Mean 500 +expl/path length Std 327 +expl/path length Max 827 +expl/path length Min 173 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0508416 +expl/Actions Std 0.853196 +expl/Actions Max 2.68146 +expl/Actions Min -2.41317 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 303965 +eval/num paths total 317 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.217688 +eval/Actions Std 0.707931 +eval/Actions Max 0.99946 +eval/Actions Min -0.997312 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.9509e-06 +time/evaluation sampling (s) 3.25895 +time/exploration sampling (s) 3.87501 +time/logging (s) 0.00754946 +time/saving (s) 0.0116555 +time/training (s) 13.4153 +time/epoch (s) 20.5685 +time/total (s) 6656.54 +Epoch -684 +------------------------------ ---------------- +2022-05-15 19:53:43.225442 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -683 finished +------------------------------ ---------------- +epoch -683 +replay_buffer/size 999047 +trainer/num train calls 318000 +trainer/QF1 Loss 1.31412 +trainer/QF2 Loss 1.32576 +trainer/Policy Loss 49.9108 +trainer/Q1 Predictions Mean -71.9058 +trainer/Q1 Predictions Std 17.3844 +trainer/Q1 Predictions Max -0.252775 +trainer/Q1 Predictions Min -86.3209 +trainer/Q2 Predictions Mean -71.935 +trainer/Q2 Predictions Std 17.3872 +trainer/Q2 Predictions Max -0.358578 +trainer/Q2 Predictions Min -86.4691 +trainer/Q Targets Mean -72.6371 +trainer/Q Targets Std 17.1212 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0888 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0105001 +trainer/policy/mean Std 0.720645 +trainer/policy/mean Max 0.999772 +trainer/policy/mean Min -0.999149 +trainer/policy/std Mean 0.424324 +trainer/policy/std Std 0.0209992 +trainer/policy/std Max 0.446314 +trainer/policy/std Min 0.388098 +trainer/Advantage Weights Mean 10.6343 +trainer/Advantage Weights Std 26.2179 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.2085e-16 +trainer/Advantage Score Mean -0.168088 +trainer/Advantage Score Std 0.543357 +trainer/Advantage Score Max 1.33247 +trainer/Advantage Score Min -3.60491 +trainer/V1 Predictions Mean -72.3953 +trainer/V1 Predictions Std 17.2501 +trainer/V1 Predictions Max 0.154623 +trainer/V1 Predictions Min -86.8749 +trainer/VF Loss 0.0730177 +expl/num steps total 318000 +expl/num paths total 358 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00726647 +expl/Actions Std 0.860931 +expl/Actions Max 2.36398 +expl/Actions Min -2.71323 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 304345 +eval/num paths total 318 +eval/path length Mean 380 +eval/path length Std 0 +eval/path length Max 380 +eval/path length Min 380 +eval/Rewards Mean 0.00263158 +eval/Rewards Std 0.0512314 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0346676 +eval/Actions Std 0.743355 +eval/Actions Max 0.999572 +eval/Actions Min -0.999759 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.65567e-06 +time/evaluation sampling (s) 3.06906 +time/exploration sampling (s) 3.53304 +time/logging (s) 0.00678714 +time/saving (s) 0.0119915 +time/training (s) 13.6546 +time/epoch (s) 20.2755 +time/total (s) 6676.82 +Epoch -683 +------------------------------ ---------------- +2022-05-15 19:54:02.655551 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -682 finished +------------------------------ ---------------- +epoch -682 +replay_buffer/size 999047 +trainer/num train calls 319000 +trainer/QF1 Loss 22.6893 +trainer/QF2 Loss 22.772 +trainer/Policy Loss 21.8105 +trainer/Q1 Predictions Mean -70.0807 +trainer/Q1 Predictions Std 20.321 +trainer/Q1 Predictions Max -0.842137 +trainer/Q1 Predictions Min -86.4993 +trainer/Q2 Predictions Mean -70.1812 +trainer/Q2 Predictions Std 20.3526 +trainer/Q2 Predictions Max 0.0305291 +trainer/Q2 Predictions Min -86.4534 +trainer/Q Targets Mean -70.667 +trainer/Q Targets Std 19.8233 +trainer/Q Targets Max -0.571273 +trainer/Q Targets Min -86.3961 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0256579 +trainer/policy/mean Std 0.703306 +trainer/policy/mean Max 0.999086 +trainer/policy/mean Min -0.999122 +trainer/policy/std Mean 0.425515 +trainer/policy/std Std 0.0217012 +trainer/policy/std Max 0.448262 +trainer/policy/std Min 0.390425 +trainer/Advantage Weights Mean 4.83014 +trainer/Advantage Weights Std 17.8616 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.15239e-14 +trainer/Advantage Score Mean -0.357378 +trainer/Advantage Score Std 0.564249 +trainer/Advantage Score Max 1.37971 +trainer/Advantage Score Min -3.14696 +trainer/V1 Predictions Mean -70.1602 +trainer/V1 Predictions Std 20.4014 +trainer/V1 Predictions Max 0.0989397 +trainer/V1 Predictions Min -86.248 +trainer/VF Loss 0.0685168 +expl/num steps total 319000 +expl/num paths total 359 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00354846 +expl/Actions Std 0.82321 +expl/Actions Max 2.36993 +expl/Actions Min -2.36873 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 305247 +eval/num paths total 319 +eval/path length Mean 902 +eval/path length Std 0 +eval/path length Max 902 +eval/path length Min 902 +eval/Rewards Mean 0.00110865 +eval/Rewards Std 0.0332779 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0304222 +eval/Actions Std 0.707277 +eval/Actions Max 0.999312 +eval/Actions Min -0.999111 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.82796e-06 +time/evaluation sampling (s) 2.72125 +time/exploration sampling (s) 3.40036 +time/logging (s) 0.0104542 +time/saving (s) 0.0111177 +time/training (s) 13.2822 +time/epoch (s) 19.4254 +time/total (s) 6696.25 +Epoch -682 +------------------------------ ---------------- +2022-05-15 19:54:22.246818 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -681 finished +------------------------------ ---------------- +epoch -681 +replay_buffer/size 999047 +trainer/num train calls 320000 +trainer/QF1 Loss 1.63718 +trainer/QF2 Loss 1.71551 +trainer/Policy Loss 19.3385 +trainer/Q1 Predictions Mean -71.774 +trainer/Q1 Predictions Std 17.598 +trainer/Q1 Predictions Max -0.964742 +trainer/Q1 Predictions Min -85.7921 +trainer/Q2 Predictions Mean -71.7922 +trainer/Q2 Predictions Std 17.592 +trainer/Q2 Predictions Max -0.991184 +trainer/Q2 Predictions Min -85.8517 +trainer/Q Targets Mean -71.5596 +trainer/Q Targets Std 17.3284 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.772 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00603243 +trainer/policy/mean Std 0.713075 +trainer/policy/mean Max 0.998175 +trainer/policy/mean Min -0.99607 +trainer/policy/std Mean 0.42734 +trainer/policy/std Std 0.0210652 +trainer/policy/std Max 0.449798 +trainer/policy/std Min 0.393027 +trainer/Advantage Weights Mean 3.99659 +trainer/Advantage Weights Std 17.8895 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.11987e-11 +trainer/Advantage Score Mean -0.46624 +trainer/Advantage Score Std 0.572707 +trainer/Advantage Score Max 2.0177 +trainer/Advantage Score Min -2.52152 +trainer/V1 Predictions Mean -71.3482 +trainer/V1 Predictions Std 17.4355 +trainer/V1 Predictions Max 0.204786 +trainer/V1 Predictions Min -85.5316 +trainer/VF Loss 0.101961 +expl/num steps total 320000 +expl/num paths total 361 +expl/path length Mean 500 +expl/path length Std 219 +expl/path length Max 719 +expl/path length Min 281 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0394777 +expl/Actions Std 0.822704 +expl/Actions Max 2.36597 +expl/Actions Min -2.27165 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 305984 +eval/num paths total 320 +eval/path length Mean 737 +eval/path length Std 0 +eval/path length Max 737 +eval/path length Min 737 +eval/Rewards Mean 0.00135685 +eval/Rewards Std 0.0368105 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0327533 +eval/Actions Std 0.737285 +eval/Actions Max 0.999871 +eval/Actions Min -0.998979 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.51597e-06 +time/evaluation sampling (s) 2.61085 +time/exploration sampling (s) 3.94723 +time/logging (s) 0.00627845 +time/saving (s) 0.0095291 +time/training (s) 13.0013 +time/epoch (s) 19.5752 +time/total (s) 6715.83 +Epoch -681 +------------------------------ ---------------- +2022-05-15 19:54:42.348677 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -680 finished +------------------------------ ---------------- +epoch -680 +replay_buffer/size 999047 +trainer/num train calls 321000 +trainer/QF1 Loss 0.693542 +trainer/QF2 Loss 0.733029 +trainer/Policy Loss 5.02489 +trainer/Q1 Predictions Mean -73.4171 +trainer/Q1 Predictions Std 16.954 +trainer/Q1 Predictions Max -0.358037 +trainer/Q1 Predictions Min -87.0234 +trainer/Q2 Predictions Mean -73.3887 +trainer/Q2 Predictions Std 16.9398 +trainer/Q2 Predictions Max -0.191786 +trainer/Q2 Predictions Min -86.6972 +trainer/Q Targets Mean -73.0018 +trainer/Q Targets Std 16.8529 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2518 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0122769 +trainer/policy/mean Std 0.71557 +trainer/policy/mean Max 0.998788 +trainer/policy/mean Min -0.998826 +trainer/policy/std Mean 0.428588 +trainer/policy/std Std 0.0201707 +trainer/policy/std Max 0.45068 +trainer/policy/std Min 0.394279 +trainer/Advantage Weights Mean 1.32702 +trainer/Advantage Weights Std 9.00536 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.18414e-12 +trainer/Advantage Score Mean -0.480197 +trainer/Advantage Score Std 0.395909 +trainer/Advantage Score Max 0.574455 +trainer/Advantage Score Min -2.7462 +trainer/V1 Predictions Mean -72.7322 +trainer/V1 Predictions Std 17.0577 +trainer/V1 Predictions Max 1.5604 +trainer/V1 Predictions Min -86.1401 +trainer/VF Loss 0.0420321 +expl/num steps total 321000 +expl/num paths total 362 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0389654 +expl/Actions Std 0.843645 +expl/Actions Max 2.68192 +expl/Actions Min -2.34205 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 306984 +eval/num paths total 321 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0359301 +eval/Actions Std 0.748703 +eval/Actions Max 0.99946 +eval/Actions Min -0.999756 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.43192e-06 +time/evaluation sampling (s) 2.90455 +time/exploration sampling (s) 3.92486 +time/logging (s) 0.00832232 +time/saving (s) 0.0112481 +time/training (s) 13.2482 +time/epoch (s) 20.0972 +time/total (s) 6735.93 +Epoch -680 +------------------------------ ---------------- +2022-05-15 19:55:02.582152 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -679 finished +------------------------------ ---------------- +epoch -679 +replay_buffer/size 999047 +trainer/num train calls 322000 +trainer/QF1 Loss 1.25235 +trainer/QF2 Loss 1.12683 +trainer/Policy Loss 13.7742 +trainer/Q1 Predictions Mean -71.2593 +trainer/Q1 Predictions Std 18.7428 +trainer/Q1 Predictions Max -0.256495 +trainer/Q1 Predictions Min -85.942 +trainer/Q2 Predictions Mean -71.2836 +trainer/Q2 Predictions Std 18.7529 +trainer/Q2 Predictions Max -0.339801 +trainer/Q2 Predictions Min -85.9618 +trainer/Q Targets Mean -71.555 +trainer/Q Targets Std 18.4852 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2427 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0216199 +trainer/policy/mean Std 0.706087 +trainer/policy/mean Max 0.997413 +trainer/policy/mean Min -0.997907 +trainer/policy/std Mean 0.428641 +trainer/policy/std Std 0.0200171 +trainer/policy/std Max 0.448753 +trainer/policy/std Min 0.394422 +trainer/Advantage Weights Mean 3.33311 +trainer/Advantage Weights Std 14.817 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.73755e-11 +trainer/Advantage Score Mean -0.322197 +trainer/Advantage Score Std 0.451744 +trainer/Advantage Score Max 0.724991 +trainer/Advantage Score Min -2.401 +trainer/V1 Predictions Mean -71.1593 +trainer/V1 Predictions Std 18.7747 +trainer/V1 Predictions Max 0.170747 +trainer/V1 Predictions Min -86.0208 +trainer/VF Loss 0.0416463 +expl/num steps total 322000 +expl/num paths total 364 +expl/path length Mean 500 +expl/path length Std 455 +expl/path length Max 955 +expl/path length Min 45 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.030616 +expl/Actions Std 0.808648 +expl/Actions Max 2.50617 +expl/Actions Min -2.37955 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 307984 +eval/num paths total 322 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.187899 +eval/Actions Std 0.62014 +eval/Actions Max 0.999377 +eval/Actions Min -0.99942 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.13111e-06 +time/evaluation sampling (s) 2.97672 +time/exploration sampling (s) 3.92456 +time/logging (s) 0.00691022 +time/saving (s) 0.00935311 +time/training (s) 13.3076 +time/epoch (s) 20.2252 +time/total (s) 6756.16 +Epoch -679 +------------------------------ ---------------- +2022-05-15 19:55:22.600598 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -678 finished +------------------------------ ---------------- +epoch -678 +replay_buffer/size 999047 +trainer/num train calls 323000 +trainer/QF1 Loss 1.59626 +trainer/QF2 Loss 1.47898 +trainer/Policy Loss 12.3096 +trainer/Q1 Predictions Mean -72.2797 +trainer/Q1 Predictions Std 17.6164 +trainer/Q1 Predictions Max -0.746035 +trainer/Q1 Predictions Min -86.7111 +trainer/Q2 Predictions Mean -72.2768 +trainer/Q2 Predictions Std 17.6278 +trainer/Q2 Predictions Max 0.066036 +trainer/Q2 Predictions Min -86.3226 +trainer/Q Targets Mean -72.1526 +trainer/Q Targets Std 17.6578 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.8171 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00888114 +trainer/policy/mean Std 0.711855 +trainer/policy/mean Max 0.99954 +trainer/policy/mean Min -0.997902 +trainer/policy/std Mean 0.426943 +trainer/policy/std Std 0.0209229 +trainer/policy/std Max 0.450077 +trainer/policy/std Min 0.389971 +trainer/Advantage Weights Mean 2.50744 +trainer/Advantage Weights Std 15.1177 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.86371e-12 +trainer/Advantage Score Mean -0.548387 +trainer/Advantage Score Std 0.46901 +trainer/Advantage Score Max 1.31987 +trainer/Advantage Score Min -2.53422 +trainer/V1 Predictions Mean -71.9774 +trainer/V1 Predictions Std 17.547 +trainer/V1 Predictions Max 0.39274 +trainer/V1 Predictions Min -85.6698 +trainer/VF Loss 0.0650488 +expl/num steps total 323000 +expl/num paths total 366 +expl/path length Mean 500 +expl/path length Std 351 +expl/path length Max 851 +expl/path length Min 149 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0485399 +expl/Actions Std 0.832483 +expl/Actions Max 2.38897 +expl/Actions Min -2.49442 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 308984 +eval/num paths total 323 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0238759 +eval/Actions Std 0.652397 +eval/Actions Max 0.99957 +eval/Actions Min -0.995724 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.94908e-06 +time/evaluation sampling (s) 3.04533 +time/exploration sampling (s) 3.60339 +time/logging (s) 0.00841735 +time/saving (s) 0.0135304 +time/training (s) 13.3427 +time/epoch (s) 20.0134 +time/total (s) 6776.18 +Epoch -678 +------------------------------ ---------------- +2022-05-15 19:55:42.665765 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -677 finished +------------------------------ ---------------- +epoch -677 +replay_buffer/size 999047 +trainer/num train calls 324000 +trainer/QF1 Loss 0.501096 +trainer/QF2 Loss 0.525403 +trainer/Policy Loss 4.75006 +trainer/Q1 Predictions Mean -71.0803 +trainer/Q1 Predictions Std 19.8962 +trainer/Q1 Predictions Max -0.28814 +trainer/Q1 Predictions Min -85.455 +trainer/Q2 Predictions Mean -71.0495 +trainer/Q2 Predictions Std 19.9247 +trainer/Q2 Predictions Max -0.259557 +trainer/Q2 Predictions Min -85.602 +trainer/Q Targets Mean -71.1043 +trainer/Q Targets Std 19.7029 +trainer/Q Targets Max 0.699195 +trainer/Q Targets Min -85.4688 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0262928 +trainer/policy/mean Std 0.717881 +trainer/policy/mean Max 0.997513 +trainer/policy/mean Min -0.995905 +trainer/policy/std Mean 0.426402 +trainer/policy/std Std 0.0213916 +trainer/policy/std Max 0.449883 +trainer/policy/std Min 0.388389 +trainer/Advantage Weights Mean 1.07394 +trainer/Advantage Weights Std 7.88882 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.35833e-16 +trainer/Advantage Score Mean -0.583697 +trainer/Advantage Score Std 0.563422 +trainer/Advantage Score Max 0.835618 +trainer/Advantage Score Min -3.59834 +trainer/V1 Predictions Mean -70.7853 +trainer/V1 Predictions Std 20.0406 +trainer/V1 Predictions Max 1.49656 +trainer/V1 Predictions Min -85.4531 +trainer/VF Loss 0.0697214 +expl/num steps total 324000 +expl/num paths total 368 +expl/path length Mean 500 +expl/path length Std 347 +expl/path length Max 847 +expl/path length Min 153 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0484134 +expl/Actions Std 0.820572 +expl/Actions Max 2.16046 +expl/Actions Min -2.38877 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 309984 +eval/num paths total 324 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0247483 +eval/Actions Std 0.758774 +eval/Actions Max 0.999755 +eval/Actions Min -0.999102 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.0091e-06 +time/evaluation sampling (s) 3.1494 +time/exploration sampling (s) 3.35387 +time/logging (s) 0.00681929 +time/saving (s) 0.00921304 +time/training (s) 13.5352 +time/epoch (s) 20.0545 +time/total (s) 6796.24 +Epoch -677 +------------------------------ ---------------- +2022-05-15 19:56:02.762905 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -676 finished +------------------------------ ---------------- +epoch -676 +replay_buffer/size 999047 +trainer/num train calls 325000 +trainer/QF1 Loss 0.44951 +trainer/QF2 Loss 0.473472 +trainer/Policy Loss 4.32522 +trainer/Q1 Predictions Mean -74.4388 +trainer/Q1 Predictions Std 14.5124 +trainer/Q1 Predictions Max -2.80455 +trainer/Q1 Predictions Min -86.4608 +trainer/Q2 Predictions Mean -74.4345 +trainer/Q2 Predictions Std 14.5247 +trainer/Q2 Predictions Max -2.60626 +trainer/Q2 Predictions Min -86.6613 +trainer/Q Targets Mean -74.2977 +trainer/Q Targets Std 14.5197 +trainer/Q Targets Max -3.46679 +trainer/Q Targets Min -86.452 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00590933 +trainer/policy/mean Std 0.721551 +trainer/policy/mean Max 0.997897 +trainer/policy/mean Min -0.998714 +trainer/policy/std Mean 0.426437 +trainer/policy/std Std 0.0208295 +trainer/policy/std Max 0.448699 +trainer/policy/std Min 0.390031 +trainer/Advantage Weights Mean 1.08881 +trainer/Advantage Weights Std 6.97169 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.93696e-14 +trainer/Advantage Score Mean -0.422258 +trainer/Advantage Score Std 0.451881 +trainer/Advantage Score Max 0.499825 +trainer/Advantage Score Min -3.01647 +trainer/V1 Predictions Mean -74.0659 +trainer/V1 Predictions Std 14.5931 +trainer/V1 Predictions Max -2.51246 +trainer/V1 Predictions Min -86.3113 +trainer/VF Loss 0.0410959 +expl/num steps total 325000 +expl/num paths total 370 +expl/path length Mean 500 +expl/path length Std 263 +expl/path length Max 763 +expl/path length Min 237 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0398537 +expl/Actions Std 0.823724 +expl/Actions Max 2.51388 +expl/Actions Min -2.35135 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 310984 +eval/num paths total 325 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0307342 +eval/Actions Std 0.72983 +eval/Actions Max 0.999775 +eval/Actions Min -0.999779 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.72691e-06 +time/evaluation sampling (s) 2.61588 +time/exploration sampling (s) 3.70621 +time/logging (s) 0.00748995 +time/saving (s) 0.0158443 +time/training (s) 13.7457 +time/epoch (s) 20.0911 +time/total (s) 6816.33 +Epoch -676 +------------------------------ ---------------- +2022-05-15 19:56:23.344426 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -675 finished +------------------------------ ---------------- +epoch -675 +replay_buffer/size 999047 +trainer/num train calls 326000 +trainer/QF1 Loss 0.639701 +trainer/QF2 Loss 0.692226 +trainer/Policy Loss 19.1455 +trainer/Q1 Predictions Mean -70.8436 +trainer/Q1 Predictions Std 20.5531 +trainer/Q1 Predictions Max -0.635111 +trainer/Q1 Predictions Min -85.93 +trainer/Q2 Predictions Mean -70.861 +trainer/Q2 Predictions Std 20.588 +trainer/Q2 Predictions Max -0.469774 +trainer/Q2 Predictions Min -85.784 +trainer/Q Targets Mean -70.7966 +trainer/Q Targets Std 20.9279 +trainer/Q Targets Max 0.0736846 +trainer/Q Targets Min -85.8572 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0168905 +trainer/policy/mean Std 0.715783 +trainer/policy/mean Max 0.998412 +trainer/policy/mean Min -0.999195 +trainer/policy/std Mean 0.427277 +trainer/policy/std Std 0.021184 +trainer/policy/std Max 0.449895 +trainer/policy/std Min 0.389353 +trainer/Advantage Weights Mean 2.83617 +trainer/Advantage Weights Std 10.2804 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.47386e-16 +trainer/Advantage Score Mean -0.456916 +trainer/Advantage Score Std 0.630079 +trainer/Advantage Score Max 0.480499 +trainer/Advantage Score Min -3.64535 +trainer/V1 Predictions Mean -70.5372 +trainer/V1 Predictions Std 21.0749 +trainer/V1 Predictions Max 1.24666 +trainer/V1 Predictions Min -85.7518 +trainer/VF Loss 0.0684168 +expl/num steps total 326000 +expl/num paths total 371 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.148053 +expl/Actions Std 0.82774 +expl/Actions Max 2.45872 +expl/Actions Min -2.0957 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 311980 +eval/num paths total 326 +eval/path length Mean 996 +eval/path length Std 0 +eval/path length Max 996 +eval/path length Min 996 +eval/Rewards Mean 0.00100402 +eval/Rewards Std 0.0316703 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0226662 +eval/Actions Std 0.749308 +eval/Actions Max 0.999638 +eval/Actions Min -0.998648 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.97278e-06 +time/evaluation sampling (s) 3.0529 +time/exploration sampling (s) 3.95994 +time/logging (s) 0.00738245 +time/saving (s) 0.0102941 +time/training (s) 13.5432 +time/epoch (s) 20.5738 +time/total (s) 6836.91 +Epoch -675 +------------------------------ ---------------- +2022-05-15 19:56:44.072953 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -674 finished +------------------------------ ---------------- +epoch -674 +replay_buffer/size 999047 +trainer/num train calls 327000 +trainer/QF1 Loss 0.829544 +trainer/QF2 Loss 0.893374 +trainer/Policy Loss 12.2849 +trainer/Q1 Predictions Mean -70.6926 +trainer/Q1 Predictions Std 19.9633 +trainer/Q1 Predictions Max -0.384256 +trainer/Q1 Predictions Min -86.4329 +trainer/Q2 Predictions Mean -70.7056 +trainer/Q2 Predictions Std 19.9602 +trainer/Q2 Predictions Max -0.355204 +trainer/Q2 Predictions Min -86.7708 +trainer/Q Targets Mean -70.5835 +trainer/Q Targets Std 19.6875 +trainer/Q Targets Max 1.53042 +trainer/Q Targets Min -86.6071 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0133209 +trainer/policy/mean Std 0.714202 +trainer/policy/mean Max 0.998847 +trainer/policy/mean Min -0.998364 +trainer/policy/std Mean 0.427176 +trainer/policy/std Std 0.0221943 +trainer/policy/std Max 0.450291 +trainer/policy/std Min 0.386045 +trainer/Advantage Weights Mean 2.74345 +trainer/Advantage Weights Std 13.5663 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.5017e-14 +trainer/Advantage Score Mean -0.430229 +trainer/Advantage Score Std 0.492997 +trainer/Advantage Score Max 1.2513 +trainer/Advantage Score Min -3.07317 +trainer/V1 Predictions Mean -70.2514 +trainer/V1 Predictions Std 19.9619 +trainer/V1 Predictions Max 1.92686 +trainer/V1 Predictions Min -86.1981 +trainer/VF Loss 0.052794 +expl/num steps total 327000 +expl/num paths total 372 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.288753 +expl/Actions Std 0.844122 +expl/Actions Max 2.33669 +expl/Actions Min -2.37884 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 312658 +eval/num paths total 327 +eval/path length Mean 678 +eval/path length Std 0 +eval/path length Max 678 +eval/path length Min 678 +eval/Rewards Mean 0.00147493 +eval/Rewards Std 0.0383764 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.04243 +eval/Actions Std 0.70666 +eval/Actions Max 0.999226 +eval/Actions Min -0.999187 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.35509e-06 +time/evaluation sampling (s) 3.0841 +time/exploration sampling (s) 3.93769 +time/logging (s) 0.00929249 +time/saving (s) 0.0148416 +time/training (s) 13.6774 +time/epoch (s) 20.7233 +time/total (s) 6857.64 +Epoch -674 +------------------------------ ---------------- +2022-05-15 19:57:03.423208 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -673 finished +------------------------------ ---------------- +epoch -673 +replay_buffer/size 999047 +trainer/num train calls 328000 +trainer/QF1 Loss 0.754994 +trainer/QF2 Loss 0.70362 +trainer/Policy Loss 7.86442 +trainer/Q1 Predictions Mean -69.1993 +trainer/Q1 Predictions Std 20.9186 +trainer/Q1 Predictions Max -0.931255 +trainer/Q1 Predictions Min -85.7036 +trainer/Q2 Predictions Mean -69.1974 +trainer/Q2 Predictions Std 20.8791 +trainer/Q2 Predictions Max -0.795085 +trainer/Q2 Predictions Min -85.7931 +trainer/Q Targets Mean -69.085 +trainer/Q Targets Std 21.0202 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7002 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0122422 +trainer/policy/mean Std 0.7097 +trainer/policy/mean Max 0.99822 +trainer/policy/mean Min -0.998755 +trainer/policy/std Mean 0.425322 +trainer/policy/std Std 0.0191335 +trainer/policy/std Max 0.447496 +trainer/policy/std Min 0.391588 +trainer/Advantage Weights Mean 1.75305 +trainer/Advantage Weights Std 11.3316 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.44847e-13 +trainer/Advantage Score Mean -0.581125 +trainer/Advantage Score Std 0.560386 +trainer/Advantage Score Max 1.22177 +trainer/Advantage Score Min -2.90381 +trainer/V1 Predictions Mean -68.8704 +trainer/V1 Predictions Std 21.1025 +trainer/V1 Predictions Max 0.134572 +trainer/V1 Predictions Min -85.6298 +trainer/VF Loss 0.0734995 +expl/num steps total 328000 +expl/num paths total 374 +expl/path length Mean 500 +expl/path length Std 253 +expl/path length Max 753 +expl/path length Min 247 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0281115 +expl/Actions Std 0.823551 +expl/Actions Max 2.19498 +expl/Actions Min -2.16382 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 313658 +eval/num paths total 328 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0674735 +eval/Actions Std 0.726725 +eval/Actions Max 0.999889 +eval/Actions Min -0.999311 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.42308e-06 +time/evaluation sampling (s) 3.07885 +time/exploration sampling (s) 3.57789 +time/logging (s) 0.00709511 +time/saving (s) 0.00919594 +time/training (s) 12.6648 +time/epoch (s) 19.3378 +time/total (s) 6876.98 +Epoch -673 +------------------------------ ---------------- +2022-05-15 19:57:23.809513 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -672 finished +------------------------------ ---------------- +epoch -672 +replay_buffer/size 999047 +trainer/num train calls 329000 +trainer/QF1 Loss 1.85828 +trainer/QF2 Loss 1.92057 +trainer/Policy Loss 3.10257 +trainer/Q1 Predictions Mean -69.6805 +trainer/Q1 Predictions Std 20.6332 +trainer/Q1 Predictions Max -2.32285 +trainer/Q1 Predictions Min -85.9541 +trainer/Q2 Predictions Mean -69.6877 +trainer/Q2 Predictions Std 20.5936 +trainer/Q2 Predictions Max -2.33372 +trainer/Q2 Predictions Min -85.7596 +trainer/Q Targets Mean -69.5877 +trainer/Q Targets Std 20.8616 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7166 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0032593 +trainer/policy/mean Std 0.711388 +trainer/policy/mean Max 0.998751 +trainer/policy/mean Min -0.998276 +trainer/policy/std Mean 0.42563 +trainer/policy/std Std 0.0204945 +trainer/policy/std Max 0.446284 +trainer/policy/std Min 0.388588 +trainer/Advantage Weights Mean 1.06438 +trainer/Advantage Weights Std 8.8578 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.06577e-16 +trainer/Advantage Score Mean -0.585582 +trainer/Advantage Score Std 0.618484 +trainer/Advantage Score Max 0.898315 +trainer/Advantage Score Min -3.50387 +trainer/V1 Predictions Mean -69.4252 +trainer/V1 Predictions Std 20.8051 +trainer/V1 Predictions Max -1.12262 +trainer/V1 Predictions Min -85.8607 +trainer/VF Loss 0.0769635 +expl/num steps total 329000 +expl/num paths total 375 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.14665 +expl/Actions Std 0.828041 +expl/Actions Max 2.51191 +expl/Actions Min -2.16155 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 314658 +eval/num paths total 329 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.227423 +eval/Actions Std 0.709403 +eval/Actions Max 0.998882 +eval/Actions Min -0.996729 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.96207e-06 +time/evaluation sampling (s) 3.1512 +time/exploration sampling (s) 3.2562 +time/logging (s) 0.0081575 +time/saving (s) 0.0125375 +time/training (s) 13.9524 +time/epoch (s) 20.3805 +time/total (s) 6897.37 +Epoch -672 +------------------------------ ---------------- +2022-05-15 19:57:43.702038 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -671 finished +------------------------------ ---------------- +epoch -671 +replay_buffer/size 999047 +trainer/num train calls 330000 +trainer/QF1 Loss 0.720163 +trainer/QF2 Loss 0.860215 +trainer/Policy Loss 13.5368 +trainer/Q1 Predictions Mean -72.0568 +trainer/Q1 Predictions Std 18.0103 +trainer/Q1 Predictions Max -1.23475 +trainer/Q1 Predictions Min -86.6402 +trainer/Q2 Predictions Mean -71.991 +trainer/Q2 Predictions Std 18.1647 +trainer/Q2 Predictions Max -0.913441 +trainer/Q2 Predictions Min -86.9805 +trainer/Q Targets Mean -71.8432 +trainer/Q Targets Std 17.723 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3814 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0142432 +trainer/policy/mean Std 0.731788 +trainer/policy/mean Max 0.999402 +trainer/policy/mean Min -0.9994 +trainer/policy/std Mean 0.427025 +trainer/policy/std Std 0.0214188 +trainer/policy/std Max 0.451763 +trainer/policy/std Min 0.390373 +trainer/Advantage Weights Mean 2.39085 +trainer/Advantage Weights Std 12.1557 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.23829e-10 +trainer/Advantage Score Mean -0.419809 +trainer/Advantage Score Std 0.421726 +trainer/Advantage Score Max 0.935436 +trainer/Advantage Score Min -2.28121 +trainer/V1 Predictions Mean -71.642 +trainer/V1 Predictions Std 17.7293 +trainer/V1 Predictions Max -0.412199 +trainer/V1 Predictions Min -86.2504 +trainer/VF Loss 0.0446293 +expl/num steps total 330000 +expl/num paths total 377 +expl/path length Mean 500 +expl/path length Std 86 +expl/path length Max 586 +expl/path length Min 414 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0682402 +expl/Actions Std 0.846092 +expl/Actions Max 2.42064 +expl/Actions Min -2.33541 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 315658 +eval/num paths total 330 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.113768 +eval/Actions Std 0.718089 +eval/Actions Max 0.999576 +eval/Actions Min -0.999888 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.52576e-06 +time/evaluation sampling (s) 2.74538 +time/exploration sampling (s) 3.88234 +time/logging (s) 0.00772298 +time/saving (s) 0.0204423 +time/training (s) 13.2273 +time/epoch (s) 19.8832 +time/total (s) 6917.25 +Epoch -671 +------------------------------ ---------------- +2022-05-15 19:58:03.880591 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -670 finished +------------------------------ --------------- +epoch -670 +replay_buffer/size 999047 +trainer/num train calls 331000 +trainer/QF1 Loss 1.28375 +trainer/QF2 Loss 1.13855 +trainer/Policy Loss 60.1136 +trainer/Q1 Predictions Mean -70.7845 +trainer/Q1 Predictions Std 18.5564 +trainer/Q1 Predictions Max -1.41599 +trainer/Q1 Predictions Min -86.1937 +trainer/Q2 Predictions Mean -70.7575 +trainer/Q2 Predictions Std 18.5883 +trainer/Q2 Predictions Max -1.92795 +trainer/Q2 Predictions Min -86.1794 +trainer/Q Targets Mean -70.9927 +trainer/Q Targets Std 18.9362 +trainer/Q Targets Max -1.65915 +trainer/Q Targets Min -86.4271 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0129381 +trainer/policy/mean Std 0.719025 +trainer/policy/mean Max 0.999977 +trainer/policy/mean Min -0.999026 +trainer/policy/std Mean 0.426108 +trainer/policy/std Std 0.0217568 +trainer/policy/std Max 0.450179 +trainer/policy/std Min 0.38782 +trainer/Advantage Weights Mean 9.71252 +trainer/Advantage Weights Std 25.6013 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.399e-16 +trainer/Advantage Score Mean -0.381158 +trainer/Advantage Score Std 0.725959 +trainer/Advantage Score Max 1.73662 +trainer/Advantage Score Min -3.47133 +trainer/V1 Predictions Mean -70.7043 +trainer/V1 Predictions Std 18.9218 +trainer/V1 Predictions Max -0.0704635 +trainer/V1 Predictions Min -86.2413 +trainer/VF Loss 0.112653 +expl/num steps total 331000 +expl/num paths total 378 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0550328 +expl/Actions Std 0.813176 +expl/Actions Max 2.39008 +expl/Actions Min -2.1814 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 316658 +eval/num paths total 331 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00684537 +eval/Actions Std 0.709587 +eval/Actions Max 0.999696 +eval/Actions Min -0.99871 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.6268e-06 +time/evaluation sampling (s) 3.24887 +time/exploration sampling (s) 3.91713 +time/logging (s) 0.00753495 +time/saving (s) 0.0118681 +time/training (s) 12.9844 +time/epoch (s) 20.1698 +time/total (s) 6937.43 +Epoch -670 +------------------------------ --------------- +2022-05-15 19:58:24.527352 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -669 finished +------------------------------ ---------------- +epoch -669 +replay_buffer/size 999047 +trainer/num train calls 332000 +trainer/QF1 Loss 0.622038 +trainer/QF2 Loss 0.70722 +trainer/Policy Loss 3.01058 +trainer/Q1 Predictions Mean -72.3474 +trainer/Q1 Predictions Std 17.8031 +trainer/Q1 Predictions Max -1.14418 +trainer/Q1 Predictions Min -85.6157 +trainer/Q2 Predictions Mean -72.4345 +trainer/Q2 Predictions Std 17.8375 +trainer/Q2 Predictions Max -1.67187 +trainer/Q2 Predictions Min -86.0466 +trainer/Q Targets Mean -71.85 +trainer/Q Targets Std 17.803 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.5961 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0141755 +trainer/policy/mean Std 0.72975 +trainer/policy/mean Max 0.999628 +trainer/policy/mean Min -0.999391 +trainer/policy/std Mean 0.425401 +trainer/policy/std Std 0.0208597 +trainer/policy/std Max 0.448922 +trainer/policy/std Min 0.389037 +trainer/Advantage Weights Mean 1.12516 +trainer/Advantage Weights Std 8.91577 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.58421e-13 +trainer/Advantage Score Mean -0.507193 +trainer/Advantage Score Std 0.516061 +trainer/Advantage Score Max 1.2176 +trainer/Advantage Score Min -2.8411 +trainer/V1 Predictions Mean -71.5222 +trainer/V1 Predictions Std 17.9434 +trainer/V1 Predictions Max -1.04352 +trainer/V1 Predictions Min -85.3654 +trainer/VF Loss 0.0612082 +expl/num steps total 332000 +expl/num paths total 379 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0274957 +expl/Actions Std 0.848363 +expl/Actions Max 2.30354 +expl/Actions Min -2.24548 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 317658 +eval/num paths total 332 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.314083 +eval/Actions Std 0.684588 +eval/Actions Max 0.998973 +eval/Actions Min -0.99929 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.84007e-06 +time/evaluation sampling (s) 3.17775 +time/exploration sampling (s) 4.0598 +time/logging (s) 0.0094865 +time/saving (s) 0.0120713 +time/training (s) 13.3813 +time/epoch (s) 20.6404 +time/total (s) 6958.07 +Epoch -669 +------------------------------ ---------------- +2022-05-15 19:58:44.491815 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -668 finished +------------------------------ ---------------- +epoch -668 +replay_buffer/size 999047 +trainer/num train calls 333000 +trainer/QF1 Loss 3.92127 +trainer/QF2 Loss 3.8639 +trainer/Policy Loss 41.177 +trainer/Q1 Predictions Mean -73.0077 +trainer/Q1 Predictions Std 16.9904 +trainer/Q1 Predictions Max -0.954885 +trainer/Q1 Predictions Min -86.1407 +trainer/Q2 Predictions Mean -72.9203 +trainer/Q2 Predictions Std 17.0363 +trainer/Q2 Predictions Max -1.05454 +trainer/Q2 Predictions Min -85.8978 +trainer/Q Targets Mean -73.0141 +trainer/Q Targets Std 17.0271 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2223 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00317463 +trainer/policy/mean Std 0.703729 +trainer/policy/mean Max 0.998665 +trainer/policy/mean Min -0.998148 +trainer/policy/std Mean 0.426664 +trainer/policy/std Std 0.0213691 +trainer/policy/std Max 0.452635 +trainer/policy/std Min 0.38888 +trainer/Advantage Weights Mean 7.16315 +trainer/Advantage Weights Std 20.7361 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.32499e-10 +trainer/Advantage Score Mean -0.197107 +trainer/Advantage Score Std 0.502102 +trainer/Advantage Score Max 1.75401 +trainer/Advantage Score Min -2.27444 +trainer/V1 Predictions Mean -72.8862 +trainer/V1 Predictions Std 17.102 +trainer/V1 Predictions Max -0.259992 +trainer/V1 Predictions Min -86.3591 +trainer/VF Loss 0.0653214 +expl/num steps total 333000 +expl/num paths total 380 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.202529 +expl/Actions Std 0.800999 +expl/Actions Max 2.27091 +expl/Actions Min -2.21294 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 318658 +eval/num paths total 333 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.119994 +eval/Actions Std 0.60155 +eval/Actions Max 0.999292 +eval/Actions Min -0.998183 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.93487e-06 +time/evaluation sampling (s) 3.17526 +time/exploration sampling (s) 3.55086 +time/logging (s) 0.00703038 +time/saving (s) 0.0100265 +time/training (s) 13.2099 +time/epoch (s) 19.9531 +time/total (s) 6978.03 +Epoch -668 +------------------------------ ---------------- +2022-05-15 19:59:02.525185 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -667 finished +------------------------------ ---------------- +epoch -667 +replay_buffer/size 999047 +trainer/num train calls 334000 +trainer/QF1 Loss 0.825357 +trainer/QF2 Loss 0.826436 +trainer/Policy Loss 25.7904 +trainer/Q1 Predictions Mean -70.6744 +trainer/Q1 Predictions Std 20.0145 +trainer/Q1 Predictions Max -0.37772 +trainer/Q1 Predictions Min -86.0758 +trainer/Q2 Predictions Mean -70.6101 +trainer/Q2 Predictions Std 20.0227 +trainer/Q2 Predictions Max -0.670954 +trainer/Q2 Predictions Min -85.992 +trainer/Q Targets Mean -70.6844 +trainer/Q Targets Std 20.0164 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.1199 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00748266 +trainer/policy/mean Std 0.710702 +trainer/policy/mean Max 0.999111 +trainer/policy/mean Min -0.99785 +trainer/policy/std Mean 0.427161 +trainer/policy/std Std 0.0212916 +trainer/policy/std Max 0.453817 +trainer/policy/std Min 0.390257 +trainer/Advantage Weights Mean 7.47934 +trainer/Advantage Weights Std 22.0671 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.90862e-17 +trainer/Advantage Score Mean -0.210059 +trainer/Advantage Score Std 0.655889 +trainer/Advantage Score Max 2.46902 +trainer/Advantage Score Min -3.73675 +trainer/V1 Predictions Mean -70.4388 +trainer/V1 Predictions Std 20.1587 +trainer/V1 Predictions Max 2.32329 +trainer/V1 Predictions Min -85.9955 +trainer/VF Loss 0.109771 +expl/num steps total 334000 +expl/num paths total 381 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00513984 +expl/Actions Std 0.81194 +expl/Actions Max 2.4145 +expl/Actions Min -2.78692 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 319658 +eval/num paths total 334 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.134884 +eval/Actions Std 0.754864 +eval/Actions Max 0.999149 +eval/Actions Min -0.999405 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.00631e-06 +time/evaluation sampling (s) 2.94876 +time/exploration sampling (s) 3.13905 +time/logging (s) 0.00677529 +time/saving (s) 0.00932663 +time/training (s) 11.9221 +time/epoch (s) 18.026 +time/total (s) 6996.06 +Epoch -667 +------------------------------ ---------------- +2022-05-15 19:59:19.755060 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -666 finished +------------------------------ ---------------- +epoch -666 +replay_buffer/size 999047 +trainer/num train calls 335000 +trainer/QF1 Loss 0.658432 +trainer/QF2 Loss 0.689313 +trainer/Policy Loss 6.69191 +trainer/Q1 Predictions Mean -73.5473 +trainer/Q1 Predictions Std 15.5009 +trainer/Q1 Predictions Max -0.557987 +trainer/Q1 Predictions Min -86.4803 +trainer/Q2 Predictions Mean -73.6227 +trainer/Q2 Predictions Std 15.4871 +trainer/Q2 Predictions Max 0.123711 +trainer/Q2 Predictions Min -86.4635 +trainer/Q Targets Mean -73.0873 +trainer/Q Targets Std 15.3807 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7837 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00846131 +trainer/policy/mean Std 0.703243 +trainer/policy/mean Max 0.999345 +trainer/policy/mean Min -0.998735 +trainer/policy/std Mean 0.425991 +trainer/policy/std Std 0.0206987 +trainer/policy/std Max 0.448388 +trainer/policy/std Min 0.389409 +trainer/Advantage Weights Mean 1.74282 +trainer/Advantage Weights Std 10.0363 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.20748e-13 +trainer/Advantage Score Mean -0.464892 +trainer/Advantage Score Std 0.465396 +trainer/Advantage Score Max 0.576645 +trainer/Advantage Score Min -2.84967 +trainer/V1 Predictions Mean -72.8345 +trainer/V1 Predictions Std 15.4971 +trainer/V1 Predictions Max 0.607002 +trainer/V1 Predictions Min -85.6109 +trainer/VF Loss 0.0477892 +expl/num steps total 335000 +expl/num paths total 383 +expl/path length Mean 500 +expl/path length Std 433 +expl/path length Max 933 +expl/path length Min 67 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0453793 +expl/Actions Std 0.854648 +expl/Actions Max 2.73337 +expl/Actions Min -2.33618 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 320578 +eval/num paths total 335 +eval/path length Mean 920 +eval/path length Std 0 +eval/path length Max 920 +eval/path length Min 920 +eval/Rewards Mean 0.00108696 +eval/Rewards Std 0.0329511 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0495424 +eval/Actions Std 0.733038 +eval/Actions Max 0.999509 +eval/Actions Min -0.998857 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.89315e-06 +time/evaluation sampling (s) 2.41796 +time/exploration sampling (s) 2.82727 +time/logging (s) 0.00674298 +time/saving (s) 0.00955113 +time/training (s) 11.9614 +time/epoch (s) 17.223 +time/total (s) 7013.29 +Epoch -666 +------------------------------ ---------------- +2022-05-15 19:59:37.171300 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -665 finished +------------------------------ ---------------- +epoch -665 +replay_buffer/size 999047 +trainer/num train calls 336000 +trainer/QF1 Loss 0.662788 +trainer/QF2 Loss 0.677381 +trainer/Policy Loss 8.9207 +trainer/Q1 Predictions Mean -73.1025 +trainer/Q1 Predictions Std 16.5859 +trainer/Q1 Predictions Max -2.21091 +trainer/Q1 Predictions Min -86.7637 +trainer/Q2 Predictions Mean -73.12 +trainer/Q2 Predictions Std 16.5787 +trainer/Q2 Predictions Max -2.59595 +trainer/Q2 Predictions Min -86.979 +trainer/Q Targets Mean -73.1142 +trainer/Q Targets Std 16.7042 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6397 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00595595 +trainer/policy/mean Std 0.719923 +trainer/policy/mean Max 0.998906 +trainer/policy/mean Min -0.998137 +trainer/policy/std Mean 0.427659 +trainer/policy/std Std 0.020003 +trainer/policy/std Max 0.450499 +trainer/policy/std Min 0.393227 +trainer/Advantage Weights Mean 1.74982 +trainer/Advantage Weights Std 9.35262 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.1988e-13 +trainer/Advantage Score Mean -0.344096 +trainer/Advantage Score Std 0.445952 +trainer/Advantage Score Max 0.606121 +trainer/Advantage Score Min -2.82852 +trainer/V1 Predictions Mean -72.9019 +trainer/V1 Predictions Std 16.7422 +trainer/V1 Predictions Max -2.79878 +trainer/V1 Predictions Min -86.5742 +trainer/VF Loss 0.0364629 +expl/num steps total 336000 +expl/num paths total 384 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0223036 +expl/Actions Std 0.826265 +expl/Actions Max 2.32577 +expl/Actions Min -2.21764 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 321578 +eval/num paths total 336 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.14326 +eval/Actions Std 0.621729 +eval/Actions Max 0.999593 +eval/Actions Min -0.99791 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.62493e-06 +time/evaluation sampling (s) 2.43758 +time/exploration sampling (s) 2.85809 +time/logging (s) 0.00678623 +time/saving (s) 0.00930588 +time/training (s) 12.0976 +time/epoch (s) 17.4093 +time/total (s) 7030.7 +Epoch -665 +------------------------------ ---------------- +2022-05-15 19:59:54.303232 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -664 finished +------------------------------ ---------------- +epoch -664 +replay_buffer/size 999047 +trainer/num train calls 337000 +trainer/QF1 Loss 0.544627 +trainer/QF2 Loss 0.552832 +trainer/Policy Loss 8.40955 +trainer/Q1 Predictions Mean -72.2697 +trainer/Q1 Predictions Std 18.185 +trainer/Q1 Predictions Max -0.253138 +trainer/Q1 Predictions Min -85.971 +trainer/Q2 Predictions Mean -72.3192 +trainer/Q2 Predictions Std 18.1967 +trainer/Q2 Predictions Max -0.362186 +trainer/Q2 Predictions Min -86.0208 +trainer/Q Targets Mean -72.2031 +trainer/Q Targets Std 18.0111 +trainer/Q Targets Max 2.08445 +trainer/Q Targets Min -85.919 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00338434 +trainer/policy/mean Std 0.710395 +trainer/policy/mean Max 0.998493 +trainer/policy/mean Min -0.997726 +trainer/policy/std Mean 0.426242 +trainer/policy/std Std 0.0195873 +trainer/policy/std Max 0.449818 +trainer/policy/std Min 0.393871 +trainer/Advantage Weights Mean 2.98503 +trainer/Advantage Weights Std 13.8162 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.87509e-19 +trainer/Advantage Score Mean -0.35396 +trainer/Advantage Score Std 0.557624 +trainer/Advantage Score Max 1.96593 +trainer/Advantage Score Min -4.31205 +trainer/V1 Predictions Mean -71.9518 +trainer/V1 Predictions Std 18.2225 +trainer/V1 Predictions Max 3.95533 +trainer/V1 Predictions Min -85.71 +trainer/VF Loss 0.0649871 +expl/num steps total 337000 +expl/num paths total 386 +expl/path length Mean 500 +expl/path length Std 369 +expl/path length Max 869 +expl/path length Min 131 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0327225 +expl/Actions Std 0.829566 +expl/Actions Max 2.28909 +expl/Actions Min -2.34061 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 322368 +eval/num paths total 337 +eval/path length Mean 790 +eval/path length Std 0 +eval/path length Max 790 +eval/path length Min 790 +eval/Rewards Mean 0.00126582 +eval/Rewards Std 0.0355559 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0447497 +eval/Actions Std 0.721398 +eval/Actions Max 0.999617 +eval/Actions Min -0.999543 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.39165e-06 +time/evaluation sampling (s) 2.41634 +time/exploration sampling (s) 2.863 +time/logging (s) 0.00732188 +time/saving (s) 0.0124484 +time/training (s) 11.8265 +time/epoch (s) 17.1256 +time/total (s) 7047.83 +Epoch -664 +------------------------------ ---------------- +2022-05-15 20:00:10.795500 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -663 finished +------------------------------ ---------------- +epoch -663 +replay_buffer/size 999047 +trainer/num train calls 338000 +trainer/QF1 Loss 0.596137 +trainer/QF2 Loss 0.620486 +trainer/Policy Loss 3.969 +trainer/Q1 Predictions Mean -71.6556 +trainer/Q1 Predictions Std 18.6044 +trainer/Q1 Predictions Max -0.653253 +trainer/Q1 Predictions Min -85.7343 +trainer/Q2 Predictions Mean -71.6749 +trainer/Q2 Predictions Std 18.5638 +trainer/Q2 Predictions Max -0.61125 +trainer/Q2 Predictions Min -85.8254 +trainer/Q Targets Mean -71.5592 +trainer/Q Targets Std 18.7486 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.3832 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0138449 +trainer/policy/mean Std 0.712685 +trainer/policy/mean Max 0.998894 +trainer/policy/mean Min -0.998635 +trainer/policy/std Mean 0.425919 +trainer/policy/std Std 0.0207477 +trainer/policy/std Max 0.449538 +trainer/policy/std Min 0.392437 +trainer/Advantage Weights Mean 0.748582 +trainer/Advantage Weights Std 3.07896 +trainer/Advantage Weights Max 41.2116 +trainer/Advantage Weights Min 2.45115e-16 +trainer/Advantage Score Mean -0.473735 +trainer/Advantage Score Std 0.614939 +trainer/Advantage Score Max 0.371872 +trainer/Advantage Score Min -3.59448 +trainer/V1 Predictions Mean -71.298 +trainer/V1 Predictions Std 18.9185 +trainer/V1 Predictions Max -0.0343687 +trainer/V1 Predictions Min -85.3123 +trainer/VF Loss 0.0623705 +expl/num steps total 338000 +expl/num paths total 387 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0333176 +expl/Actions Std 0.828648 +expl/Actions Max 2.15019 +expl/Actions Min -2.5154 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 323031 +eval/num paths total 338 +eval/path length Mean 663 +eval/path length Std 0 +eval/path length Max 663 +eval/path length Min 663 +eval/Rewards Mean 0.0015083 +eval/Rewards Std 0.0388075 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0291316 +eval/Actions Std 0.730364 +eval/Actions Max 0.999941 +eval/Actions Min -0.999476 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.42284e-06 +time/evaluation sampling (s) 2.43528 +time/exploration sampling (s) 2.68748 +time/logging (s) 0.00692155 +time/saving (s) 0.0125297 +time/training (s) 11.341 +time/epoch (s) 16.4832 +time/total (s) 7064.32 +Epoch -663 +------------------------------ ---------------- +2022-05-15 20:00:27.662170 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -662 finished +------------------------------ ---------------- +epoch -662 +replay_buffer/size 999047 +trainer/num train calls 339000 +trainer/QF1 Loss 0.610249 +trainer/QF2 Loss 0.520433 +trainer/Policy Loss 6.13264 +trainer/Q1 Predictions Mean -73.8446 +trainer/Q1 Predictions Std 17.0306 +trainer/Q1 Predictions Max -1.0149 +trainer/Q1 Predictions Min -86.4612 +trainer/Q2 Predictions Mean -73.8403 +trainer/Q2 Predictions Std 17.0577 +trainer/Q2 Predictions Max -0.338792 +trainer/Q2 Predictions Min -86.7401 +trainer/Q Targets Mean -73.7915 +trainer/Q Targets Std 17.4654 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4584 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0132794 +trainer/policy/mean Std 0.714366 +trainer/policy/mean Max 0.998961 +trainer/policy/mean Min -0.999351 +trainer/policy/std Mean 0.426511 +trainer/policy/std Std 0.0209655 +trainer/policy/std Max 0.450918 +trainer/policy/std Min 0.393408 +trainer/Advantage Weights Mean 1.86491 +trainer/Advantage Weights Std 11.2894 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.7295e-12 +trainer/Advantage Score Mean -0.431495 +trainer/Advantage Score Std 0.522326 +trainer/Advantage Score Max 0.718048 +trainer/Advantage Score Min -2.70832 +trainer/V1 Predictions Mean -73.5367 +trainer/V1 Predictions Std 17.518 +trainer/V1 Predictions Max 0.590267 +trainer/V1 Predictions Min -86.3085 +trainer/VF Loss 0.0513368 +expl/num steps total 339000 +expl/num paths total 389 +expl/path length Mean 500 +expl/path length Std 371 +expl/path length Max 871 +expl/path length Min 129 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0379211 +expl/Actions Std 0.83175 +expl/Actions Max 2.39485 +expl/Actions Min -2.30688 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 324031 +eval/num paths total 339 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0430258 +eval/Actions Std 0.743698 +eval/Actions Max 0.999618 +eval/Actions Min -0.998791 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.52506e-06 +time/evaluation sampling (s) 2.52556 +time/exploration sampling (s) 2.70122 +time/logging (s) 0.00688468 +time/saving (s) 0.00948019 +time/training (s) 11.6148 +time/epoch (s) 16.858 +time/total (s) 7081.18 +Epoch -662 +------------------------------ ---------------- +2022-05-15 20:00:44.328777 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -661 finished +------------------------------ ---------------- +epoch -661 +replay_buffer/size 999047 +trainer/num train calls 340000 +trainer/QF1 Loss 1.04228 +trainer/QF2 Loss 1.0157 +trainer/Policy Loss 4.6228 +trainer/Q1 Predictions Mean -72.892 +trainer/Q1 Predictions Std 18.0247 +trainer/Q1 Predictions Max -1.43305 +trainer/Q1 Predictions Min -85.7903 +trainer/Q2 Predictions Mean -72.9344 +trainer/Q2 Predictions Std 18.0015 +trainer/Q2 Predictions Max -0.356808 +trainer/Q2 Predictions Min -85.8708 +trainer/Q Targets Mean -72.5065 +trainer/Q Targets Std 18.1605 +trainer/Q Targets Max 0.253309 +trainer/Q Targets Min -85.3676 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0225044 +trainer/policy/mean Std 0.719476 +trainer/policy/mean Max 0.999813 +trainer/policy/mean Min -0.999314 +trainer/policy/std Mean 0.427558 +trainer/policy/std Std 0.0219928 +trainer/policy/std Max 0.450949 +trainer/policy/std Min 0.389738 +trainer/Advantage Weights Mean 1.64217 +trainer/Advantage Weights Std 10.9275 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.74136e-16 +trainer/Advantage Score Mean -0.506937 +trainer/Advantage Score Std 0.542793 +trainer/Advantage Score Max 0.955991 +trainer/Advantage Score Min -3.62867 +trainer/V1 Predictions Mean -72.2682 +trainer/V1 Predictions Std 18.2545 +trainer/V1 Predictions Max 0.13528 +trainer/V1 Predictions Min -85.1936 +trainer/VF Loss 0.0622697 +expl/num steps total 340000 +expl/num paths total 390 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0420292 +expl/Actions Std 0.825746 +expl/Actions Max 2.75468 +expl/Actions Min -2.16197 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 325031 +eval/num paths total 340 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00529762 +eval/Actions Std 0.730708 +eval/Actions Max 0.999866 +eval/Actions Min -0.999365 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.65101e-06 +time/evaluation sampling (s) 2.42254 +time/exploration sampling (s) 2.52737 +time/logging (s) 0.00672386 +time/saving (s) 0.00925192 +time/training (s) 11.6936 +time/epoch (s) 16.6595 +time/total (s) 7097.84 +Epoch -661 +------------------------------ ---------------- +2022-05-15 20:01:01.157022 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -660 finished +------------------------------ ---------------- +epoch -660 +replay_buffer/size 999047 +trainer/num train calls 341000 +trainer/QF1 Loss 0.7877 +trainer/QF2 Loss 0.731309 +trainer/Policy Loss 12.4829 +trainer/Q1 Predictions Mean -72.0433 +trainer/Q1 Predictions Std 17.8719 +trainer/Q1 Predictions Max -0.334994 +trainer/Q1 Predictions Min -86.1195 +trainer/Q2 Predictions Mean -72.0438 +trainer/Q2 Predictions Std 17.9315 +trainer/Q2 Predictions Max -0.458163 +trainer/Q2 Predictions Min -85.972 +trainer/Q Targets Mean -71.8414 +trainer/Q Targets Std 18.0654 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.5644 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000236097 +trainer/policy/mean Std 0.704818 +trainer/policy/mean Max 0.999147 +trainer/policy/mean Min -0.997941 +trainer/policy/std Mean 0.426493 +trainer/policy/std Std 0.022176 +trainer/policy/std Max 0.448696 +trainer/policy/std Min 0.388813 +trainer/Advantage Weights Mean 3.70626 +trainer/Advantage Weights Std 15.1636 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.03063e-11 +trainer/Advantage Score Mean -0.406012 +trainer/Advantage Score Std 0.517329 +trainer/Advantage Score Max 0.9505 +trainer/Advantage Score Min -2.42197 +trainer/V1 Predictions Mean -71.6433 +trainer/V1 Predictions Std 17.9516 +trainer/V1 Predictions Max 0.385957 +trainer/V1 Predictions Min -85.5889 +trainer/VF Loss 0.0555121 +expl/num steps total 341000 +expl/num paths total 391 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0488155 +expl/Actions Std 0.817084 +expl/Actions Max 2.46155 +expl/Actions Min -2.37809 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 326031 +eval/num paths total 341 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.230255 +eval/Actions Std 0.64679 +eval/Actions Max 0.999515 +eval/Actions Min -0.999508 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.29203e-06 +time/evaluation sampling (s) 2.41008 +time/exploration sampling (s) 2.53003 +time/logging (s) 0.0070299 +time/saving (s) 0.0101247 +time/training (s) 11.8645 +time/epoch (s) 16.8217 +time/total (s) 7114.67 +Epoch -660 +------------------------------ ---------------- +2022-05-15 20:01:17.923199 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -659 finished +------------------------------ ---------------- +epoch -659 +replay_buffer/size 999047 +trainer/num train calls 342000 +trainer/QF1 Loss 9.3536 +trainer/QF2 Loss 9.33208 +trainer/Policy Loss 43.4083 +trainer/Q1 Predictions Mean -71.9866 +trainer/Q1 Predictions Std 17.9085 +trainer/Q1 Predictions Max -0.627703 +trainer/Q1 Predictions Min -86.6912 +trainer/Q2 Predictions Mean -72.036 +trainer/Q2 Predictions Std 17.9095 +trainer/Q2 Predictions Max -0.556333 +trainer/Q2 Predictions Min -86.4734 +trainer/Q Targets Mean -72.4917 +trainer/Q Targets Std 17.5708 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9386 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0149223 +trainer/policy/mean Std 0.713493 +trainer/policy/mean Max 0.999172 +trainer/policy/mean Min -0.996341 +trainer/policy/std Mean 0.427491 +trainer/policy/std Std 0.0204808 +trainer/policy/std Max 0.450143 +trainer/policy/std Min 0.392536 +trainer/Advantage Weights Mean 9.13144 +trainer/Advantage Weights Std 25.178 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.54193e-16 +trainer/Advantage Score Mean -0.253116 +trainer/Advantage Score Std 0.568184 +trainer/Advantage Score Max 1.48179 +trainer/Advantage Score Min -3.5328 +trainer/V1 Predictions Mean -72.037 +trainer/V1 Predictions Std 17.9299 +trainer/V1 Predictions Max -0.351843 +trainer/V1 Predictions Min -86.7847 +trainer/VF Loss 0.0781612 +expl/num steps total 342000 +expl/num paths total 393 +expl/path length Mean 500 +expl/path length Std 191 +expl/path length Max 691 +expl/path length Min 309 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0323743 +expl/Actions Std 0.828533 +expl/Actions Max 2.40967 +expl/Actions Min -2.33573 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 327031 +eval/num paths total 342 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0154182 +eval/Actions Std 0.637568 +eval/Actions Max 0.997548 +eval/Actions Min -0.994364 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.50991e-06 +time/evaluation sampling (s) 2.3171 +time/exploration sampling (s) 2.5861 +time/logging (s) 0.00684663 +time/saving (s) 0.0092093 +time/training (s) 11.8398 +time/epoch (s) 16.759 +time/total (s) 7131.43 +Epoch -659 +------------------------------ ---------------- +2022-05-15 20:01:34.426543 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -658 finished +------------------------------ ---------------- +epoch -658 +replay_buffer/size 999047 +trainer/num train calls 343000 +trainer/QF1 Loss 1.16178 +trainer/QF2 Loss 1.14652 +trainer/Policy Loss 25.5245 +trainer/Q1 Predictions Mean -71.7432 +trainer/Q1 Predictions Std 17.8256 +trainer/Q1 Predictions Max -2.14984 +trainer/Q1 Predictions Min -87.0715 +trainer/Q2 Predictions Mean -71.8128 +trainer/Q2 Predictions Std 17.7697 +trainer/Q2 Predictions Max -2.76539 +trainer/Q2 Predictions Min -87.1427 +trainer/Q Targets Mean -71.9636 +trainer/Q Targets Std 17.7334 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0616 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00179715 +trainer/policy/mean Std 0.701476 +trainer/policy/mean Max 0.999424 +trainer/policy/mean Min -0.998992 +trainer/policy/std Mean 0.427641 +trainer/policy/std Std 0.0205178 +trainer/policy/std Max 0.450514 +trainer/policy/std Min 0.393793 +trainer/Advantage Weights Mean 5.5714 +trainer/Advantage Weights Std 19.8533 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.29472e-09 +trainer/Advantage Score Mean -0.21537 +trainer/Advantage Score Std 0.461948 +trainer/Advantage Score Max 2.02427 +trainer/Advantage Score Min -2.0465 +trainer/V1 Predictions Mean -71.735 +trainer/V1 Predictions Std 17.5957 +trainer/V1 Predictions Max -3.90471 +trainer/V1 Predictions Min -86.7506 +trainer/VF Loss 0.074369 +expl/num steps total 343000 +expl/num paths total 394 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0314538 +expl/Actions Std 0.858151 +expl/Actions Max 2.5866 +expl/Actions Min -2.36846 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 327614 +eval/num paths total 343 +eval/path length Mean 583 +eval/path length Std 0 +eval/path length Max 583 +eval/path length Min 583 +eval/Rewards Mean 0.00171527 +eval/Rewards Std 0.0413802 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0455594 +eval/Actions Std 0.732736 +eval/Actions Max 0.999818 +eval/Actions Min -0.999772 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.76975e-06 +time/evaluation sampling (s) 2.28648 +time/exploration sampling (s) 2.81618 +time/logging (s) 0.00570087 +time/saving (s) 0.00933711 +time/training (s) 11.3778 +time/epoch (s) 16.4955 +time/total (s) 7147.93 +Epoch -658 +------------------------------ ---------------- +2022-05-15 20:01:51.084484 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -657 finished +------------------------------ --------------- +epoch -657 +replay_buffer/size 999047 +trainer/num train calls 344000 +trainer/QF1 Loss 0.629462 +trainer/QF2 Loss 0.704537 +trainer/Policy Loss 40.0723 +trainer/Q1 Predictions Mean -70.8834 +trainer/Q1 Predictions Std 19.483 +trainer/Q1 Predictions Max -3.13953 +trainer/Q1 Predictions Min -85.7835 +trainer/Q2 Predictions Mean -70.9215 +trainer/Q2 Predictions Std 19.4787 +trainer/Q2 Predictions Max -2.7372 +trainer/Q2 Predictions Min -86.0105 +trainer/Q Targets Mean -70.8516 +trainer/Q Targets Std 19.6846 +trainer/Q Targets Max -1.80341 +trainer/Q Targets Min -86.4106 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0131688 +trainer/policy/mean Std 0.711769 +trainer/policy/mean Max 0.99897 +trainer/policy/mean Min -0.998831 +trainer/policy/std Mean 0.426237 +trainer/policy/std Std 0.0219643 +trainer/policy/std Max 0.449163 +trainer/policy/std Min 0.389006 +trainer/Advantage Weights Mean 7.29152 +trainer/Advantage Weights Std 21.6575 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.8774e-19 +trainer/Advantage Score Mean -0.328639 +trainer/Advantage Score Std 0.699185 +trainer/Advantage Score Max 1.04208 +trainer/Advantage Score Min -4.21645 +trainer/V1 Predictions Mean -70.471 +trainer/V1 Predictions Std 19.9712 +trainer/V1 Predictions Max -1.33427 +trainer/V1 Predictions Min -86.0637 +trainer/VF Loss 0.0823067 +expl/num steps total 344000 +expl/num paths total 395 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0226792 +expl/Actions Std 0.80862 +expl/Actions Max 2.30704 +expl/Actions Min -2.23145 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 328428 +eval/num paths total 344 +eval/path length Mean 814 +eval/path length Std 0 +eval/path length Max 814 +eval/path length Min 814 +eval/Rewards Mean 0.0012285 +eval/Rewards Std 0.0350284 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0264516 +eval/Actions Std 0.730724 +eval/Actions Max 0.999601 +eval/Actions Min -0.998977 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.6268e-06 +time/evaluation sampling (s) 2.26722 +time/exploration sampling (s) 2.6671 +time/logging (s) 0.00624868 +time/saving (s) 0.00925002 +time/training (s) 11.7023 +time/epoch (s) 16.6521 +time/total (s) 7164.59 +Epoch -657 +------------------------------ --------------- +2022-05-15 20:02:07.157600 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -656 finished +------------------------------ ---------------- +epoch -656 +replay_buffer/size 999047 +trainer/num train calls 345000 +trainer/QF1 Loss 1.31756 +trainer/QF2 Loss 1.3228 +trainer/Policy Loss 20.4792 +trainer/Q1 Predictions Mean -71.9326 +trainer/Q1 Predictions Std 18.7117 +trainer/Q1 Predictions Max -0.368592 +trainer/Q1 Predictions Min -86.5995 +trainer/Q2 Predictions Mean -71.9515 +trainer/Q2 Predictions Std 18.7054 +trainer/Q2 Predictions Max -0.576096 +trainer/Q2 Predictions Min -86.3431 +trainer/Q Targets Mean -72.0266 +trainer/Q Targets Std 18.9518 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6193 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0320689 +trainer/policy/mean Std 0.720609 +trainer/policy/mean Max 0.998112 +trainer/policy/mean Min -0.998571 +trainer/policy/std Mean 0.423779 +trainer/policy/std Std 0.0195011 +trainer/policy/std Max 0.443821 +trainer/policy/std Min 0.39115 +trainer/Advantage Weights Mean 4.686 +trainer/Advantage Weights Std 17.6229 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.18929e-14 +trainer/Advantage Score Mean -0.313497 +trainer/Advantage Score Std 0.559406 +trainer/Advantage Score Max 2.02719 +trainer/Advantage Score Min -3.10764 +trainer/V1 Predictions Mean -71.8444 +trainer/V1 Predictions Std 18.8033 +trainer/V1 Predictions Max 0.0883524 +trainer/V1 Predictions Min -86.4837 +trainer/VF Loss 0.0695687 +expl/num steps total 345000 +expl/num paths total 397 +expl/path length Mean 500 +expl/path length Std 48 +expl/path length Max 548 +expl/path length Min 452 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00784674 +expl/Actions Std 0.821456 +expl/Actions Max 2.88856 +expl/Actions Min -2.53238 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 329428 +eval/num paths total 345 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.168014 +eval/Actions Std 0.717449 +eval/Actions Max 0.999463 +eval/Actions Min -0.997246 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.58908e-06 +time/evaluation sampling (s) 2.35257 +time/exploration sampling (s) 2.60976 +time/logging (s) 0.00702271 +time/saving (s) 0.0100149 +time/training (s) 11.0878 +time/epoch (s) 16.0672 +time/total (s) 7180.66 +Epoch -656 +------------------------------ ---------------- +2022-05-15 20:02:23.999606 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -655 finished +------------------------------ ---------------- +epoch -655 +replay_buffer/size 999047 +trainer/num train calls 346000 +trainer/QF1 Loss 0.859551 +trainer/QF2 Loss 0.766862 +trainer/Policy Loss 17.5718 +trainer/Q1 Predictions Mean -72.8377 +trainer/Q1 Predictions Std 18.5388 +trainer/Q1 Predictions Max -1.03694 +trainer/Q1 Predictions Min -85.6585 +trainer/Q2 Predictions Mean -72.8731 +trainer/Q2 Predictions Std 18.5139 +trainer/Q2 Predictions Max -0.336626 +trainer/Q2 Predictions Min -85.6936 +trainer/Q Targets Mean -72.8654 +trainer/Q Targets Std 18.6703 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.5044 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0480882 +trainer/policy/mean Std 0.713369 +trainer/policy/mean Max 0.99893 +trainer/policy/mean Min -0.997305 +trainer/policy/std Mean 0.425768 +trainer/policy/std Std 0.020926 +trainer/policy/std Max 0.44633 +trainer/policy/std Min 0.391292 +trainer/Advantage Weights Mean 4.13198 +trainer/Advantage Weights Std 18.0662 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.9721e-16 +trainer/Advantage Score Mean -0.41217 +trainer/Advantage Score Std 0.5307 +trainer/Advantage Score Max 1.65173 +trainer/Advantage Score Min -3.61623 +trainer/V1 Predictions Mean -72.7108 +trainer/V1 Predictions Std 18.5007 +trainer/V1 Predictions Max -0.668685 +trainer/V1 Predictions Min -85.5216 +trainer/VF Loss 0.067263 +expl/num steps total 346000 +expl/num paths total 399 +expl/path length Mean 500 +expl/path length Std 254 +expl/path length Max 754 +expl/path length Min 246 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0268275 +expl/Actions Std 0.832849 +expl/Actions Max 2.54712 +expl/Actions Min -2.30242 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 330197 +eval/num paths total 346 +eval/path length Mean 769 +eval/path length Std 0 +eval/path length Max 769 +eval/path length Min 769 +eval/Rewards Mean 0.00130039 +eval/Rewards Std 0.0360375 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0282178 +eval/Actions Std 0.735837 +eval/Actions Max 0.999352 +eval/Actions Min -0.998657 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.74507e-06 +time/evaluation sampling (s) 2.2609 +time/exploration sampling (s) 2.6416 +time/logging (s) 0.00633816 +time/saving (s) 0.00949204 +time/training (s) 11.9157 +time/epoch (s) 16.8341 +time/total (s) 7197.49 +Epoch -655 +------------------------------ ---------------- +2022-05-15 20:02:40.676313 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -654 finished +------------------------------ ---------------- +epoch -654 +replay_buffer/size 999047 +trainer/num train calls 347000 +trainer/QF1 Loss 1.18578 +trainer/QF2 Loss 1.08132 +trainer/Policy Loss 19.6384 +trainer/Q1 Predictions Mean -70.2824 +trainer/Q1 Predictions Std 19.7328 +trainer/Q1 Predictions Max -1.35665 +trainer/Q1 Predictions Min -86.8506 +trainer/Q2 Predictions Mean -70.3665 +trainer/Q2 Predictions Std 19.7092 +trainer/Q2 Predictions Max -1.36346 +trainer/Q2 Predictions Min -86.3565 +trainer/Q Targets Mean -70.69 +trainer/Q Targets Std 19.3652 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2654 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0153593 +trainer/policy/mean Std 0.714216 +trainer/policy/mean Max 0.997647 +trainer/policy/mean Min -0.997556 +trainer/policy/std Mean 0.425571 +trainer/policy/std Std 0.0211279 +trainer/policy/std Max 0.447508 +trainer/policy/std Min 0.392125 +trainer/Advantage Weights Mean 4.24036 +trainer/Advantage Weights Std 17.0688 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.47992e-11 +trainer/Advantage Score Mean -0.309735 +trainer/Advantage Score Std 0.492424 +trainer/Advantage Score Max 1.57337 +trainer/Advantage Score Min -2.49364 +trainer/V1 Predictions Mean -70.3548 +trainer/V1 Predictions Std 19.5708 +trainer/V1 Predictions Max -1.97387 +trainer/V1 Predictions Min -87.1148 +trainer/VF Loss 0.0576066 +expl/num steps total 347000 +expl/num paths total 400 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.184756 +expl/Actions Std 0.904016 +expl/Actions Max 2.48932 +expl/Actions Min -2.50271 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 331197 +eval/num paths total 347 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0424293 +eval/Actions Std 0.750701 +eval/Actions Max 0.999158 +eval/Actions Min -0.998417 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.841e-06 +time/evaluation sampling (s) 2.28614 +time/exploration sampling (s) 2.71834 +time/logging (s) 0.00669335 +time/saving (s) 0.00956498 +time/training (s) 11.6493 +time/epoch (s) 16.6701 +time/total (s) 7214.17 +Epoch -654 +------------------------------ ---------------- +2022-05-15 20:02:57.478095 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -653 finished +------------------------------ ---------------- +epoch -653 +replay_buffer/size 999047 +trainer/num train calls 348000 +trainer/QF1 Loss 1.73062 +trainer/QF2 Loss 1.81015 +trainer/Policy Loss 21.0724 +trainer/Q1 Predictions Mean -71.0477 +trainer/Q1 Predictions Std 18.187 +trainer/Q1 Predictions Max -0.309397 +trainer/Q1 Predictions Min -85.4832 +trainer/Q2 Predictions Mean -71.0413 +trainer/Q2 Predictions Std 18.1757 +trainer/Q2 Predictions Max -0.0340384 +trainer/Q2 Predictions Min -85.599 +trainer/Q Targets Mean -71.2694 +trainer/Q Targets Std 18.1213 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4704 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0115157 +trainer/policy/mean Std 0.719964 +trainer/policy/mean Max 0.999097 +trainer/policy/mean Min -0.998652 +trainer/policy/std Mean 0.423981 +trainer/policy/std Std 0.0206318 +trainer/policy/std Max 0.44466 +trainer/policy/std Min 0.390231 +trainer/Advantage Weights Mean 6.53324 +trainer/Advantage Weights Std 21.7574 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.43459e-14 +trainer/Advantage Score Mean -0.284627 +trainer/Advantage Score Std 0.547515 +trainer/Advantage Score Max 1.15365 +trainer/Advantage Score Min -3.18753 +trainer/V1 Predictions Mean -70.9365 +trainer/V1 Predictions Std 18.3072 +trainer/V1 Predictions Max 0.51534 +trainer/V1 Predictions Min -85.7528 +trainer/VF Loss 0.0659027 +expl/num steps total 348000 +expl/num paths total 401 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0858102 +expl/Actions Std 0.818241 +expl/Actions Max 2.50273 +expl/Actions Min -2.4756 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 332197 +eval/num paths total 348 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0392026 +eval/Actions Std 0.711131 +eval/Actions Max 0.999514 +eval/Actions Min -0.999813 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.84286e-06 +time/evaluation sampling (s) 2.33427 +time/exploration sampling (s) 2.7326 +time/logging (s) 0.0069278 +time/saving (s) 0.00945604 +time/training (s) 11.7115 +time/epoch (s) 16.7947 +time/total (s) 7230.97 +Epoch -653 +------------------------------ ---------------- +2022-05-15 20:03:14.494906 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -652 finished +------------------------------ ---------------- +epoch -652 +replay_buffer/size 999047 +trainer/num train calls 349000 +trainer/QF1 Loss 3.82982 +trainer/QF2 Loss 3.70255 +trainer/Policy Loss 13.5236 +trainer/Q1 Predictions Mean -73.0863 +trainer/Q1 Predictions Std 18.3453 +trainer/Q1 Predictions Max -0.252526 +trainer/Q1 Predictions Min -85.9959 +trainer/Q2 Predictions Mean -73.0835 +trainer/Q2 Predictions Std 18.3265 +trainer/Q2 Predictions Max -0.360676 +trainer/Q2 Predictions Min -86.2045 +trainer/Q Targets Mean -72.711 +trainer/Q Targets Std 18.3776 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0015 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0359756 +trainer/policy/mean Std 0.711734 +trainer/policy/mean Max 0.99907 +trainer/policy/mean Min -0.998804 +trainer/policy/std Mean 0.426973 +trainer/policy/std Std 0.0215445 +trainer/policy/std Max 0.448769 +trainer/policy/std Min 0.390936 +trainer/Advantage Weights Mean 3.34063 +trainer/Advantage Weights Std 15.161 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.10427e-09 +trainer/Advantage Score Mean -0.382641 +trainer/Advantage Score Std 0.445876 +trainer/Advantage Score Max 2.08784 +trainer/Advantage Score Min -2.06241 +trainer/V1 Predictions Mean -72.5542 +trainer/V1 Predictions Std 18.3484 +trainer/V1 Predictions Max -0.140303 +trainer/V1 Predictions Min -85.5381 +trainer/VF Loss 0.0603782 +expl/num steps total 349000 +expl/num paths total 402 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.068279 +expl/Actions Std 0.832951 +expl/Actions Max 2.47886 +expl/Actions Min -2.14629 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 332982 +eval/num paths total 349 +eval/path length Mean 785 +eval/path length Std 0 +eval/path length Max 785 +eval/path length Min 785 +eval/Rewards Mean 0.00127389 +eval/Rewards Std 0.0356688 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0705743 +eval/Actions Std 0.733258 +eval/Actions Max 0.999599 +eval/Actions Min -0.998769 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.17534e-06 +time/evaluation sampling (s) 2.25832 +time/exploration sampling (s) 2.65551 +time/logging (s) 0.00678371 +time/saving (s) 0.00993539 +time/training (s) 12.0791 +time/epoch (s) 17.0097 +time/total (s) 7247.98 +Epoch -652 +------------------------------ ---------------- +2022-05-15 20:03:31.091742 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -651 finished +------------------------------ ---------------- +epoch -651 +replay_buffer/size 999047 +trainer/num train calls 350000 +trainer/QF1 Loss 0.695439 +trainer/QF2 Loss 0.7417 +trainer/Policy Loss 7.94565 +trainer/Q1 Predictions Mean -73.0584 +trainer/Q1 Predictions Std 16.8436 +trainer/Q1 Predictions Max -7.06728 +trainer/Q1 Predictions Min -85.8711 +trainer/Q2 Predictions Mean -73.1365 +trainer/Q2 Predictions Std 16.8933 +trainer/Q2 Predictions Max -6.11834 +trainer/Q2 Predictions Min -86.0697 +trainer/Q Targets Mean -73.0049 +trainer/Q Targets Std 16.7851 +trainer/Q Targets Max -7.29213 +trainer/Q Targets Min -85.898 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000955083 +trainer/policy/mean Std 0.720686 +trainer/policy/mean Max 0.998482 +trainer/policy/mean Min -0.999226 +trainer/policy/std Mean 0.424785 +trainer/policy/std Std 0.020416 +trainer/policy/std Max 0.447139 +trainer/policy/std Min 0.389889 +trainer/Advantage Weights Mean 2.14564 +trainer/Advantage Weights Std 11.4872 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.13949e-17 +trainer/Advantage Score Mean -0.378742 +trainer/Advantage Score Std 0.468838 +trainer/Advantage Score Max 1.29993 +trainer/Advantage Score Min -3.90134 +trainer/V1 Predictions Mean -72.7448 +trainer/V1 Predictions Std 16.8443 +trainer/V1 Predictions Max -6.66823 +trainer/V1 Predictions Min -85.8006 +trainer/VF Loss 0.0462992 +expl/num steps total 350000 +expl/num paths total 404 +expl/path length Mean 500 +expl/path length Std 233 +expl/path length Max 733 +expl/path length Min 267 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0264218 +expl/Actions Std 0.839406 +expl/Actions Max 2.42231 +expl/Actions Min -2.49796 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 333982 +eval/num paths total 350 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0221607 +eval/Actions Std 0.72707 +eval/Actions Max 0.999101 +eval/Actions Min -0.999046 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.55182e-06 +time/evaluation sampling (s) 2.24718 +time/exploration sampling (s) 2.74976 +time/logging (s) 0.00669862 +time/saving (s) 0.00897329 +time/training (s) 11.5766 +time/epoch (s) 16.5893 +time/total (s) 7264.57 +Epoch -651 +------------------------------ ---------------- +2022-05-15 20:03:47.935298 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -650 finished +------------------------------ ---------------- +epoch -650 +replay_buffer/size 999047 +trainer/num train calls 351000 +trainer/QF1 Loss 1.10836 +trainer/QF2 Loss 0.984502 +trainer/Policy Loss 17.4555 +trainer/Q1 Predictions Mean -71.5724 +trainer/Q1 Predictions Std 19.1461 +trainer/Q1 Predictions Max -0.25196 +trainer/Q1 Predictions Min -86.804 +trainer/Q2 Predictions Mean -71.5013 +trainer/Q2 Predictions Std 19.2161 +trainer/Q2 Predictions Max -0.350699 +trainer/Q2 Predictions Min -86.3157 +trainer/Q Targets Mean -71.1042 +trainer/Q Targets Std 18.8585 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4394 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00970594 +trainer/policy/mean Std 0.710132 +trainer/policy/mean Max 0.998794 +trainer/policy/mean Min -0.998003 +trainer/policy/std Mean 0.424658 +trainer/policy/std Std 0.0211803 +trainer/policy/std Max 0.447324 +trainer/policy/std Min 0.389374 +trainer/Advantage Weights Mean 4.35931 +trainer/Advantage Weights Std 17.9933 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.73603e-10 +trainer/Advantage Score Mean -0.534183 +trainer/Advantage Score Std 0.522779 +trainer/Advantage Score Max 1.13214 +trainer/Advantage Score Min -2.24742 +trainer/V1 Predictions Mean -70.8239 +trainer/V1 Predictions Std 18.9406 +trainer/V1 Predictions Max -0.331736 +trainer/V1 Predictions Min -86.3382 +trainer/VF Loss 0.074189 +expl/num steps total 351000 +expl/num paths total 405 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0484842 +expl/Actions Std 0.831992 +expl/Actions Max 2.48149 +expl/Actions Min -2.57633 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 334982 +eval/num paths total 351 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0528324 +eval/Actions Std 0.697078 +eval/Actions Max 0.999518 +eval/Actions Min -0.999481 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.48011e-06 +time/evaluation sampling (s) 2.27452 +time/exploration sampling (s) 2.78431 +time/logging (s) 0.00713103 +time/saving (s) 0.0193477 +time/training (s) 11.7519 +time/epoch (s) 16.8372 +time/total (s) 7281.41 +Epoch -650 +------------------------------ ---------------- +2022-05-15 20:04:04.606054 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -649 finished +------------------------------ ---------------- +epoch -649 +replay_buffer/size 999047 +trainer/num train calls 352000 +trainer/QF1 Loss 0.504216 +trainer/QF2 Loss 0.585701 +trainer/Policy Loss 24.7681 +trainer/Q1 Predictions Mean -73.0366 +trainer/Q1 Predictions Std 17.5185 +trainer/Q1 Predictions Max -0.375896 +trainer/Q1 Predictions Min -85.427 +trainer/Q2 Predictions Mean -73.0623 +trainer/Q2 Predictions Std 17.5736 +trainer/Q2 Predictions Max -0.401813 +trainer/Q2 Predictions Min -85.9164 +trainer/Q Targets Mean -73.1375 +trainer/Q Targets Std 17.4881 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.8658 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0064367 +trainer/policy/mean Std 0.702119 +trainer/policy/mean Max 0.999684 +trainer/policy/mean Min -0.997003 +trainer/policy/std Mean 0.425084 +trainer/policy/std Std 0.0199419 +trainer/policy/std Max 0.448139 +trainer/policy/std Min 0.389738 +trainer/Advantage Weights Mean 4.10995 +trainer/Advantage Weights Std 17.5973 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.21413e-11 +trainer/Advantage Score Mean -0.297213 +trainer/Advantage Score Std 0.498462 +trainer/Advantage Score Max 2.38858 +trainer/Advantage Score Min -2.36771 +trainer/V1 Predictions Mean -72.9414 +trainer/V1 Predictions Std 17.6226 +trainer/V1 Predictions Max 1.57393 +trainer/V1 Predictions Min -85.7554 +trainer/VF Loss 0.0755744 +expl/num steps total 352000 +expl/num paths total 407 +expl/path length Mean 500 +expl/path length Std 174 +expl/path length Max 674 +expl/path length Min 326 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0354186 +expl/Actions Std 0.818731 +expl/Actions Max 2.23569 +expl/Actions Min -2.30737 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 335737 +eval/num paths total 352 +eval/path length Mean 755 +eval/path length Std 0 +eval/path length Max 755 +eval/path length Min 755 +eval/Rewards Mean 0.0013245 +eval/Rewards Std 0.0363696 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0488777 +eval/Actions Std 0.731222 +eval/Actions Max 0.999249 +eval/Actions Min -0.999438 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.64263e-06 +time/evaluation sampling (s) 2.46629 +time/exploration sampling (s) 2.9004 +time/logging (s) 0.00649042 +time/saving (s) 0.00951075 +time/training (s) 11.2802 +time/epoch (s) 16.6629 +time/total (s) 7298.08 +Epoch -649 +------------------------------ ---------------- +2022-05-15 20:04:20.116124 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -648 finished +------------------------------ ---------------- +epoch -648 +replay_buffer/size 999047 +trainer/num train calls 353000 +trainer/QF1 Loss 0.791843 +trainer/QF2 Loss 1.02368 +trainer/Policy Loss 17.8267 +trainer/Q1 Predictions Mean -73.8509 +trainer/Q1 Predictions Std 15.8812 +trainer/Q1 Predictions Max -3.27006 +trainer/Q1 Predictions Min -85.1505 +trainer/Q2 Predictions Mean -73.7041 +trainer/Q2 Predictions Std 16.0025 +trainer/Q2 Predictions Max -3.41773 +trainer/Q2 Predictions Min -85.3721 +trainer/Q Targets Mean -73.9289 +trainer/Q Targets Std 15.6278 +trainer/Q Targets Max -4.8807 +trainer/Q Targets Min -85.4654 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0288864 +trainer/policy/mean Std 0.719273 +trainer/policy/mean Max 0.999568 +trainer/policy/mean Min -0.999391 +trainer/policy/std Mean 0.425638 +trainer/policy/std Std 0.0193758 +trainer/policy/std Max 0.448392 +trainer/policy/std Min 0.392481 +trainer/Advantage Weights Mean 3.39335 +trainer/Advantage Weights Std 14.9591 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.75599e-15 +trainer/Advantage Score Mean -0.316674 +trainer/Advantage Score Std 0.461158 +trainer/Advantage Score Max 0.822228 +trainer/Advantage Score Min -3.22609 +trainer/V1 Predictions Mean -73.6729 +trainer/V1 Predictions Std 15.9588 +trainer/V1 Predictions Max -2.93034 +trainer/V1 Predictions Min -85.3258 +trainer/VF Loss 0.0416609 +expl/num steps total 353000 +expl/num paths total 409 +expl/path length Mean 500 +expl/path length Std 25 +expl/path length Max 525 +expl/path length Min 475 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00220759 +expl/Actions Std 0.834547 +expl/Actions Max 2.20709 +expl/Actions Min -2.41045 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 336737 +eval/num paths total 353 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0174856 +eval/Actions Std 0.730738 +eval/Actions Max 0.999876 +eval/Actions Min -0.999144 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.7488e-06 +time/evaluation sampling (s) 2.28728 +time/exploration sampling (s) 2.78251 +time/logging (s) 0.00731347 +time/saving (s) 0.00982741 +time/training (s) 10.4167 +time/epoch (s) 15.5036 +time/total (s) 7313.59 +Epoch -648 +------------------------------ ---------------- +2022-05-15 20:04:37.198629 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -647 finished +------------------------------ ---------------- +epoch -647 +replay_buffer/size 999047 +trainer/num train calls 354000 +trainer/QF1 Loss 0.745506 +trainer/QF2 Loss 0.811299 +trainer/Policy Loss 17.0108 +trainer/Q1 Predictions Mean -71.8191 +trainer/Q1 Predictions Std 16.958 +trainer/Q1 Predictions Max -1.53047 +trainer/Q1 Predictions Min -85.5662 +trainer/Q2 Predictions Mean -71.9825 +trainer/Q2 Predictions Std 16.8624 +trainer/Q2 Predictions Max -1.51491 +trainer/Q2 Predictions Min -85.823 +trainer/Q Targets Mean -72.0535 +trainer/Q Targets Std 17.2067 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1218 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0202047 +trainer/policy/mean Std 0.712651 +trainer/policy/mean Max 0.999366 +trainer/policy/mean Min -0.997962 +trainer/policy/std Mean 0.425504 +trainer/policy/std Std 0.0202851 +trainer/policy/std Max 0.448402 +trainer/policy/std Min 0.390464 +trainer/Advantage Weights Mean 3.18241 +trainer/Advantage Weights Std 14.0828 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.54089e-17 +trainer/Advantage Score Mean -0.446062 +trainer/Advantage Score Std 0.695543 +trainer/Advantage Score Max 0.55838 +trainer/Advantage Score Min -3.72659 +trainer/V1 Predictions Mean -71.8091 +trainer/V1 Predictions Std 17.3639 +trainer/V1 Predictions Max -0.224167 +trainer/V1 Predictions Min -86.3639 +trainer/VF Loss 0.0756455 +expl/num steps total 354000 +expl/num paths total 410 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0472151 +expl/Actions Std 0.82342 +expl/Actions Max 2.15103 +expl/Actions Min -2.65942 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 337737 +eval/num paths total 354 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.118136 +eval/Actions Std 0.683268 +eval/Actions Max 0.999277 +eval/Actions Min -0.998973 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.10875e-06 +time/evaluation sampling (s) 2.3351 +time/exploration sampling (s) 2.74067 +time/logging (s) 0.00679854 +time/saving (s) 0.00943312 +time/training (s) 11.9825 +time/epoch (s) 17.0745 +time/total (s) 7330.67 +Epoch -647 +------------------------------ ---------------- +2022-05-15 20:04:58.390031 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -646 finished +------------------------------ ---------------- +epoch -646 +replay_buffer/size 999047 +trainer/num train calls 355000 +trainer/QF1 Loss 1.39582 +trainer/QF2 Loss 1.48019 +trainer/Policy Loss 29.0863 +trainer/Q1 Predictions Mean -73.1983 +trainer/Q1 Predictions Std 17.4088 +trainer/Q1 Predictions Max -1.56174 +trainer/Q1 Predictions Min -87.5184 +trainer/Q2 Predictions Mean -73.1787 +trainer/Q2 Predictions Std 17.4714 +trainer/Q2 Predictions Max -1.63861 +trainer/Q2 Predictions Min -87.3111 +trainer/Q Targets Mean -72.8788 +trainer/Q Targets Std 17.2328 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4177 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00757046 +trainer/policy/mean Std 0.723898 +trainer/policy/mean Max 0.998746 +trainer/policy/mean Min -0.997972 +trainer/policy/std Mean 0.425616 +trainer/policy/std Std 0.0210965 +trainer/policy/std Max 0.447726 +trainer/policy/std Min 0.389834 +trainer/Advantage Weights Mean 7.10745 +trainer/Advantage Weights Std 23.3845 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.07388e-12 +trainer/Advantage Score Mean -0.280543 +trainer/Advantage Score Std 0.474721 +trainer/Advantage Score Max 1.96062 +trainer/Advantage Score Min -2.75597 +trainer/V1 Predictions Mean -72.6019 +trainer/V1 Predictions Std 17.2723 +trainer/V1 Predictions Max -1.357 +trainer/V1 Predictions Min -86.181 +trainer/VF Loss 0.0705875 +expl/num steps total 355000 +expl/num paths total 411 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.010024 +expl/Actions Std 0.848085 +expl/Actions Max 2.32719 +expl/Actions Min -2.5146 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 338737 +eval/num paths total 355 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0775586 +eval/Actions Std 0.818368 +eval/Actions Max 0.998868 +eval/Actions Min -0.998437 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78279e-06 +time/evaluation sampling (s) 2.36248 +time/exploration sampling (s) 2.92993 +time/logging (s) 0.012006 +time/saving (s) 0.0155521 +time/training (s) 15.8697 +time/epoch (s) 21.1897 +time/total (s) 7351.86 +Epoch -646 +------------------------------ ---------------- +2022-05-15 20:05:26.988617 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -645 finished +------------------------------ ---------------- +epoch -645 +replay_buffer/size 999047 +trainer/num train calls 356000 +trainer/QF1 Loss 1.15329 +trainer/QF2 Loss 1.19231 +trainer/Policy Loss 22.04 +trainer/Q1 Predictions Mean -71.6699 +trainer/Q1 Predictions Std 19.4511 +trainer/Q1 Predictions Max -1.067 +trainer/Q1 Predictions Min -86.1576 +trainer/Q2 Predictions Mean -71.7903 +trainer/Q2 Predictions Std 19.4124 +trainer/Q2 Predictions Max -0.701116 +trainer/Q2 Predictions Min -86.1147 +trainer/Q Targets Mean -71.3151 +trainer/Q Targets Std 19.529 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.5497 +trainer/rewards Mean -0.976562 +trainer/rewards Std 0.151288 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0234375 +trainer/terminals Std 0.151288 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0168578 +trainer/policy/mean Std 0.714262 +trainer/policy/mean Max 0.998854 +trainer/policy/mean Min -0.998686 +trainer/policy/std Mean 0.42606 +trainer/policy/std Std 0.0213119 +trainer/policy/std Max 0.447009 +trainer/policy/std Min 0.386562 +trainer/Advantage Weights Mean 3.63485 +trainer/Advantage Weights Std 17.4679 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.70545e-21 +trainer/Advantage Score Mean -0.48598 +trainer/Advantage Score Std 0.643526 +trainer/Advantage Score Max 1.64174 +trainer/Advantage Score Min -4.61903 +trainer/V1 Predictions Mean -71.1455 +trainer/V1 Predictions Std 19.3669 +trainer/V1 Predictions Max -1.65662 +trainer/V1 Predictions Min -85.4035 +trainer/VF Loss 0.092375 +expl/num steps total 356000 +expl/num paths total 413 +expl/path length Mean 500 +expl/path length Std 309 +expl/path length Max 809 +expl/path length Min 191 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0235297 +expl/Actions Std 0.835236 +expl/Actions Max 2.25003 +expl/Actions Min -2.50877 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 339737 +eval/num paths total 356 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.222163 +eval/Actions Std 0.67151 +eval/Actions Max 0.998957 +eval/Actions Min -0.999632 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.93903e-06 +time/evaluation sampling (s) 3.42376 +time/exploration sampling (s) 3.75899 +time/logging (s) 0.0132166 +time/saving (s) 0.0190305 +time/training (s) 21.3707 +time/epoch (s) 28.5857 +time/total (s) 7380.46 +Epoch -645 +------------------------------ ---------------- +2022-05-15 20:05:51.275664 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -644 finished +------------------------------ ---------------- +epoch -644 +replay_buffer/size 999047 +trainer/num train calls 357000 +trainer/QF1 Loss 0.930966 +trainer/QF2 Loss 0.982134 +trainer/Policy Loss 23.3665 +trainer/Q1 Predictions Mean -72.0706 +trainer/Q1 Predictions Std 17.8765 +trainer/Q1 Predictions Max -3.07976 +trainer/Q1 Predictions Min -86.6297 +trainer/Q2 Predictions Mean -72.0541 +trainer/Q2 Predictions Std 17.8176 +trainer/Q2 Predictions Max -3.39973 +trainer/Q2 Predictions Min -86.6529 +trainer/Q Targets Mean -71.94 +trainer/Q Targets Std 18.2649 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2095 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00207503 +trainer/policy/mean Std 0.718466 +trainer/policy/mean Max 0.999462 +trainer/policy/mean Min -0.998889 +trainer/policy/std Mean 0.426225 +trainer/policy/std Std 0.021416 +trainer/policy/std Max 0.448341 +trainer/policy/std Min 0.388951 +trainer/Advantage Weights Mean 7.42365 +trainer/Advantage Weights Std 22.3014 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.22574e-14 +trainer/Advantage Score Mean -0.288725 +trainer/Advantage Score Std 0.605631 +trainer/Advantage Score Max 1.98919 +trainer/Advantage Score Min -3.1065 +trainer/V1 Predictions Mean -71.6488 +trainer/V1 Predictions Std 18.2069 +trainer/V1 Predictions Max -1.60917 +trainer/V1 Predictions Min -85.991 +trainer/VF Loss 0.0818527 +expl/num steps total 357000 +expl/num paths total 414 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0582239 +expl/Actions Std 0.834559 +expl/Actions Max 2.61663 +expl/Actions Min -2.42591 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 340421 +eval/num paths total 357 +eval/path length Mean 684 +eval/path length Std 0 +eval/path length Max 684 +eval/path length Min 684 +eval/Rewards Mean 0.00146199 +eval/Rewards Std 0.038208 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0688491 +eval/Actions Std 0.733753 +eval/Actions Max 0.999347 +eval/Actions Min -0.998568 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.66572e-05 +time/evaluation sampling (s) 4.80137 +time/exploration sampling (s) 4.38811 +time/logging (s) 0.0111458 +time/saving (s) 0.0145975 +time/training (s) 15.0428 +time/epoch (s) 24.2581 +time/total (s) 7404.73 +Epoch -644 +------------------------------ ---------------- +2022-05-15 20:06:10.035921 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -643 finished +------------------------------ ---------------- +epoch -643 +replay_buffer/size 999047 +trainer/num train calls 358000 +trainer/QF1 Loss 2.53408 +trainer/QF2 Loss 2.59135 +trainer/Policy Loss 3.58744 +trainer/Q1 Predictions Mean -71.6926 +trainer/Q1 Predictions Std 18.821 +trainer/Q1 Predictions Max -1.01896 +trainer/Q1 Predictions Min -85.6586 +trainer/Q2 Predictions Mean -71.6698 +trainer/Q2 Predictions Std 18.8719 +trainer/Q2 Predictions Max -1.31469 +trainer/Q2 Predictions Min -85.7878 +trainer/Q Targets Mean -71.2691 +trainer/Q Targets Std 18.8836 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.0059 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0330228 +trainer/policy/mean Std 0.719051 +trainer/policy/mean Max 0.998808 +trainer/policy/mean Min -0.999497 +trainer/policy/std Mean 0.426429 +trainer/policy/std Std 0.0216697 +trainer/policy/std Max 0.449926 +trainer/policy/std Min 0.389553 +trainer/Advantage Weights Mean 1.30102 +trainer/Advantage Weights Std 10.7822 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.35911e-20 +trainer/Advantage Score Mean -0.706332 +trainer/Advantage Score Std 0.570414 +trainer/Advantage Score Max 1.10119 +trainer/Advantage Score Min -4.42018 +trainer/V1 Predictions Mean -71.037 +trainer/V1 Predictions Std 18.881 +trainer/V1 Predictions Max -1.18942 +trainer/V1 Predictions Min -84.997 +trainer/VF Loss 0.0898942 +expl/num steps total 358000 +expl/num paths total 415 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0109502 +expl/Actions Std 0.829138 +expl/Actions Max 2.48363 +expl/Actions Min -2.3581 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 341421 +eval/num paths total 358 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0406895 +eval/Actions Std 0.728442 +eval/Actions Max 0.999532 +eval/Actions Min -0.999178 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.05556e-05 +time/evaluation sampling (s) 2.34639 +time/exploration sampling (s) 2.99908 +time/logging (s) 0.00898308 +time/saving (s) 0.0125199 +time/training (s) 13.3759 +time/epoch (s) 18.7429 +time/total (s) 7423.48 +Epoch -643 +------------------------------ ---------------- +2022-05-15 20:06:29.895699 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -642 finished +------------------------------ ---------------- +epoch -642 +replay_buffer/size 999047 +trainer/num train calls 359000 +trainer/QF1 Loss 0.577007 +trainer/QF2 Loss 0.688019 +trainer/Policy Loss 47.8379 +trainer/Q1 Predictions Mean -74.8181 +trainer/Q1 Predictions Std 12.9808 +trainer/Q1 Predictions Max -2.28553 +trainer/Q1 Predictions Min -86.2215 +trainer/Q2 Predictions Mean -74.7378 +trainer/Q2 Predictions Std 12.9572 +trainer/Q2 Predictions Max -2.11435 +trainer/Q2 Predictions Min -85.9557 +trainer/Q Targets Mean -75.2963 +trainer/Q Targets Std 12.9091 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2873 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0059508 +trainer/policy/mean Std 0.719256 +trainer/policy/mean Max 0.999377 +trainer/policy/mean Min -0.996476 +trainer/policy/std Mean 0.425904 +trainer/policy/std Std 0.0218811 +trainer/policy/std Max 0.450778 +trainer/policy/std Min 0.388511 +trainer/Advantage Weights Mean 9.19949 +trainer/Advantage Weights Std 24.1476 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.00794e-11 +trainer/Advantage Score Mean -0.0916119 +trainer/Advantage Score Std 0.447811 +trainer/Advantage Score Max 1.99691 +trainer/Advantage Score Min -2.42272 +trainer/V1 Predictions Mean -75.0829 +trainer/V1 Predictions Std 12.9632 +trainer/V1 Predictions Max -1.23212 +trainer/V1 Predictions Min -86.1696 +trainer/VF Loss 0.073358 +expl/num steps total 359000 +expl/num paths total 416 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0393195 +expl/Actions Std 0.833482 +expl/Actions Max 2.22378 +expl/Actions Min -2.37825 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 342421 +eval/num paths total 359 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.109843 +eval/Actions Std 0.642839 +eval/Actions Max 0.999717 +eval/Actions Min -0.999041 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.63005e-06 +time/evaluation sampling (s) 2.69787 +time/exploration sampling (s) 2.90776 +time/logging (s) 0.011657 +time/saving (s) 0.0154954 +time/training (s) 14.2206 +time/epoch (s) 19.8534 +time/total (s) 7443.34 +Epoch -642 +------------------------------ ---------------- +2022-05-15 20:06:48.920431 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -641 finished +------------------------------ ---------------- +epoch -641 +replay_buffer/size 999047 +trainer/num train calls 360000 +trainer/QF1 Loss 1.03982 +trainer/QF2 Loss 1.05998 +trainer/Policy Loss 10.7991 +trainer/Q1 Predictions Mean -71.5854 +trainer/Q1 Predictions Std 19.0332 +trainer/Q1 Predictions Max -0.933698 +trainer/Q1 Predictions Min -86.257 +trainer/Q2 Predictions Mean -71.5823 +trainer/Q2 Predictions Std 19.1096 +trainer/Q2 Predictions Max -0.749018 +trainer/Q2 Predictions Min -86.42 +trainer/Q Targets Mean -71.4075 +trainer/Q Targets Std 19.168 +trainer/Q Targets Max -1.02153 +trainer/Q Targets Min -86.4734 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0315681 +trainer/policy/mean Std 0.724427 +trainer/policy/mean Max 0.99965 +trainer/policy/mean Min -0.99677 +trainer/policy/std Mean 0.425016 +trainer/policy/std Std 0.020453 +trainer/policy/std Max 0.445733 +trainer/policy/std Min 0.387153 +trainer/Advantage Weights Mean 2.50464 +trainer/Advantage Weights Std 11.5592 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.14556e-15 +trainer/Advantage Score Mean -0.38951 +trainer/Advantage Score Std 0.601902 +trainer/Advantage Score Max 2.77964 +trainer/Advantage Score Min -3.44029 +trainer/V1 Predictions Mean -71.168 +trainer/V1 Predictions Std 19.2624 +trainer/V1 Predictions Max -0.193519 +trainer/V1 Predictions Min -86.0307 +trainer/VF Loss 0.080561 +expl/num steps total 360000 +expl/num paths total 417 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.299615 +expl/Actions Std 0.79685 +expl/Actions Max 2.4729 +expl/Actions Min -2.34031 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 343421 +eval/num paths total 360 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.098844 +eval/Actions Std 0.718739 +eval/Actions Max 0.99903 +eval/Actions Min -0.999383 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.63983e-06 +time/evaluation sampling (s) 2.58361 +time/exploration sampling (s) 2.94292 +time/logging (s) 0.00855134 +time/saving (s) 0.0127117 +time/training (s) 13.4608 +time/epoch (s) 19.0086 +time/total (s) 7462.36 +Epoch -641 +------------------------------ ---------------- +2022-05-15 20:07:07.604990 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -640 finished +------------------------------ ---------------- +epoch -640 +replay_buffer/size 999047 +trainer/num train calls 361000 +trainer/QF1 Loss 0.820451 +trainer/QF2 Loss 0.876409 +trainer/Policy Loss 12.42 +trainer/Q1 Predictions Mean -73.6509 +trainer/Q1 Predictions Std 15.311 +trainer/Q1 Predictions Max -1.26398 +trainer/Q1 Predictions Min -86.3944 +trainer/Q2 Predictions Mean -73.6095 +trainer/Q2 Predictions Std 15.2931 +trainer/Q2 Predictions Max -1.45102 +trainer/Q2 Predictions Min -86.4939 +trainer/Q Targets Mean -73.5771 +trainer/Q Targets Std 15.5807 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7736 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0258697 +trainer/policy/mean Std 0.724326 +trainer/policy/mean Max 0.999641 +trainer/policy/mean Min -0.998654 +trainer/policy/std Mean 0.425305 +trainer/policy/std Std 0.0206942 +trainer/policy/std Max 0.448414 +trainer/policy/std Min 0.389698 +trainer/Advantage Weights Mean 3.10634 +trainer/Advantage Weights Std 15.6322 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.25206e-12 +trainer/Advantage Score Mean -0.449019 +trainer/Advantage Score Std 0.526703 +trainer/Advantage Score Max 1.69769 +trainer/Advantage Score Min -2.74062 +trainer/V1 Predictions Mean -73.3845 +trainer/V1 Predictions Std 15.4691 +trainer/V1 Predictions Max -1.01407 +trainer/V1 Predictions Min -86.5912 +trainer/VF Loss 0.0750466 +expl/num steps total 361000 +expl/num paths total 418 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0149979 +expl/Actions Std 0.826572 +expl/Actions Max 2.2989 +expl/Actions Min -2.23081 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 344159 +eval/num paths total 361 +eval/path length Mean 738 +eval/path length Std 0 +eval/path length Max 738 +eval/path length Min 738 +eval/Rewards Mean 0.00135501 +eval/Rewards Std 0.0367856 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0143586 +eval/Actions Std 0.735823 +eval/Actions Max 0.999795 +eval/Actions Min -0.999533 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.80002e-06 +time/evaluation sampling (s) 2.49266 +time/exploration sampling (s) 2.93084 +time/logging (s) 0.00679821 +time/saving (s) 0.0105605 +time/training (s) 13.2321 +time/epoch (s) 18.6729 +time/total (s) 7481.03 +Epoch -640 +------------------------------ ---------------- +2022-05-15 20:07:26.248873 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -639 finished +------------------------------ ---------------- +epoch -639 +replay_buffer/size 999047 +trainer/num train calls 362000 +trainer/QF1 Loss 0.705195 +trainer/QF2 Loss 0.582915 +trainer/Policy Loss 31.8312 +trainer/Q1 Predictions Mean -72.6155 +trainer/Q1 Predictions Std 16.6744 +trainer/Q1 Predictions Max -4.58391 +trainer/Q1 Predictions Min -85.6761 +trainer/Q2 Predictions Mean -72.6528 +trainer/Q2 Predictions Std 16.797 +trainer/Q2 Predictions Max -3.85072 +trainer/Q2 Predictions Min -85.8597 +trainer/Q Targets Mean -72.9515 +trainer/Q Targets Std 16.9196 +trainer/Q Targets Max -3.00213 +trainer/Q Targets Min -86.0411 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00742688 +trainer/policy/mean Std 0.713393 +trainer/policy/mean Max 0.999226 +trainer/policy/mean Min -0.99938 +trainer/policy/std Mean 0.427279 +trainer/policy/std Std 0.0211547 +trainer/policy/std Max 0.450556 +trainer/policy/std Min 0.388943 +trainer/Advantage Weights Mean 9.22843 +trainer/Advantage Weights Std 23.9639 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.47869e-15 +trainer/Advantage Score Mean -0.104757 +trainer/Advantage Score Std 0.526323 +trainer/Advantage Score Max 1.3541 +trainer/Advantage Score Min -3.25267 +trainer/V1 Predictions Mean -72.7298 +trainer/V1 Predictions Std 17.0569 +trainer/V1 Predictions Max -1.11909 +trainer/V1 Predictions Min -85.8418 +trainer/VF Loss 0.0794326 +expl/num steps total 362000 +expl/num paths total 419 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.163603 +expl/Actions Std 0.891319 +expl/Actions Max 2.69018 +expl/Actions Min -2.41206 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 345159 +eval/num paths total 362 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.235044 +eval/Actions Std 0.788326 +eval/Actions Max 0.998753 +eval/Actions Min -0.995223 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.0091e-06 +time/evaluation sampling (s) 2.53293 +time/exploration sampling (s) 2.92868 +time/logging (s) 0.0105603 +time/saving (s) 0.0169689 +time/training (s) 13.1505 +time/epoch (s) 18.6397 +time/total (s) 7499.68 +Epoch -639 +------------------------------ ---------------- +2022-05-15 20:07:44.452625 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -638 finished +------------------------------ ---------------- +epoch -638 +replay_buffer/size 999047 +trainer/num train calls 363000 +trainer/QF1 Loss 1.05724 +trainer/QF2 Loss 1.00162 +trainer/Policy Loss 20.5371 +trainer/Q1 Predictions Mean -72.7964 +trainer/Q1 Predictions Std 18.3093 +trainer/Q1 Predictions Max -1.11037 +trainer/Q1 Predictions Min -86.8409 +trainer/Q2 Predictions Mean -72.8011 +trainer/Q2 Predictions Std 18.3149 +trainer/Q2 Predictions Max -0.673503 +trainer/Q2 Predictions Min -86.7286 +trainer/Q Targets Mean -72.8324 +trainer/Q Targets Std 18.2546 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6236 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0461564 +trainer/policy/mean Std 0.725954 +trainer/policy/mean Max 0.998312 +trainer/policy/mean Min -0.995876 +trainer/policy/std Mean 0.427519 +trainer/policy/std Std 0.0212119 +trainer/policy/std Max 0.452568 +trainer/policy/std Min 0.388624 +trainer/Advantage Weights Mean 4.51437 +trainer/Advantage Weights Std 18.7784 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.07412e-16 +trainer/Advantage Score Mean -0.360876 +trainer/Advantage Score Std 0.51254 +trainer/Advantage Score Max 1.54068 +trainer/Advantage Score Min -3.52172 +trainer/V1 Predictions Mean -72.6089 +trainer/V1 Predictions Std 18.3747 +trainer/V1 Predictions Max -0.887147 +trainer/V1 Predictions Min -86.5567 +trainer/VF Loss 0.064998 +expl/num steps total 363000 +expl/num paths total 420 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0332877 +expl/Actions Std 0.86311 +expl/Actions Max 2.40843 +expl/Actions Min -2.34154 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 346159 +eval/num paths total 363 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0938318 +eval/Actions Std 0.631821 +eval/Actions Max 0.998725 +eval/Actions Min -0.999592 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.21819e-06 +time/evaluation sampling (s) 2.50627 +time/exploration sampling (s) 2.85772 +time/logging (s) 0.00695822 +time/saving (s) 0.00944955 +time/training (s) 12.8107 +time/epoch (s) 18.1911 +time/total (s) 7517.87 +Epoch -638 +------------------------------ ---------------- +2022-05-15 20:08:03.494595 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -637 finished +------------------------------ ---------------- +epoch -637 +replay_buffer/size 999047 +trainer/num train calls 364000 +trainer/QF1 Loss 1.05289 +trainer/QF2 Loss 0.848269 +trainer/Policy Loss 30.6971 +trainer/Q1 Predictions Mean -70.1493 +trainer/Q1 Predictions Std 20.09 +trainer/Q1 Predictions Max -0.542344 +trainer/Q1 Predictions Min -86.6531 +trainer/Q2 Predictions Mean -70.1142 +trainer/Q2 Predictions Std 20.1924 +trainer/Q2 Predictions Max -0.480885 +trainer/Q2 Predictions Min -86.5894 +trainer/Q Targets Mean -70.1697 +trainer/Q Targets Std 20.356 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5867 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0175602 +trainer/policy/mean Std 0.708305 +trainer/policy/mean Max 0.996344 +trainer/policy/mean Min -0.998643 +trainer/policy/std Mean 0.426446 +trainer/policy/std Std 0.0196237 +trainer/policy/std Max 0.450486 +trainer/policy/std Min 0.392094 +trainer/Advantage Weights Mean 4.85862 +trainer/Advantage Weights Std 17.4751 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.93855e-18 +trainer/Advantage Score Mean -0.369462 +trainer/Advantage Score Std 0.643697 +trainer/Advantage Score Max 0.996065 +trainer/Advantage Score Min -4.03686 +trainer/V1 Predictions Mean -69.8477 +trainer/V1 Predictions Std 20.4261 +trainer/V1 Predictions Max 0.430138 +trainer/V1 Predictions Min -86.4714 +trainer/VF Loss 0.0736633 +expl/num steps total 364000 +expl/num paths total 421 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00743847 +expl/Actions Std 0.82644 +expl/Actions Max 2.38985 +expl/Actions Min -2.35636 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 347159 +eval/num paths total 364 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.22644 +eval/Actions Std 0.646561 +eval/Actions Max 0.999882 +eval/Actions Min -0.999374 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.22703e-06 +time/evaluation sampling (s) 2.45361 +time/exploration sampling (s) 2.85396 +time/logging (s) 0.00838018 +time/saving (s) 0.0160682 +time/training (s) 13.704 +time/epoch (s) 19.036 +time/total (s) 7536.91 +Epoch -637 +------------------------------ ---------------- +2022-05-15 20:08:23.557941 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -636 finished +------------------------------ ---------------- +epoch -636 +replay_buffer/size 999047 +trainer/num train calls 365000 +trainer/QF1 Loss 0.922043 +trainer/QF2 Loss 0.958286 +trainer/Policy Loss 40.3109 +trainer/Q1 Predictions Mean -71.266 +trainer/Q1 Predictions Std 20.4246 +trainer/Q1 Predictions Max -0.502253 +trainer/Q1 Predictions Min -86.3747 +trainer/Q2 Predictions Mean -71.3233 +trainer/Q2 Predictions Std 20.4705 +trainer/Q2 Predictions Max 0.786106 +trainer/Q2 Predictions Min -86.2139 +trainer/Q Targets Mean -71.5176 +trainer/Q Targets Std 20.5147 +trainer/Q Targets Max 0.201944 +trainer/Q Targets Min -85.9739 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00798168 +trainer/policy/mean Std 0.716946 +trainer/policy/mean Max 0.999826 +trainer/policy/mean Min -0.998486 +trainer/policy/std Mean 0.42732 +trainer/policy/std Std 0.0217906 +trainer/policy/std Max 0.451565 +trainer/policy/std Min 0.390475 +trainer/Advantage Weights Mean 7.14333 +trainer/Advantage Weights Std 22.7232 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.33227e-15 +trainer/Advantage Score Mean -0.2885 +trainer/Advantage Score Std 0.521086 +trainer/Advantage Score Max 1.33833 +trainer/Advantage Score Min -3.23053 +trainer/V1 Predictions Mean -71.2784 +trainer/V1 Predictions Std 20.6324 +trainer/V1 Predictions Max 0.44888 +trainer/V1 Predictions Min -86.0964 +trainer/VF Loss 0.0594109 +expl/num steps total 365000 +expl/num paths total 422 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.25273 +expl/Actions Std 0.847703 +expl/Actions Max 2.24277 +expl/Actions Min -2.51656 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 347824 +eval/num paths total 365 +eval/path length Mean 665 +eval/path length Std 0 +eval/path length Max 665 +eval/path length Min 665 +eval/Rewards Mean 0.00150376 +eval/Rewards Std 0.0387492 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0127212 +eval/Actions Std 0.731429 +eval/Actions Max 0.999822 +eval/Actions Min -0.99974 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.44519e-06 +time/evaluation sampling (s) 2.85995 +time/exploration sampling (s) 3.53834 +time/logging (s) 0.00649806 +time/saving (s) 0.0146761 +time/training (s) 13.632 +time/epoch (s) 20.0515 +time/total (s) 7556.97 +Epoch -636 +------------------------------ ---------------- +2022-05-15 20:08:43.417252 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -635 finished +------------------------------ ---------------- +epoch -635 +replay_buffer/size 999047 +trainer/num train calls 366000 +trainer/QF1 Loss 0.733984 +trainer/QF2 Loss 0.798732 +trainer/Policy Loss 12.7904 +trainer/Q1 Predictions Mean -72.4157 +trainer/Q1 Predictions Std 18.1965 +trainer/Q1 Predictions Max -0.399317 +trainer/Q1 Predictions Min -86.007 +trainer/Q2 Predictions Mean -72.4242 +trainer/Q2 Predictions Std 18.2331 +trainer/Q2 Predictions Max -0.387047 +trainer/Q2 Predictions Min -86.528 +trainer/Q Targets Mean -72.1435 +trainer/Q Targets Std 18.4815 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0153 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00360248 +trainer/policy/mean Std 0.725223 +trainer/policy/mean Max 0.998279 +trainer/policy/mean Min -0.998634 +trainer/policy/std Mean 0.424976 +trainer/policy/std Std 0.0215859 +trainer/policy/std Max 0.448969 +trainer/policy/std Min 0.389439 +trainer/Advantage Weights Mean 3.27907 +trainer/Advantage Weights Std 14.6134 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.85121e-13 +trainer/Advantage Score Mean -0.394906 +trainer/Advantage Score Std 0.54503 +trainer/Advantage Score Max 0.602167 +trainer/Advantage Score Min -2.83544 +trainer/V1 Predictions Mean -71.9037 +trainer/V1 Predictions Std 18.4518 +trainer/V1 Predictions Max -0.0606077 +trainer/V1 Predictions Min -85.7918 +trainer/VF Loss 0.0533669 +expl/num steps total 366000 +expl/num paths total 423 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0350073 +expl/Actions Std 0.843721 +expl/Actions Max 2.21317 +expl/Actions Min -2.31244 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 348824 +eval/num paths total 366 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.452756 +eval/Actions Std 0.499135 +eval/Actions Max 0.999033 +eval/Actions Min -0.998656 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.91831e-06 +time/evaluation sampling (s) 2.84798 +time/exploration sampling (s) 3.39667 +time/logging (s) 0.00749174 +time/saving (s) 0.0115995 +time/training (s) 13.5876 +time/epoch (s) 19.8514 +time/total (s) 7576.83 +Epoch -635 +------------------------------ ---------------- +2022-05-15 20:09:03.409976 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -634 finished +------------------------------ ---------------- +epoch -634 +replay_buffer/size 999047 +trainer/num train calls 367000 +trainer/QF1 Loss 0.728834 +trainer/QF2 Loss 0.6347 +trainer/Policy Loss 30.1496 +trainer/Q1 Predictions Mean -72.7881 +trainer/Q1 Predictions Std 17.3905 +trainer/Q1 Predictions Max -2.14009 +trainer/Q1 Predictions Min -86.7683 +trainer/Q2 Predictions Mean -72.8304 +trainer/Q2 Predictions Std 17.4697 +trainer/Q2 Predictions Max -1.794 +trainer/Q2 Predictions Min -86.68 +trainer/Q Targets Mean -72.7453 +trainer/Q Targets Std 17.4178 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2445 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00043868 +trainer/policy/mean Std 0.721296 +trainer/policy/mean Max 0.996699 +trainer/policy/mean Min -0.998137 +trainer/policy/std Mean 0.424676 +trainer/policy/std Std 0.0207569 +trainer/policy/std Max 0.447214 +trainer/policy/std Min 0.386868 +trainer/Advantage Weights Mean 5.66576 +trainer/Advantage Weights Std 20.619 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.3461e-15 +trainer/Advantage Score Mean -0.282047 +trainer/Advantage Score Std 0.534702 +trainer/Advantage Score Max 1.22661 +trainer/Advantage Score Min -3.3686 +trainer/V1 Predictions Mean -72.5056 +trainer/V1 Predictions Std 17.5133 +trainer/V1 Predictions Max -1.45505 +trainer/V1 Predictions Min -86.159 +trainer/VF Loss 0.0626297 +expl/num steps total 367000 +expl/num paths total 424 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0373904 +expl/Actions Std 0.841695 +expl/Actions Max 2.47388 +expl/Actions Min -2.243 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 349824 +eval/num paths total 367 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.182505 +eval/Actions Std 0.613289 +eval/Actions Max 0.999802 +eval/Actions Min -0.998919 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.37722e-06 +time/evaluation sampling (s) 2.93267 +time/exploration sampling (s) 3.44849 +time/logging (s) 0.00849972 +time/saving (s) 0.0178987 +time/training (s) 13.5783 +time/epoch (s) 19.9859 +time/total (s) 7596.82 +Epoch -634 +------------------------------ ---------------- +2022-05-15 20:09:26.143518 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -633 finished +------------------------------ ---------------- +epoch -633 +replay_buffer/size 999047 +trainer/num train calls 368000 +trainer/QF1 Loss 3.17184 +trainer/QF2 Loss 3.37949 +trainer/Policy Loss 29.5095 +trainer/Q1 Predictions Mean -69.9504 +trainer/Q1 Predictions Std 20.1392 +trainer/Q1 Predictions Max -0.849628 +trainer/Q1 Predictions Min -85.7494 +trainer/Q2 Predictions Mean -69.8795 +trainer/Q2 Predictions Std 20.1923 +trainer/Q2 Predictions Max -1.34613 +trainer/Q2 Predictions Min -85.8286 +trainer/Q Targets Mean -70.3894 +trainer/Q Targets Std 19.8726 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.7522 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0201905 +trainer/policy/mean Std 0.71997 +trainer/policy/mean Max 0.999603 +trainer/policy/mean Min -0.998492 +trainer/policy/std Mean 0.427059 +trainer/policy/std Std 0.0207811 +trainer/policy/std Max 0.448459 +trainer/policy/std Min 0.390882 +trainer/Advantage Weights Mean 8.59689 +trainer/Advantage Weights Std 26.2445 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.78859e-18 +trainer/Advantage Score Mean -0.293146 +trainer/Advantage Score Std 0.569186 +trainer/Advantage Score Max 3.05662 +trainer/Advantage Score Min -3.98803 +trainer/V1 Predictions Mean -70.1749 +trainer/V1 Predictions Std 19.8782 +trainer/V1 Predictions Max -0.501985 +trainer/V1 Predictions Min -85.3459 +trainer/VF Loss 0.112015 +expl/num steps total 368000 +expl/num paths total 425 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0478444 +expl/Actions Std 0.821592 +expl/Actions Max 2.2935 +expl/Actions Min -2.46832 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 350824 +eval/num paths total 368 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.181977 +eval/Actions Std 0.67665 +eval/Actions Max 0.999287 +eval/Actions Min -0.998887 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.57604e-06 +time/evaluation sampling (s) 2.82021 +time/exploration sampling (s) 3.50481 +time/logging (s) 0.0114417 +time/saving (s) 0.0156186 +time/training (s) 16.3753 +time/epoch (s) 22.7274 +time/total (s) 7619.55 +Epoch -633 +------------------------------ ---------------- +2022-05-15 20:09:55.222223 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -632 finished +------------------------------ ---------------- +epoch -632 +replay_buffer/size 999047 +trainer/num train calls 369000 +trainer/QF1 Loss 1.34391 +trainer/QF2 Loss 1.27988 +trainer/Policy Loss 22.7118 +trainer/Q1 Predictions Mean -73.2102 +trainer/Q1 Predictions Std 17.83 +trainer/Q1 Predictions Max -1.30126 +trainer/Q1 Predictions Min -85.5378 +trainer/Q2 Predictions Mean -73.1676 +trainer/Q2 Predictions Std 17.8931 +trainer/Q2 Predictions Max -0.715349 +trainer/Q2 Predictions Min -85.6098 +trainer/Q Targets Mean -73.0307 +trainer/Q Targets Std 18.4112 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0456 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0210335 +trainer/policy/mean Std 0.704659 +trainer/policy/mean Max 0.999342 +trainer/policy/mean Min -0.997161 +trainer/policy/std Mean 0.426455 +trainer/policy/std Std 0.0212236 +trainer/policy/std Max 0.449389 +trainer/policy/std Min 0.390447 +trainer/Advantage Weights Mean 6.52112 +trainer/Advantage Weights Std 21.1122 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.94445e-22 +trainer/Advantage Score Mean -0.294549 +trainer/Advantage Score Std 0.638386 +trainer/Advantage Score Max 1.14721 +trainer/Advantage Score Min -4.87189 +trainer/V1 Predictions Mean -72.8782 +trainer/V1 Predictions Std 18.3652 +trainer/V1 Predictions Max 0.261653 +trainer/V1 Predictions Min -86.3822 +trainer/VF Loss 0.0726294 +expl/num steps total 369000 +expl/num paths total 426 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0165954 +expl/Actions Std 0.814485 +expl/Actions Max 2.23252 +expl/Actions Min -2.36839 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 351824 +eval/num paths total 369 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.332777 +eval/Actions Std 0.667953 +eval/Actions Max 0.999676 +eval/Actions Min -0.998223 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.1188e-05 +time/evaluation sampling (s) 4.57244 +time/exploration sampling (s) 6.07458 +time/logging (s) 0.0117309 +time/saving (s) 0.0154572 +time/training (s) 18.3916 +time/epoch (s) 29.0658 +time/total (s) 7648.62 +Epoch -632 +------------------------------ ---------------- +2022-05-15 20:10:24.778706 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -631 finished +------------------------------ ---------------- +epoch -631 +replay_buffer/size 999047 +trainer/num train calls 370000 +trainer/QF1 Loss 11.061 +trainer/QF2 Loss 11.2641 +trainer/Policy Loss 5.3613 +trainer/Q1 Predictions Mean -71.1747 +trainer/Q1 Predictions Std 19.6688 +trainer/Q1 Predictions Max -1.15033 +trainer/Q1 Predictions Min -86.7163 +trainer/Q2 Predictions Mean -71.2224 +trainer/Q2 Predictions Std 19.6654 +trainer/Q2 Predictions Max -0.350309 +trainer/Q2 Predictions Min -86.8176 +trainer/Q Targets Mean -70.7218 +trainer/Q Targets Std 19.4072 +trainer/Q Targets Max 0.0598445 +trainer/Q Targets Min -86.1486 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0103282 +trainer/policy/mean Std 0.70045 +trainer/policy/mean Max 0.99792 +trainer/policy/mean Min -0.997137 +trainer/policy/std Mean 0.426599 +trainer/policy/std Std 0.0199833 +trainer/policy/std Max 0.44797 +trainer/policy/std Min 0.393734 +trainer/Advantage Weights Mean 0.909001 +trainer/Advantage Weights Std 8.86618 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.74468e-16 +trainer/Advantage Score Mean -0.856892 +trainer/Advantage Score Std 0.576831 +trainer/Advantage Score Max 1.33001 +trainer/Advantage Score Min -3.62848 +trainer/V1 Predictions Mean -70.2414 +trainer/V1 Predictions Std 19.7765 +trainer/V1 Predictions Max 0.720809 +trainer/V1 Predictions Min -86.2372 +trainer/VF Loss 0.113746 +expl/num steps total 370000 +expl/num paths total 428 +expl/path length Mean 500 +expl/path length Std 385 +expl/path length Max 885 +expl/path length Min 115 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0327607 +expl/Actions Std 0.813884 +expl/Actions Max 2.2857 +expl/Actions Min -2.32113 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 352578 +eval/num paths total 370 +eval/path length Mean 754 +eval/path length Std 0 +eval/path length Max 754 +eval/path length Min 754 +eval/Rewards Mean 0.00132626 +eval/Rewards Std 0.0363937 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0294193 +eval/Actions Std 0.731505 +eval/Actions Max 0.999597 +eval/Actions Min -0.999162 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.29309e-05 +time/evaluation sampling (s) 4.72762 +time/exploration sampling (s) 5.9548 +time/logging (s) 0.0101684 +time/saving (s) 0.0167541 +time/training (s) 18.8327 +time/epoch (s) 29.542 +time/total (s) 7678.17 +Epoch -631 +------------------------------ ---------------- +2022-05-15 20:10:54.420498 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -630 finished +------------------------------ ---------------- +epoch -630 +replay_buffer/size 999047 +trainer/num train calls 371000 +trainer/QF1 Loss 0.781266 +trainer/QF2 Loss 0.763821 +trainer/Policy Loss 30.1418 +trainer/Q1 Predictions Mean -71.9855 +trainer/Q1 Predictions Std 18.1963 +trainer/Q1 Predictions Max -0.251175 +trainer/Q1 Predictions Min -86.4468 +trainer/Q2 Predictions Mean -72.0263 +trainer/Q2 Predictions Std 18.1808 +trainer/Q2 Predictions Max 0.0492507 +trainer/Q2 Predictions Min -86.221 +trainer/Q Targets Mean -72.3764 +trainer/Q Targets Std 18.0797 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2018 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0105469 +trainer/policy/mean Std 0.720585 +trainer/policy/mean Max 0.998595 +trainer/policy/mean Min -0.997093 +trainer/policy/std Mean 0.426553 +trainer/policy/std Std 0.019785 +trainer/policy/std Max 0.447826 +trainer/policy/std Min 0.392898 +trainer/Advantage Weights Mean 7.18322 +trainer/Advantage Weights Std 22.4747 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.40007e-16 +trainer/Advantage Score Mean -0.228485 +trainer/Advantage Score Std 0.482729 +trainer/Advantage Score Max 1.08697 +trainer/Advantage Score Min -3.59659 +trainer/V1 Predictions Mean -72.1157 +trainer/V1 Predictions Std 18.1117 +trainer/V1 Predictions Max -0.143927 +trainer/V1 Predictions Min -85.9115 +trainer/VF Loss 0.0569772 +expl/num steps total 371000 +expl/num paths total 430 +expl/path length Mean 500 +expl/path length Std 195 +expl/path length Max 695 +expl/path length Min 305 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0148787 +expl/Actions Std 0.825198 +expl/Actions Max 2.53544 +expl/Actions Min -2.44666 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 353578 +eval/num paths total 371 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0998428 +eval/Actions Std 0.659903 +eval/Actions Max 0.999754 +eval/Actions Min -0.998834 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.92414e-06 +time/evaluation sampling (s) 4.6218 +time/exploration sampling (s) 6.41389 +time/logging (s) 0.012865 +time/saving (s) 0.0119587 +time/training (s) 18.5716 +time/epoch (s) 29.6321 +time/total (s) 7707.81 +Epoch -630 +------------------------------ ---------------- +2022-05-15 20:11:25.204153 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -629 finished +------------------------------ ---------------- +epoch -629 +replay_buffer/size 999047 +trainer/num train calls 372000 +trainer/QF1 Loss 1.05994 +trainer/QF2 Loss 1.0027 +trainer/Policy Loss 34.8959 +trainer/Q1 Predictions Mean -71.5934 +trainer/Q1 Predictions Std 18.9485 +trainer/Q1 Predictions Max -3.29591 +trainer/Q1 Predictions Min -86.4198 +trainer/Q2 Predictions Mean -71.6454 +trainer/Q2 Predictions Std 18.9517 +trainer/Q2 Predictions Max -3.49615 +trainer/Q2 Predictions Min -86.124 +trainer/Q Targets Mean -71.9442 +trainer/Q Targets Std 18.5413 +trainer/Q Targets Max -3.74618 +trainer/Q Targets Min -87.0088 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0165767 +trainer/policy/mean Std 0.704736 +trainer/policy/mean Max 0.996761 +trainer/policy/mean Min -0.999647 +trainer/policy/std Mean 0.427084 +trainer/policy/std Std 0.0206104 +trainer/policy/std Max 0.450544 +trainer/policy/std Min 0.389691 +trainer/Advantage Weights Mean 7.07434 +trainer/Advantage Weights Std 21.3113 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.16975e-09 +trainer/Advantage Score Mean -0.185445 +trainer/Advantage Score Std 0.47234 +trainer/Advantage Score Max 1.99026 +trainer/Advantage Score Min -1.99487 +trainer/V1 Predictions Mean -71.6616 +trainer/V1 Predictions Std 18.7542 +trainer/V1 Predictions Max -3.24954 +trainer/V1 Predictions Min -86.1421 +trainer/VF Loss 0.0770023 +expl/num steps total 372000 +expl/num paths total 431 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0405882 +expl/Actions Std 0.820184 +expl/Actions Max 2.62354 +expl/Actions Min -2.63441 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 354578 +eval/num paths total 372 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0259197 +eval/Actions Std 0.759595 +eval/Actions Max 0.999896 +eval/Actions Min -0.999773 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.12089e-05 +time/evaluation sampling (s) 4.94639 +time/exploration sampling (s) 7.0276 +time/logging (s) 0.011613 +time/saving (s) 0.015326 +time/training (s) 18.7711 +time/epoch (s) 30.772 +time/total (s) 7738.58 +Epoch -629 +------------------------------ ---------------- +2022-05-15 20:11:56.138196 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -628 finished +------------------------------ ---------------- +epoch -628 +replay_buffer/size 999047 +trainer/num train calls 373000 +trainer/QF1 Loss 0.57674 +trainer/QF2 Loss 0.565531 +trainer/Policy Loss 9.23233 +trainer/Q1 Predictions Mean -72.308 +trainer/Q1 Predictions Std 17.7286 +trainer/Q1 Predictions Max -0.542985 +trainer/Q1 Predictions Min -86.005 +trainer/Q2 Predictions Mean -72.3157 +trainer/Q2 Predictions Std 17.6297 +trainer/Q2 Predictions Max -0.713776 +trainer/Q2 Predictions Min -86.3474 +trainer/Q Targets Mean -72.1265 +trainer/Q Targets Std 17.6507 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5152 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.013215 +trainer/policy/mean Std 0.724231 +trainer/policy/mean Max 0.999307 +trainer/policy/mean Min -0.998989 +trainer/policy/std Mean 0.42623 +trainer/policy/std Std 0.0204893 +trainer/policy/std Max 0.448316 +trainer/policy/std Min 0.3865 +trainer/Advantage Weights Mean 2.52593 +trainer/Advantage Weights Std 14.6776 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.86844e-14 +trainer/Advantage Score Mean -0.513203 +trainer/Advantage Score Std 0.508073 +trainer/Advantage Score Max 1.24539 +trainer/Advantage Score Min -3.00537 +trainer/V1 Predictions Mean -71.8044 +trainer/V1 Predictions Std 17.8989 +trainer/V1 Predictions Max 0.00292289 +trainer/V1 Predictions Min -86.6099 +trainer/VF Loss 0.0620193 +expl/num steps total 373000 +expl/num paths total 432 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.165468 +expl/Actions Std 0.847188 +expl/Actions Max 2.61534 +expl/Actions Min -2.56033 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 355311 +eval/num paths total 373 +eval/path length Mean 733 +eval/path length Std 0 +eval/path length Max 733 +eval/path length Min 733 +eval/Rewards Mean 0.00136426 +eval/Rewards Std 0.0369106 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0286461 +eval/Actions Std 0.747133 +eval/Actions Max 0.999656 +eval/Actions Min -0.998193 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.17838e-06 +time/evaluation sampling (s) 5.03761 +time/exploration sampling (s) 6.85106 +time/logging (s) 0.0115992 +time/saving (s) 0.0183776 +time/training (s) 18.9996 +time/epoch (s) 30.9182 +time/total (s) 7769.51 +Epoch -628 +------------------------------ ---------------- +2022-05-15 20:12:27.041920 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -627 finished +------------------------------ ---------------- +epoch -627 +replay_buffer/size 999047 +trainer/num train calls 374000 +trainer/QF1 Loss 0.638652 +trainer/QF2 Loss 0.547958 +trainer/Policy Loss 7.23782 +trainer/Q1 Predictions Mean -71.1916 +trainer/Q1 Predictions Std 19.7408 +trainer/Q1 Predictions Max -0.484257 +trainer/Q1 Predictions Min -86.2395 +trainer/Q2 Predictions Mean -71.1444 +trainer/Q2 Predictions Std 19.7035 +trainer/Q2 Predictions Max -0.428121 +trainer/Q2 Predictions Min -85.9987 +trainer/Q Targets Mean -71.0522 +trainer/Q Targets Std 19.8272 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0032 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000796474 +trainer/policy/mean Std 0.711179 +trainer/policy/mean Max 0.999558 +trainer/policy/mean Min -0.995581 +trainer/policy/std Mean 0.425822 +trainer/policy/std Std 0.0197523 +trainer/policy/std Max 0.4469 +trainer/policy/std Min 0.388781 +trainer/Advantage Weights Mean 1.864 +trainer/Advantage Weights Std 10.2013 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.28487e-15 +trainer/Advantage Score Mean -0.536218 +trainer/Advantage Score Std 0.591122 +trainer/Advantage Score Max 0.722382 +trainer/Advantage Score Min -3.30837 +trainer/V1 Predictions Mean -70.7965 +trainer/V1 Predictions Std 19.9893 +trainer/V1 Predictions Max 0.99024 +trainer/V1 Predictions Min -85.8621 +trainer/VF Loss 0.0691321 +expl/num steps total 374000 +expl/num paths total 433 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00993898 +expl/Actions Std 0.82261 +expl/Actions Max 2.39271 +expl/Actions Min -2.30867 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 356311 +eval/num paths total 374 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.580583 +eval/Actions Std 0.486301 +eval/Actions Max 0.998472 +eval/Actions Min -0.99951 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.04569e-05 +time/evaluation sampling (s) 5.09529 +time/exploration sampling (s) 6.7929 +time/logging (s) 0.0123715 +time/saving (s) 0.0170911 +time/training (s) 18.9725 +time/epoch (s) 30.8902 +time/total (s) 7800.41 +Epoch -627 +------------------------------ ---------------- +2022-05-15 20:12:58.801234 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -626 finished +------------------------------ ---------------- +epoch -626 +replay_buffer/size 999047 +trainer/num train calls 375000 +trainer/QF1 Loss 1.58802 +trainer/QF2 Loss 1.38672 +trainer/Policy Loss 20.2989 +trainer/Q1 Predictions Mean -69.3183 +trainer/Q1 Predictions Std 21.7329 +trainer/Q1 Predictions Max -0.418765 +trainer/Q1 Predictions Min -85.9292 +trainer/Q2 Predictions Mean -69.4423 +trainer/Q2 Predictions Std 21.7316 +trainer/Q2 Predictions Max -0.340348 +trainer/Q2 Predictions Min -85.799 +trainer/Q Targets Mean -69.4482 +trainer/Q Targets Std 22.0552 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.1288 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00112826 +trainer/policy/mean Std 0.727406 +trainer/policy/mean Max 0.999293 +trainer/policy/mean Min -0.998646 +trainer/policy/std Mean 0.426124 +trainer/policy/std Std 0.0207729 +trainer/policy/std Max 0.446476 +trainer/policy/std Min 0.389346 +trainer/Advantage Weights Mean 4.17727 +trainer/Advantage Weights Std 17.4274 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.47758e-16 +trainer/Advantage Score Mean -0.352086 +trainer/Advantage Score Std 0.534649 +trainer/Advantage Score Max 1.02072 +trainer/Advantage Score Min -3.6451 +trainer/V1 Predictions Mean -69.2158 +trainer/V1 Predictions Std 22.0351 +trainer/V1 Predictions Max 1.64364 +trainer/V1 Predictions Min -85.9558 +trainer/VF Loss 0.0581981 +expl/num steps total 375000 +expl/num paths total 434 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.274466 +expl/Actions Std 0.796867 +expl/Actions Max 2.24847 +expl/Actions Min -2.47702 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 357311 +eval/num paths total 375 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0298315 +eval/Actions Std 0.591035 +eval/Actions Max 0.998711 +eval/Actions Min -0.999859 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.3167e-05 +time/evaluation sampling (s) 5.34173 +time/exploration sampling (s) 7.50351 +time/logging (s) 0.00995065 +time/saving (s) 0.0119449 +time/training (s) 18.8755 +time/epoch (s) 31.7427 +time/total (s) 7832.16 +Epoch -626 +------------------------------ ---------------- +2022-05-15 20:13:29.299446 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -625 finished +------------------------------ ---------------- +epoch -625 +replay_buffer/size 999047 +trainer/num train calls 376000 +trainer/QF1 Loss 0.608561 +trainer/QF2 Loss 0.752752 +trainer/Policy Loss 11.333 +trainer/Q1 Predictions Mean -72.8871 +trainer/Q1 Predictions Std 17.2274 +trainer/Q1 Predictions Max -1.74331 +trainer/Q1 Predictions Min -86.2006 +trainer/Q2 Predictions Mean -72.8862 +trainer/Q2 Predictions Std 17.2011 +trainer/Q2 Predictions Max -1.3601 +trainer/Q2 Predictions Min -86.2061 +trainer/Q Targets Mean -72.9787 +trainer/Q Targets Std 17.2873 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0926 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0199506 +trainer/policy/mean Std 0.714473 +trainer/policy/mean Max 0.997795 +trainer/policy/mean Min -0.999681 +trainer/policy/std Mean 0.425388 +trainer/policy/std Std 0.0213755 +trainer/policy/std Max 0.447731 +trainer/policy/std Min 0.38743 +trainer/Advantage Weights Mean 3.36879 +trainer/Advantage Weights Std 14.5037 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.87437e-17 +trainer/Advantage Score Mean -0.384525 +trainer/Advantage Score Std 0.52145 +trainer/Advantage Score Max 1.15693 +trainer/Advantage Score Min -3.80881 +trainer/V1 Predictions Mean -72.7669 +trainer/V1 Predictions Std 17.3641 +trainer/V1 Predictions Max -0.966972 +trainer/V1 Predictions Min -86.0714 +trainer/VF Loss 0.056089 +expl/num steps total 376000 +expl/num paths total 435 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00785928 +expl/Actions Std 0.81657 +expl/Actions Max 2.65859 +expl/Actions Min -2.42038 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 358311 +eval/num paths total 376 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0101111 +eval/Actions Std 0.732251 +eval/Actions Max 0.999717 +eval/Actions Min -0.999532 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0476e-05 +time/evaluation sampling (s) 4.82415 +time/exploration sampling (s) 6.67346 +time/logging (s) 0.0100395 +time/saving (s) 0.0169962 +time/training (s) 18.9623 +time/epoch (s) 30.4869 +time/total (s) 7862.65 +Epoch -625 +------------------------------ ---------------- +2022-05-15 20:13:59.869073 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -624 finished +------------------------------ ---------------- +epoch -624 +replay_buffer/size 999047 +trainer/num train calls 377000 +trainer/QF1 Loss 0.698463 +trainer/QF2 Loss 0.663924 +trainer/Policy Loss 12.5052 +trainer/Q1 Predictions Mean -72.2427 +trainer/Q1 Predictions Std 17.6402 +trainer/Q1 Predictions Max -3.65268 +trainer/Q1 Predictions Min -86.2304 +trainer/Q2 Predictions Mean -72.252 +trainer/Q2 Predictions Std 17.7085 +trainer/Q2 Predictions Max -2.50709 +trainer/Q2 Predictions Min -86.5035 +trainer/Q Targets Mean -72.3873 +trainer/Q Targets Std 17.4469 +trainer/Q Targets Max -3.534 +trainer/Q Targets Min -86.31 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.012957 +trainer/policy/mean Std 0.715507 +trainer/policy/mean Max 0.999448 +trainer/policy/mean Min -0.999747 +trainer/policy/std Mean 0.425962 +trainer/policy/std Std 0.0204941 +trainer/policy/std Max 0.44571 +trainer/policy/std Min 0.389628 +trainer/Advantage Weights Mean 3.88231 +trainer/Advantage Weights Std 17.9237 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.73788e-11 +trainer/Advantage Score Mean -0.397268 +trainer/Advantage Score Std 0.516688 +trainer/Advantage Score Max 2.14026 +trainer/Advantage Score Min -2.47758 +trainer/V1 Predictions Mean -72.0995 +trainer/V1 Predictions Std 17.6552 +trainer/V1 Predictions Max -3.52631 +trainer/V1 Predictions Min -86.2274 +trainer/VF Loss 0.076369 +expl/num steps total 377000 +expl/num paths total 437 +expl/path length Mean 500 +expl/path length Std 35 +expl/path length Max 535 +expl/path length Min 465 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0373535 +expl/Actions Std 0.826775 +expl/Actions Max 2.39801 +expl/Actions Min -2.44165 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 359311 +eval/num paths total 377 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0931484 +eval/Actions Std 0.711323 +eval/Actions Max 0.998571 +eval/Actions Min -0.997226 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0869e-05 +time/evaluation sampling (s) 5.1451 +time/exploration sampling (s) 6.90113 +time/logging (s) 0.0125238 +time/saving (s) 0.0184488 +time/training (s) 18.483 +time/epoch (s) 30.5602 +time/total (s) 7893.22 +Epoch -624 +------------------------------ ---------------- +2022-05-15 20:14:30.593063 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -623 finished +------------------------------ ---------------- +epoch -623 +replay_buffer/size 999047 +trainer/num train calls 378000 +trainer/QF1 Loss 1.15418 +trainer/QF2 Loss 1.34521 +trainer/Policy Loss 96.421 +trainer/Q1 Predictions Mean -73.4598 +trainer/Q1 Predictions Std 15.983 +trainer/Q1 Predictions Max -1.81422 +trainer/Q1 Predictions Min -86.3831 +trainer/Q2 Predictions Mean -73.4422 +trainer/Q2 Predictions Std 15.944 +trainer/Q2 Predictions Max -2.97988 +trainer/Q2 Predictions Min -85.5085 +trainer/Q Targets Mean -74.1052 +trainer/Q Targets Std 16.3 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9037 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0125392 +trainer/policy/mean Std 0.718746 +trainer/policy/mean Max 0.996964 +trainer/policy/mean Min -0.999231 +trainer/policy/std Mean 0.427232 +trainer/policy/std Std 0.0202953 +trainer/policy/std Max 0.446872 +trainer/policy/std Min 0.388995 +trainer/Advantage Weights Mean 15.8711 +trainer/Advantage Weights Std 31.8279 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.89429e-14 +trainer/Advantage Score Mean -0.141609 +trainer/Advantage Score Std 0.650654 +trainer/Advantage Score Max 1.01403 +trainer/Advantage Score Min -3.15973 +trainer/V1 Predictions Mean -73.8931 +trainer/V1 Predictions Std 16.3223 +trainer/V1 Predictions Max -1.28377 +trainer/V1 Predictions Min -86.3309 +trainer/VF Loss 0.0997085 +expl/num steps total 378000 +expl/num paths total 439 +expl/path length Mean 500 +expl/path length Std 338 +expl/path length Max 838 +expl/path length Min 162 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0332886 +expl/Actions Std 0.85269 +expl/Actions Max 2.38082 +expl/Actions Min -2.61111 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 360311 +eval/num paths total 378 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.134279 +eval/Actions Std 0.802676 +eval/Actions Max 0.998677 +eval/Actions Min -0.998817 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.07652e-05 +time/evaluation sampling (s) 5.47883 +time/exploration sampling (s) 6.31423 +time/logging (s) 0.0095674 +time/saving (s) 0.0164785 +time/training (s) 18.8884 +time/epoch (s) 30.7075 +time/total (s) 7923.93 +Epoch -623 +------------------------------ ---------------- +2022-05-15 20:15:01.057495 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -622 finished +------------------------------ ---------------- +epoch -622 +replay_buffer/size 999047 +trainer/num train calls 379000 +trainer/QF1 Loss 0.387403 +trainer/QF2 Loss 0.427775 +trainer/Policy Loss 32.1714 +trainer/Q1 Predictions Mean -75.1196 +trainer/Q1 Predictions Std 14.3893 +trainer/Q1 Predictions Max -0.430257 +trainer/Q1 Predictions Min -86.69 +trainer/Q2 Predictions Mean -75.1607 +trainer/Q2 Predictions Std 14.3742 +trainer/Q2 Predictions Max -0.592465 +trainer/Q2 Predictions Min -86.5412 +trainer/Q Targets Mean -75.2065 +trainer/Q Targets Std 14.4897 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9452 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0331687 +trainer/policy/mean Std 0.719201 +trainer/policy/mean Max 0.998149 +trainer/policy/mean Min -0.995738 +trainer/policy/std Mean 0.426438 +trainer/policy/std Std 0.0203481 +trainer/policy/std Max 0.445267 +trainer/policy/std Min 0.388404 +trainer/Advantage Weights Mean 4.8694 +trainer/Advantage Weights Std 16.5388 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.5634e-12 +trainer/Advantage Score Mean -0.213166 +trainer/Advantage Score Std 0.426346 +trainer/Advantage Score Max 0.612927 +trainer/Advantage Score Min -2.57495 +trainer/V1 Predictions Mean -74.9913 +trainer/V1 Predictions Std 14.4692 +trainer/V1 Predictions Max -0.149756 +trainer/V1 Predictions Min -86.7509 +trainer/VF Loss 0.0351571 +expl/num steps total 379000 +expl/num paths total 440 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.158914 +expl/Actions Std 0.832753 +expl/Actions Max 2.54131 +expl/Actions Min -2.29332 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 361311 +eval/num paths total 379 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0637944 +eval/Actions Std 0.631233 +eval/Actions Max 0.999618 +eval/Actions Min -0.998814 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.31202e-06 +time/evaluation sampling (s) 5.01626 +time/exploration sampling (s) 6.2904 +time/logging (s) 0.0113731 +time/saving (s) 0.0169929 +time/training (s) 19.122 +time/epoch (s) 30.457 +time/total (s) 7954.39 +Epoch -622 +------------------------------ ---------------- +2022-05-15 20:15:31.729996 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -621 finished +------------------------------ ---------------- +epoch -621 +replay_buffer/size 999047 +trainer/num train calls 380000 +trainer/QF1 Loss 0.858643 +trainer/QF2 Loss 0.873997 +trainer/Policy Loss 8.11245 +trainer/Q1 Predictions Mean -71.7161 +trainer/Q1 Predictions Std 19.5785 +trainer/Q1 Predictions Max -0.339899 +trainer/Q1 Predictions Min -86.7964 +trainer/Q2 Predictions Mean -71.6939 +trainer/Q2 Predictions Std 19.4886 +trainer/Q2 Predictions Max -0.444799 +trainer/Q2 Predictions Min -86.5755 +trainer/Q Targets Mean -71.5078 +trainer/Q Targets Std 19.7351 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00250474 +trainer/policy/mean Std 0.719587 +trainer/policy/mean Max 0.998622 +trainer/policy/mean Min -0.998666 +trainer/policy/std Mean 0.425231 +trainer/policy/std Std 0.0210838 +trainer/policy/std Max 0.445137 +trainer/policy/std Min 0.388517 +trainer/Advantage Weights Mean 2.88737 +trainer/Advantage Weights Std 13.7602 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.25681e-18 +trainer/Advantage Score Mean -0.380219 +trainer/Advantage Score Std 0.612264 +trainer/Advantage Score Max 1.13301 +trainer/Advantage Score Min -4.1218 +trainer/V1 Predictions Mean -71.2768 +trainer/V1 Predictions Std 19.8395 +trainer/V1 Predictions Max 0.47047 +trainer/V1 Predictions Min -86.4591 +trainer/VF Loss 0.0672766 +expl/num steps total 380000 +expl/num paths total 441 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.239337 +expl/Actions Std 0.864024 +expl/Actions Max 2.45722 +expl/Actions Min -2.28309 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 361972 +eval/num paths total 380 +eval/path length Mean 661 +eval/path length Std 0 +eval/path length Max 661 +eval/path length Min 661 +eval/Rewards Mean 0.00151286 +eval/Rewards Std 0.0388661 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0255202 +eval/Actions Std 0.732952 +eval/Actions Max 0.999865 +eval/Actions Min -0.998308 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.13384e-05 +time/evaluation sampling (s) 5.06818 +time/exploration sampling (s) 6.25033 +time/logging (s) 0.00957896 +time/saving (s) 0.0175424 +time/training (s) 19.3159 +time/epoch (s) 30.6616 +time/total (s) 7985.06 +Epoch -621 +------------------------------ ---------------- +2022-05-15 20:16:01.981774 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -620 finished +------------------------------ ---------------- +epoch -620 +replay_buffer/size 999047 +trainer/num train calls 381000 +trainer/QF1 Loss 0.828833 +trainer/QF2 Loss 0.850771 +trainer/Policy Loss 22.9438 +trainer/Q1 Predictions Mean -72.6326 +trainer/Q1 Predictions Std 16.8883 +trainer/Q1 Predictions Max -0.918386 +trainer/Q1 Predictions Min -86.1235 +trainer/Q2 Predictions Mean -72.6344 +trainer/Q2 Predictions Std 16.9263 +trainer/Q2 Predictions Max -1.64172 +trainer/Q2 Predictions Min -86.2204 +trainer/Q Targets Mean -72.7275 +trainer/Q Targets Std 16.4741 +trainer/Q Targets Max -2.55452 +trainer/Q Targets Min -86.3072 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00111626 +trainer/policy/mean Std 0.723603 +trainer/policy/mean Max 0.998987 +trainer/policy/mean Min -0.998268 +trainer/policy/std Mean 0.424742 +trainer/policy/std Std 0.0201139 +trainer/policy/std Max 0.444944 +trainer/policy/std Min 0.389111 +trainer/Advantage Weights Mean 4.81408 +trainer/Advantage Weights Std 19.399 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.51424e-12 +trainer/Advantage Score Mean -0.345478 +trainer/Advantage Score Std 0.554884 +trainer/Advantage Score Max 2.71782 +trainer/Advantage Score Min -2.72161 +trainer/V1 Predictions Mean -72.5276 +trainer/V1 Predictions Std 16.6002 +trainer/V1 Predictions Max -1.41576 +trainer/V1 Predictions Min -86.2336 +trainer/VF Loss 0.101972 +expl/num steps total 381000 +expl/num paths total 442 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0324397 +expl/Actions Std 0.826757 +expl/Actions Max 2.25556 +expl/Actions Min -2.5532 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 362972 +eval/num paths total 381 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.130624 +eval/Actions Std 0.642276 +eval/Actions Max 0.998193 +eval/Actions Min -0.999362 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.49599e-06 +time/evaluation sampling (s) 4.75783 +time/exploration sampling (s) 6.47759 +time/logging (s) 0.00835042 +time/saving (s) 0.0138917 +time/training (s) 18.9809 +time/epoch (s) 30.2386 +time/total (s) 8015.3 +Epoch -620 +------------------------------ ---------------- +2022-05-15 20:16:31.823222 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -619 finished +------------------------------ ---------------- +epoch -619 +replay_buffer/size 999047 +trainer/num train calls 382000 +trainer/QF1 Loss 0.912606 +trainer/QF2 Loss 1.13697 +trainer/Policy Loss 13.5916 +trainer/Q1 Predictions Mean -71.9329 +trainer/Q1 Predictions Std 19.4914 +trainer/Q1 Predictions Max -0.441102 +trainer/Q1 Predictions Min -87.1891 +trainer/Q2 Predictions Mean -72.0153 +trainer/Q2 Predictions Std 19.5648 +trainer/Q2 Predictions Max -0.864576 +trainer/Q2 Predictions Min -87.5769 +trainer/Q Targets Mean -71.4421 +trainer/Q Targets Std 19.6026 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6527 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0134543 +trainer/policy/mean Std 0.73251 +trainer/policy/mean Max 0.999365 +trainer/policy/mean Min -0.998665 +trainer/policy/std Mean 0.424846 +trainer/policy/std Std 0.0203827 +trainer/policy/std Max 0.447811 +trainer/policy/std Min 0.389254 +trainer/Advantage Weights Mean 2.43447 +trainer/Advantage Weights Std 14.0984 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.95908e-14 +trainer/Advantage Score Mean -0.485525 +trainer/Advantage Score Std 0.492028 +trainer/Advantage Score Max 1.97702 +trainer/Advantage Score Min -3.02961 +trainer/V1 Predictions Mean -71.1979 +trainer/V1 Predictions Std 19.6351 +trainer/V1 Predictions Max 1.68673 +trainer/V1 Predictions Min -86.3244 +trainer/VF Loss 0.0730721 +expl/num steps total 382000 +expl/num paths total 443 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.143923 +expl/Actions Std 0.824246 +expl/Actions Max 2.27205 +expl/Actions Min -2.18838 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 363972 +eval/num paths total 382 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0271321 +eval/Actions Std 0.725349 +eval/Actions Max 0.998648 +eval/Actions Min -0.999493 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.86009e-06 +time/evaluation sampling (s) 4.56291 +time/exploration sampling (s) 6.46817 +time/logging (s) 0.0119904 +time/saving (s) 0.0158498 +time/training (s) 18.7751 +time/epoch (s) 29.8341 +time/total (s) 8045.14 +Epoch -619 +------------------------------ ---------------- +2022-05-15 20:17:01.973626 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -618 finished +------------------------------ ---------------- +epoch -618 +replay_buffer/size 999047 +trainer/num train calls 383000 +trainer/QF1 Loss 0.601471 +trainer/QF2 Loss 0.507721 +trainer/Policy Loss 22.2977 +trainer/Q1 Predictions Mean -71.7436 +trainer/Q1 Predictions Std 18.0554 +trainer/Q1 Predictions Max -0.88708 +trainer/Q1 Predictions Min -86.4827 +trainer/Q2 Predictions Mean -71.7477 +trainer/Q2 Predictions Std 18.0005 +trainer/Q2 Predictions Max -1.24977 +trainer/Q2 Predictions Min -86.5349 +trainer/Q Targets Mean -71.8411 +trainer/Q Targets Std 18.0205 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.956 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0274531 +trainer/policy/mean Std 0.717703 +trainer/policy/mean Max 0.998915 +trainer/policy/mean Min -0.997503 +trainer/policy/std Mean 0.424501 +trainer/policy/std Std 0.0205226 +trainer/policy/std Max 0.444472 +trainer/policy/std Min 0.390017 +trainer/Advantage Weights Mean 4.90548 +trainer/Advantage Weights Std 17.486 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.59269e-17 +trainer/Advantage Score Mean -0.284596 +trainer/Advantage Score Std 0.523916 +trainer/Advantage Score Max 1.45813 +trainer/Advantage Score Min -3.74225 +trainer/V1 Predictions Mean -71.601 +trainer/V1 Predictions Std 18.0727 +trainer/V1 Predictions Max 0.0854734 +trainer/V1 Predictions Min -86.7978 +trainer/VF Loss 0.057987 +expl/num steps total 383000 +expl/num paths total 444 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.000265913 +expl/Actions Std 0.845212 +expl/Actions Max 2.30561 +expl/Actions Min -2.29252 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 364972 +eval/num paths total 383 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.159356 +eval/Actions Std 0.708012 +eval/Actions Max 0.99956 +eval/Actions Min -0.998427 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.38958e-05 +time/evaluation sampling (s) 4.47422 +time/exploration sampling (s) 6.93785 +time/logging (s) 0.00815736 +time/saving (s) 0.0122577 +time/training (s) 18.7017 +time/epoch (s) 30.1342 +time/total (s) 8075.28 +Epoch -618 +------------------------------ ---------------- +2022-05-15 20:17:32.379317 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -617 finished +------------------------------ ---------------- +epoch -617 +replay_buffer/size 999047 +trainer/num train calls 384000 +trainer/QF1 Loss 0.615013 +trainer/QF2 Loss 0.671887 +trainer/Policy Loss 16.9046 +trainer/Q1 Predictions Mean -72.5693 +trainer/Q1 Predictions Std 19.5507 +trainer/Q1 Predictions Max -1.15198 +trainer/Q1 Predictions Min -87.23 +trainer/Q2 Predictions Mean -72.6188 +trainer/Q2 Predictions Std 19.5981 +trainer/Q2 Predictions Max -0.339151 +trainer/Q2 Predictions Min -87.254 +trainer/Q Targets Mean -72.348 +trainer/Q Targets Std 19.4678 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0259 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.01488 +trainer/policy/mean Std 0.729394 +trainer/policy/mean Max 0.999182 +trainer/policy/mean Min -0.999615 +trainer/policy/std Mean 0.426193 +trainer/policy/std Std 0.0206906 +trainer/policy/std Max 0.446652 +trainer/policy/std Min 0.388013 +trainer/Advantage Weights Mean 4.52006 +trainer/Advantage Weights Std 19.0748 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.60734e-21 +trainer/Advantage Score Mean -0.384399 +trainer/Advantage Score Std 0.612731 +trainer/Advantage Score Max 2.54153 +trainer/Advantage Score Min -4.66302 +trainer/V1 Predictions Mean -72.1073 +trainer/V1 Predictions Std 19.6083 +trainer/V1 Predictions Max -0.889487 +trainer/V1 Predictions Min -86.8808 +trainer/VF Loss 0.0864844 +expl/num steps total 384000 +expl/num paths total 446 +expl/path length Mean 500 +expl/path length Std 375 +expl/path length Max 875 +expl/path length Min 125 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0305346 +expl/Actions Std 0.831559 +expl/Actions Max 2.4112 +expl/Actions Min -2.54635 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 365972 +eval/num paths total 384 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0622089 +eval/Actions Std 0.692945 +eval/Actions Max 0.999467 +eval/Actions Min -0.999502 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.0958e-05 +time/evaluation sampling (s) 4.95898 +time/exploration sampling (s) 6.4966 +time/logging (s) 0.0115884 +time/saving (s) 0.0166275 +time/training (s) 18.9149 +time/epoch (s) 30.3988 +time/total (s) 8105.69 +Epoch -617 +------------------------------ ---------------- +2022-05-15 20:18:03.046918 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -616 finished +------------------------------ ---------------- +epoch -616 +replay_buffer/size 999047 +trainer/num train calls 385000 +trainer/QF1 Loss 0.652754 +trainer/QF2 Loss 0.580772 +trainer/Policy Loss 37.9128 +trainer/Q1 Predictions Mean -73.3352 +trainer/Q1 Predictions Std 16.3448 +trainer/Q1 Predictions Max -0.404913 +trainer/Q1 Predictions Min -86.3902 +trainer/Q2 Predictions Mean -73.3677 +trainer/Q2 Predictions Std 16.3426 +trainer/Q2 Predictions Max -0.429814 +trainer/Q2 Predictions Min -86.7478 +trainer/Q Targets Mean -73.206 +trainer/Q Targets Std 16.4731 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6234 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00266935 +trainer/policy/mean Std 0.717951 +trainer/policy/mean Max 0.998342 +trainer/policy/mean Min -0.995351 +trainer/policy/std Mean 0.424244 +trainer/policy/std Std 0.0201584 +trainer/policy/std Max 0.446334 +trainer/policy/std Min 0.388966 +trainer/Advantage Weights Mean 10.396 +trainer/Advantage Weights Std 25.5736 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.05722e-11 +trainer/Advantage Score Mean -0.170355 +trainer/Advantage Score Std 0.530767 +trainer/Advantage Score Max 1.60394 +trainer/Advantage Score Min -2.52728 +trainer/V1 Predictions Mean -72.9578 +trainer/V1 Predictions Std 16.5713 +trainer/V1 Predictions Max -0.185719 +trainer/V1 Predictions Min -86.4204 +trainer/VF Loss 0.0695919 +expl/num steps total 385000 +expl/num paths total 447 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.261056 +expl/Actions Std 0.874376 +expl/Actions Max 2.4586 +expl/Actions Min -2.45905 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 366750 +eval/num paths total 385 +eval/path length Mean 778 +eval/path length Std 0 +eval/path length Max 778 +eval/path length Min 778 +eval/Rewards Mean 0.00128535 +eval/Rewards Std 0.0358287 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0377804 +eval/Actions Std 0.72723 +eval/Actions Max 0.999926 +eval/Actions Min -0.999575 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.12387e-05 +time/evaluation sampling (s) 4.74268 +time/exploration sampling (s) 7.11259 +time/logging (s) 0.00812103 +time/saving (s) 0.0109521 +time/training (s) 18.7779 +time/epoch (s) 30.6523 +time/total (s) 8136.35 +Epoch -616 +------------------------------ ---------------- +2022-05-15 20:18:33.246813 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -615 finished +------------------------------ ---------------- +epoch -615 +replay_buffer/size 999047 +trainer/num train calls 386000 +trainer/QF1 Loss 0.918143 +trainer/QF2 Loss 0.966109 +trainer/Policy Loss 9.47318 +trainer/Q1 Predictions Mean -71.567 +trainer/Q1 Predictions Std 18.6857 +trainer/Q1 Predictions Max -1.02046 +trainer/Q1 Predictions Min -87.0184 +trainer/Q2 Predictions Mean -71.5747 +trainer/Q2 Predictions Std 18.6406 +trainer/Q2 Predictions Max -0.696911 +trainer/Q2 Predictions Min -87.0643 +trainer/Q Targets Mean -71.4766 +trainer/Q Targets Std 18.9824 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8898 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00633755 +trainer/policy/mean Std 0.721771 +trainer/policy/mean Max 0.99955 +trainer/policy/mean Min -0.999399 +trainer/policy/std Mean 0.423714 +trainer/policy/std Std 0.0196447 +trainer/policy/std Max 0.445695 +trainer/policy/std Min 0.391621 +trainer/Advantage Weights Mean 2.73078 +trainer/Advantage Weights Std 13.7325 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.18572e-20 +trainer/Advantage Score Mean -0.532042 +trainer/Advantage Score Std 0.671524 +trainer/Advantage Score Max 1.39746 +trainer/Advantage Score Min -4.39493 +trainer/V1 Predictions Mean -71.2479 +trainer/V1 Predictions Std 19.0711 +trainer/V1 Predictions Max -0.546033 +trainer/V1 Predictions Min -86.9082 +trainer/VF Loss 0.0856195 +expl/num steps total 386000 +expl/num paths total 448 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0730764 +expl/Actions Std 0.832028 +expl/Actions Max 2.35918 +expl/Actions Min -2.59004 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 367198 +eval/num paths total 386 +eval/path length Mean 448 +eval/path length Std 0 +eval/path length Max 448 +eval/path length Min 448 +eval/Rewards Mean 0.00223214 +eval/Rewards Std 0.0471928 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0501321 +eval/Actions Std 0.741256 +eval/Actions Max 0.999317 +eval/Actions Min -0.999165 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.29263e-05 +time/evaluation sampling (s) 4.80392 +time/exploration sampling (s) 6.68953 +time/logging (s) 0.00704324 +time/saving (s) 0.013392 +time/training (s) 18.6737 +time/epoch (s) 30.1876 +time/total (s) 8166.54 +Epoch -615 +------------------------------ ---------------- +2022-05-15 20:19:03.097714 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -614 finished +------------------------------ ---------------- +epoch -614 +replay_buffer/size 999047 +trainer/num train calls 387000 +trainer/QF1 Loss 0.770802 +trainer/QF2 Loss 0.821651 +trainer/Policy Loss 12.1902 +trainer/Q1 Predictions Mean -74.9228 +trainer/Q1 Predictions Std 16.1384 +trainer/Q1 Predictions Max -0.251712 +trainer/Q1 Predictions Min -86.8178 +trainer/Q2 Predictions Mean -74.9752 +trainer/Q2 Predictions Std 16.1357 +trainer/Q2 Predictions Max -0.125806 +trainer/Q2 Predictions Min -86.8609 +trainer/Q Targets Mean -74.5322 +trainer/Q Targets Std 16.0489 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3564 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0037796 +trainer/policy/mean Std 0.723715 +trainer/policy/mean Max 0.999409 +trainer/policy/mean Min -0.996344 +trainer/policy/std Mean 0.423415 +trainer/policy/std Std 0.0202614 +trainer/policy/std Max 0.444694 +trainer/policy/std Min 0.390501 +trainer/Advantage Weights Mean 2.57305 +trainer/Advantage Weights Std 12.9186 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.72003e-18 +trainer/Advantage Score Mean -0.427411 +trainer/Advantage Score Std 0.554128 +trainer/Advantage Score Max 1.8302 +trainer/Advantage Score Min -3.94027 +trainer/V1 Predictions Mean -74.2982 +trainer/V1 Predictions Std 16.1716 +trainer/V1 Predictions Max 0.810909 +trainer/V1 Predictions Min -86.3945 +trainer/VF Loss 0.0677058 +expl/num steps total 387000 +expl/num paths total 449 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0922965 +expl/Actions Std 0.809342 +expl/Actions Max 2.25635 +expl/Actions Min -2.34517 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 368198 +eval/num paths total 387 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.180857 +eval/Actions Std 0.611031 +eval/Actions Max 0.999838 +eval/Actions Min -0.997444 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.64264e-06 +time/evaluation sampling (s) 4.82525 +time/exploration sampling (s) 5.67245 +time/logging (s) 0.0125024 +time/saving (s) 0.0182116 +time/training (s) 19.3145 +time/epoch (s) 29.8429 +time/total (s) 8196.39 +Epoch -614 +------------------------------ ---------------- +2022-05-15 20:19:32.583062 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -613 finished +------------------------------ ---------------- +epoch -613 +replay_buffer/size 999047 +trainer/num train calls 388000 +trainer/QF1 Loss 0.711294 +trainer/QF2 Loss 0.721637 +trainer/Policy Loss 24.5511 +trainer/Q1 Predictions Mean -71.4361 +trainer/Q1 Predictions Std 19.1279 +trainer/Q1 Predictions Max -0.625253 +trainer/Q1 Predictions Min -87.9252 +trainer/Q2 Predictions Mean -71.3807 +trainer/Q2 Predictions Std 19.1447 +trainer/Q2 Predictions Max -0.601598 +trainer/Q2 Predictions Min -87.9426 +trainer/Q Targets Mean -71.275 +trainer/Q Targets Std 18.7706 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5353 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0163387 +trainer/policy/mean Std 0.728529 +trainer/policy/mean Max 0.998892 +trainer/policy/mean Min -0.99821 +trainer/policy/std Mean 0.423163 +trainer/policy/std Std 0.0209668 +trainer/policy/std Max 0.446563 +trainer/policy/std Min 0.387756 +trainer/Advantage Weights Mean 5.0891 +trainer/Advantage Weights Std 19.6599 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.01821e-09 +trainer/Advantage Score Mean -0.331469 +trainer/Advantage Score Std 0.48634 +trainer/Advantage Score Max 1.49508 +trainer/Advantage Score Min -2.07052 +trainer/V1 Predictions Mean -70.9334 +trainer/V1 Predictions Std 18.9393 +trainer/V1 Predictions Max -0.220649 +trainer/V1 Predictions Min -86.7341 +trainer/VF Loss 0.0638406 +expl/num steps total 388000 +expl/num paths total 450 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0271294 +expl/Actions Std 0.812847 +expl/Actions Max 2.20618 +expl/Actions Min -2.69922 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 369198 +eval/num paths total 388 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0604389 +eval/Actions Std 0.723187 +eval/Actions Max 0.999918 +eval/Actions Min -0.999066 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.97513e-06 +time/evaluation sampling (s) 4.16989 +time/exploration sampling (s) 6.00495 +time/logging (s) 0.011406 +time/saving (s) 0.0147937 +time/training (s) 19.2686 +time/epoch (s) 29.4697 +time/total (s) 8225.87 +Epoch -613 +------------------------------ ---------------- +2022-05-15 20:20:02.243219 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -612 finished +------------------------------ ---------------- +epoch -612 +replay_buffer/size 999047 +trainer/num train calls 389000 +trainer/QF1 Loss 1.48498 +trainer/QF2 Loss 1.50248 +trainer/Policy Loss 5.46618 +trainer/Q1 Predictions Mean -73.9579 +trainer/Q1 Predictions Std 18.4259 +trainer/Q1 Predictions Max -0.377881 +trainer/Q1 Predictions Min -87.6476 +trainer/Q2 Predictions Mean -73.9344 +trainer/Q2 Predictions Std 18.4705 +trainer/Q2 Predictions Max -0.367243 +trainer/Q2 Predictions Min -87.494 +trainer/Q Targets Mean -73.0286 +trainer/Q Targets Std 18.5211 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9231 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0025704 +trainer/policy/mean Std 0.722397 +trainer/policy/mean Max 0.999765 +trainer/policy/mean Min -0.996603 +trainer/policy/std Mean 0.423352 +trainer/policy/std Std 0.0207493 +trainer/policy/std Max 0.443671 +trainer/policy/std Min 0.389074 +trainer/Advantage Weights Mean 2.2241 +trainer/Advantage Weights Std 12.8528 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.24264e-15 +trainer/Advantage Score Mean -0.569809 +trainer/Advantage Score Std 0.514594 +trainer/Advantage Score Max 0.841358 +trainer/Advantage Score Min -3.33624 +trainer/V1 Predictions Mean -72.8012 +trainer/V1 Predictions Std 18.5887 +trainer/V1 Predictions Max 0.38399 +trainer/V1 Predictions Min -86.5871 +trainer/VF Loss 0.0655364 +expl/num steps total 389000 +expl/num paths total 452 +expl/path length Mean 500 +expl/path length Std 296 +expl/path length Max 796 +expl/path length Min 204 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00271595 +expl/Actions Std 0.812261 +expl/Actions Max 2.18147 +expl/Actions Min -2.32521 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 369968 +eval/num paths total 389 +eval/path length Mean 770 +eval/path length Std 0 +eval/path length Max 770 +eval/path length Min 770 +eval/Rewards Mean 0.0012987 +eval/Rewards Std 0.0360141 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00527715 +eval/Actions Std 0.730193 +eval/Actions Max 0.99955 +eval/Actions Min -0.999469 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.21207e-05 +time/evaluation sampling (s) 4.66805 +time/exploration sampling (s) 6.30518 +time/logging (s) 0.01143 +time/saving (s) 0.0179559 +time/training (s) 18.6439 +time/epoch (s) 29.6465 +time/total (s) 8255.52 +Epoch -612 +------------------------------ ---------------- +2022-05-15 20:20:33.432787 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -611 finished +------------------------------ ---------------- +epoch -611 +replay_buffer/size 999047 +trainer/num train calls 390000 +trainer/QF1 Loss 1.5891 +trainer/QF2 Loss 1.55659 +trainer/Policy Loss 24.2452 +trainer/Q1 Predictions Mean -71.7983 +trainer/Q1 Predictions Std 19.675 +trainer/Q1 Predictions Max -4.64848 +trainer/Q1 Predictions Min -86.3054 +trainer/Q2 Predictions Mean -71.7963 +trainer/Q2 Predictions Std 19.6576 +trainer/Q2 Predictions Max -4.31176 +trainer/Q2 Predictions Min -86.5777 +trainer/Q Targets Mean -71.4102 +trainer/Q Targets Std 19.943 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.549 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0128314 +trainer/policy/mean Std 0.715382 +trainer/policy/mean Max 0.999483 +trainer/policy/mean Min -0.99834 +trainer/policy/std Mean 0.425041 +trainer/policy/std Std 0.0190694 +trainer/policy/std Max 0.444251 +trainer/policy/std Min 0.392125 +trainer/Advantage Weights Mean 5.93424 +trainer/Advantage Weights Std 21.179 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.46761e-12 +trainer/Advantage Score Mean -0.378968 +trainer/Advantage Score Std 0.565748 +trainer/Advantage Score Max 1.30087 +trainer/Advantage Score Min -2.61342 +trainer/V1 Predictions Mean -71.2694 +trainer/V1 Predictions Std 19.8644 +trainer/V1 Predictions Max -2.38933 +trainer/V1 Predictions Min -85.2023 +trainer/VF Loss 0.0696312 +expl/num steps total 390000 +expl/num paths total 454 +expl/path length Mean 500 +expl/path length Std 103 +expl/path length Max 603 +expl/path length Min 397 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0321845 +expl/Actions Std 0.824843 +expl/Actions Max 2.66529 +expl/Actions Min -2.36849 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 370968 +eval/num paths total 390 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.02544 +eval/Actions Std 0.678192 +eval/Actions Max 0.998941 +eval/Actions Min -0.999514 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.1107e-05 +time/evaluation sampling (s) 5.13987 +time/exploration sampling (s) 6.81698 +time/logging (s) 0.00910247 +time/saving (s) 0.01198 +time/training (s) 19.1949 +time/epoch (s) 31.1728 +time/total (s) 8286.7 +Epoch -611 +------------------------------ ---------------- +2022-05-15 20:21:04.005926 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -610 finished +------------------------------ ---------------- +epoch -610 +replay_buffer/size 999047 +trainer/num train calls 391000 +trainer/QF1 Loss 1.0796 +trainer/QF2 Loss 1.06419 +trainer/Policy Loss 7.16242 +trainer/Q1 Predictions Mean -73.6069 +trainer/Q1 Predictions Std 17.1907 +trainer/Q1 Predictions Max -2.51262 +trainer/Q1 Predictions Min -86.8768 +trainer/Q2 Predictions Mean -73.5748 +trainer/Q2 Predictions Std 17.2601 +trainer/Q2 Predictions Max -3.07812 +trainer/Q2 Predictions Min -86.5968 +trainer/Q Targets Mean -73.2651 +trainer/Q Targets Std 17.3002 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3448 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00947921 +trainer/policy/mean Std 0.7257 +trainer/policy/mean Max 0.999568 +trainer/policy/mean Min -0.998785 +trainer/policy/std Mean 0.423151 +trainer/policy/std Std 0.0205233 +trainer/policy/std Max 0.445363 +trainer/policy/std Min 0.386765 +trainer/Advantage Weights Mean 1.7423 +trainer/Advantage Weights Std 11.1193 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.71948e-19 +trainer/Advantage Score Mean -0.598501 +trainer/Advantage Score Std 0.624389 +trainer/Advantage Score Max 1.07834 +trainer/Advantage Score Min -4.24355 +trainer/V1 Predictions Mean -72.9634 +trainer/V1 Predictions Std 17.5328 +trainer/V1 Predictions Max -0.748964 +trainer/V1 Predictions Min -86.2807 +trainer/VF Loss 0.0820113 +expl/num steps total 391000 +expl/num paths total 455 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0423392 +expl/Actions Std 0.837331 +expl/Actions Max 2.43921 +expl/Actions Min -2.37566 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 371968 +eval/num paths total 391 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0241395 +eval/Actions Std 0.452421 +eval/Actions Max 0.999681 +eval/Actions Min -0.998674 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.88618e-06 +time/evaluation sampling (s) 4.98403 +time/exploration sampling (s) 6.68722 +time/logging (s) 0.0128376 +time/saving (s) 0.0170239 +time/training (s) 18.865 +time/epoch (s) 30.5662 +time/total (s) 8317.27 +Epoch -610 +------------------------------ ---------------- +2022-05-15 20:21:34.300341 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -609 finished +------------------------------ ---------------- +epoch -609 +replay_buffer/size 999047 +trainer/num train calls 392000 +trainer/QF1 Loss 10.8408 +trainer/QF2 Loss 11.2234 +trainer/Policy Loss 21.2334 +trainer/Q1 Predictions Mean -71.7463 +trainer/Q1 Predictions Std 19.1886 +trainer/Q1 Predictions Max -0.943727 +trainer/Q1 Predictions Min -86.4023 +trainer/Q2 Predictions Mean -71.7649 +trainer/Q2 Predictions Std 19.2324 +trainer/Q2 Predictions Max -1.08167 +trainer/Q2 Predictions Min -86.4349 +trainer/Q Targets Mean -71.4954 +trainer/Q Targets Std 19.2605 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2341 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0174235 +trainer/policy/mean Std 0.725045 +trainer/policy/mean Max 0.999176 +trainer/policy/mean Min -0.998616 +trainer/policy/std Mean 0.423575 +trainer/policy/std Std 0.0205766 +trainer/policy/std Max 0.443307 +trainer/policy/std Min 0.387868 +trainer/Advantage Weights Mean 4.97902 +trainer/Advantage Weights Std 19.1703 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.02895e-15 +trainer/Advantage Score Mean -0.33964 +trainer/Advantage Score Std 0.550981 +trainer/Advantage Score Max 1.14588 +trainer/Advantage Score Min -3.29236 +trainer/V1 Predictions Mean -71.4541 +trainer/V1 Predictions Std 19.3022 +trainer/V1 Predictions Max -0.683387 +trainer/V1 Predictions Min -86.029 +trainer/VF Loss 0.0616038 +expl/num steps total 392000 +expl/num paths total 457 +expl/path length Mean 500 +expl/path length Std 245 +expl/path length Max 745 +expl/path length Min 255 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0173242 +expl/Actions Std 0.8201 +expl/Actions Max 2.33846 +expl/Actions Min -2.54965 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 372968 +eval/num paths total 392 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0273746 +eval/Actions Std 0.738282 +eval/Actions Max 0.999447 +eval/Actions Min -0.99478 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0916e-05 +time/evaluation sampling (s) 4.46703 +time/exploration sampling (s) 6.63889 +time/logging (s) 0.0109886 +time/saving (s) 0.0160108 +time/training (s) 19.1488 +time/epoch (s) 30.2817 +time/total (s) 8347.56 +Epoch -609 +------------------------------ ---------------- +2022-05-15 20:22:04.924511 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -608 finished +------------------------------ ---------------- +epoch -608 +replay_buffer/size 999047 +trainer/num train calls 393000 +trainer/QF1 Loss 0.623624 +trainer/QF2 Loss 0.693606 +trainer/Policy Loss 5.16856 +trainer/Q1 Predictions Mean -73.8438 +trainer/Q1 Predictions Std 17.3095 +trainer/Q1 Predictions Max -0.685405 +trainer/Q1 Predictions Min -85.9429 +trainer/Q2 Predictions Mean -73.8724 +trainer/Q2 Predictions Std 17.2882 +trainer/Q2 Predictions Max -0.837644 +trainer/Q2 Predictions Min -86.3445 +trainer/Q Targets Mean -73.9021 +trainer/Q Targets Std 17.5717 +trainer/Q Targets Max 0.843393 +trainer/Q Targets Min -86.139 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0153221 +trainer/policy/mean Std 0.716818 +trainer/policy/mean Max 0.998654 +trainer/policy/mean Min -0.999656 +trainer/policy/std Mean 0.425046 +trainer/policy/std Std 0.0210072 +trainer/policy/std Max 0.447553 +trainer/policy/std Min 0.388682 +trainer/Advantage Weights Mean 2.1402 +trainer/Advantage Weights Std 10.8821 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.66707e-19 +trainer/Advantage Score Mean -0.454913 +trainer/Advantage Score Std 0.630854 +trainer/Advantage Score Max 0.860016 +trainer/Advantage Score Min -4.32381 +trainer/V1 Predictions Mean -73.7137 +trainer/V1 Predictions Std 17.6228 +trainer/V1 Predictions Max 0.491554 +trainer/V1 Predictions Min -86.0509 +trainer/VF Loss 0.0678218 +expl/num steps total 393000 +expl/num paths total 458 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.151034 +expl/Actions Std 0.893771 +expl/Actions Max 2.40758 +expl/Actions Min -2.55268 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 373968 +eval/num paths total 393 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0483252 +eval/Actions Std 0.749367 +eval/Actions Max 0.9995 +eval/Actions Min -0.998744 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.1777e-05 +time/evaluation sampling (s) 4.91359 +time/exploration sampling (s) 6.4661 +time/logging (s) 0.0098817 +time/saving (s) 0.0120532 +time/training (s) 19.2075 +time/epoch (s) 30.6091 +time/total (s) 8378.18 +Epoch -608 +------------------------------ ---------------- +2022-05-15 20:22:35.502702 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -607 finished +------------------------------ ---------------- +epoch -607 +replay_buffer/size 999047 +trainer/num train calls 394000 +trainer/QF1 Loss 0.811821 +trainer/QF2 Loss 0.771552 +trainer/Policy Loss 24.6054 +trainer/Q1 Predictions Mean -69.589 +trainer/Q1 Predictions Std 20.7384 +trainer/Q1 Predictions Max -0.370577 +trainer/Q1 Predictions Min -86.5602 +trainer/Q2 Predictions Mean -69.6807 +trainer/Q2 Predictions Std 20.7187 +trainer/Q2 Predictions Max -0.442348 +trainer/Q2 Predictions Min -86.1314 +trainer/Q Targets Mean -69.783 +trainer/Q Targets Std 20.8708 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9282 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0171201 +trainer/policy/mean Std 0.72831 +trainer/policy/mean Max 0.999375 +trainer/policy/mean Min -0.998768 +trainer/policy/std Mean 0.422314 +trainer/policy/std Std 0.0211555 +trainer/policy/std Max 0.447915 +trainer/policy/std Min 0.386786 +trainer/Advantage Weights Mean 5.00751 +trainer/Advantage Weights Std 19.3791 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.58097e-22 +trainer/Advantage Score Mean -0.427178 +trainer/Advantage Score Std 0.75241 +trainer/Advantage Score Max 0.992218 +trainer/Advantage Score Min -4.93812 +trainer/V1 Predictions Mean -69.429 +trainer/V1 Predictions Std 21.0629 +trainer/V1 Predictions Max 0.63796 +trainer/V1 Predictions Min -86.6639 +trainer/VF Loss 0.0927627 +expl/num steps total 394000 +expl/num paths total 460 +expl/path length Mean 500 +expl/path length Std 294 +expl/path length Max 794 +expl/path length Min 206 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0416215 +expl/Actions Std 0.821928 +expl/Actions Max 2.32509 +expl/Actions Min -2.23157 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 374968 +eval/num paths total 394 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0896552 +eval/Actions Std 0.756911 +eval/Actions Max 0.999775 +eval/Actions Min -0.997767 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.26092e-05 +time/evaluation sampling (s) 4.93394 +time/exploration sampling (s) 6.47212 +time/logging (s) 0.0115767 +time/saving (s) 0.0158806 +time/training (s) 19.1337 +time/epoch (s) 30.5672 +time/total (s) 8408.75 +Epoch -607 +------------------------------ ---------------- +2022-05-15 20:23:05.978376 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -606 finished +------------------------------ ---------------- +epoch -606 +replay_buffer/size 999047 +trainer/num train calls 395000 +trainer/QF1 Loss 0.85901 +trainer/QF2 Loss 0.856635 +trainer/Policy Loss 31.6074 +trainer/Q1 Predictions Mean -71.7175 +trainer/Q1 Predictions Std 19.1323 +trainer/Q1 Predictions Max -0.691196 +trainer/Q1 Predictions Min -86.274 +trainer/Q2 Predictions Mean -71.7243 +trainer/Q2 Predictions Std 19.1022 +trainer/Q2 Predictions Max -0.577624 +trainer/Q2 Predictions Min -86.2371 +trainer/Q Targets Mean -71.478 +trainer/Q Targets Std 19.3465 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5638 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000472361 +trainer/policy/mean Std 0.718596 +trainer/policy/mean Max 0.999922 +trainer/policy/mean Min -0.998027 +trainer/policy/std Mean 0.423476 +trainer/policy/std Std 0.0210869 +trainer/policy/std Max 0.446786 +trainer/policy/std Min 0.388839 +trainer/Advantage Weights Mean 5.4417 +trainer/Advantage Weights Std 20.7327 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.37893e-15 +trainer/Advantage Score Mean -0.360243 +trainer/Advantage Score Std 0.599296 +trainer/Advantage Score Max 1.52913 +trainer/Advantage Score Min -3.26858 +trainer/V1 Predictions Mean -71.2338 +trainer/V1 Predictions Std 19.4542 +trainer/V1 Predictions Max 1.62414 +trainer/V1 Predictions Min -86.4361 +trainer/VF Loss 0.0838095 +expl/num steps total 395000 +expl/num paths total 462 +expl/path length Mean 500 +expl/path length Std 321 +expl/path length Max 821 +expl/path length Min 179 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0257855 +expl/Actions Std 0.813587 +expl/Actions Max 2.33287 +expl/Actions Min -2.42264 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 375822 +eval/num paths total 395 +eval/path length Mean 854 +eval/path length Std 0 +eval/path length Max 854 +eval/path length Min 854 +eval/Rewards Mean 0.00117096 +eval/Rewards Std 0.0341993 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0462786 +eval/Actions Std 0.749234 +eval/Actions Max 0.999887 +eval/Actions Min -0.999395 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.33789e-05 +time/evaluation sampling (s) 4.91954 +time/exploration sampling (s) 6.37648 +time/logging (s) 0.00797621 +time/saving (s) 0.0133053 +time/training (s) 19.1412 +time/epoch (s) 30.4585 +time/total (s) 8439.22 +Epoch -606 +------------------------------ ---------------- +2022-05-15 20:23:36.328744 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -605 finished +------------------------------ ---------------- +epoch -605 +replay_buffer/size 999047 +trainer/num train calls 396000 +trainer/QF1 Loss 1.01882 +trainer/QF2 Loss 1.17339 +trainer/Policy Loss 66.0064 +trainer/Q1 Predictions Mean -73.7565 +trainer/Q1 Predictions Std 15.5507 +trainer/Q1 Predictions Max -6.57302 +trainer/Q1 Predictions Min -86.1612 +trainer/Q2 Predictions Mean -73.6736 +trainer/Q2 Predictions Std 15.6117 +trainer/Q2 Predictions Max -4.88253 +trainer/Q2 Predictions Min -86.2966 +trainer/Q Targets Mean -74.3872 +trainer/Q Targets Std 15.65 +trainer/Q Targets Max -6.65148 +trainer/Q Targets Min -87.0581 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00941359 +trainer/policy/mean Std 0.713924 +trainer/policy/mean Max 0.999717 +trainer/policy/mean Min -0.999591 +trainer/policy/std Mean 0.423021 +trainer/policy/std Std 0.0198376 +trainer/policy/std Max 0.443405 +trainer/policy/std Min 0.389812 +trainer/Advantage Weights Mean 15.3922 +trainer/Advantage Weights Std 29.7461 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.24318e-07 +trainer/Advantage Score Mean -0.0792423 +trainer/Advantage Score Std 0.468164 +trainer/Advantage Score Max 2.37875 +trainer/Advantage Score Min -1.59004 +trainer/V1 Predictions Mean -74.2045 +trainer/V1 Predictions Std 15.6515 +trainer/V1 Predictions Max -6.96198 +trainer/V1 Predictions Min -87.0532 +trainer/VF Loss 0.0858205 +expl/num steps total 396000 +expl/num paths total 464 +expl/path length Mean 500 +expl/path length Std 382 +expl/path length Max 882 +expl/path length Min 118 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0445159 +expl/Actions Std 0.847952 +expl/Actions Max 2.57174 +expl/Actions Min -2.31367 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 376822 +eval/num paths total 396 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.203759 +eval/Actions Std 0.633827 +eval/Actions Max 0.999366 +eval/Actions Min -0.997647 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.31433e-05 +time/evaluation sampling (s) 4.9617 +time/exploration sampling (s) 6.21106 +time/logging (s) 0.0120706 +time/saving (s) 0.0155633 +time/training (s) 19.1439 +time/epoch (s) 30.3443 +time/total (s) 8469.57 +Epoch -605 +------------------------------ ---------------- +2022-05-15 20:24:06.914381 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -604 finished +------------------------------ ---------------- +epoch -604 +replay_buffer/size 999047 +trainer/num train calls 397000 +trainer/QF1 Loss 0.593809 +trainer/QF2 Loss 0.811366 +trainer/Policy Loss 5.9292 +trainer/Q1 Predictions Mean -73.0764 +trainer/Q1 Predictions Std 17.9357 +trainer/Q1 Predictions Max -2.01367 +trainer/Q1 Predictions Min -86.5638 +trainer/Q2 Predictions Mean -73.1713 +trainer/Q2 Predictions Std 17.949 +trainer/Q2 Predictions Max -1.84385 +trainer/Q2 Predictions Min -86.8269 +trainer/Q Targets Mean -72.8263 +trainer/Q Targets Std 17.7764 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0916 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0127241 +trainer/policy/mean Std 0.724538 +trainer/policy/mean Max 0.998792 +trainer/policy/mean Min -0.997375 +trainer/policy/std Mean 0.423276 +trainer/policy/std Std 0.0206696 +trainer/policy/std Max 0.442895 +trainer/policy/std Min 0.38797 +trainer/Advantage Weights Mean 1.53857 +trainer/Advantage Weights Std 11.2138 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.97575e-20 +trainer/Advantage Score Mean -0.639228 +trainer/Advantage Score Std 0.613083 +trainer/Advantage Score Max 0.856232 +trainer/Advantage Score Min -4.4264 +trainer/V1 Predictions Mean -72.5328 +trainer/V1 Predictions Std 18.0439 +trainer/V1 Predictions Max -1.2331 +trainer/V1 Predictions Min -86.3082 +trainer/VF Loss 0.0846887 +expl/num steps total 397000 +expl/num paths total 466 +expl/path length Mean 500 +expl/path length Std 466 +expl/path length Max 966 +expl/path length Min 34 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0126369 +expl/Actions Std 0.817597 +expl/Actions Max 2.19828 +expl/Actions Min -2.31668 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 377822 +eval/num paths total 397 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.212538 +eval/Actions Std 0.824482 +eval/Actions Max 0.999971 +eval/Actions Min -0.99982 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.7523e-06 +time/evaluation sampling (s) 4.77781 +time/exploration sampling (s) 6.22834 +time/logging (s) 0.00775062 +time/saving (s) 0.0137034 +time/training (s) 19.5416 +time/epoch (s) 30.5692 +time/total (s) 8500.14 +Epoch -604 +------------------------------ ---------------- +2022-05-15 20:24:37.080354 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -603 finished +------------------------------ ---------------- +epoch -603 +replay_buffer/size 999047 +trainer/num train calls 398000 +trainer/QF1 Loss 0.552508 +trainer/QF2 Loss 0.473019 +trainer/Policy Loss 12.7679 +trainer/Q1 Predictions Mean -74.8419 +trainer/Q1 Predictions Std 14.905 +trainer/Q1 Predictions Max -0.616753 +trainer/Q1 Predictions Min -86.8897 +trainer/Q2 Predictions Mean -74.9447 +trainer/Q2 Predictions Std 14.9687 +trainer/Q2 Predictions Max -0.500848 +trainer/Q2 Predictions Min -86.7839 +trainer/Q Targets Mean -74.9589 +trainer/Q Targets Std 14.8315 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8452 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0112219 +trainer/policy/mean Std 0.707204 +trainer/policy/mean Max 0.998836 +trainer/policy/mean Min -0.994763 +trainer/policy/std Mean 0.42257 +trainer/policy/std Std 0.0217867 +trainer/policy/std Max 0.445816 +trainer/policy/std Min 0.386301 +trainer/Advantage Weights Mean 2.54807 +trainer/Advantage Weights Std 13.3644 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.81228e-10 +trainer/Advantage Score Mean -0.322622 +trainer/Advantage Score Std 0.424723 +trainer/Advantage Score Max 1.06554 +trainer/Advantage Score Min -2.09702 +trainer/V1 Predictions Mean -74.722 +trainer/V1 Predictions Std 15.0241 +trainer/V1 Predictions Max -0.731235 +trainer/V1 Predictions Min -86.8372 +trainer/VF Loss 0.0375887 +expl/num steps total 398000 +expl/num paths total 467 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0757746 +expl/Actions Std 0.824596 +expl/Actions Max 2.29663 +expl/Actions Min -2.25038 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 378435 +eval/num paths total 398 +eval/path length Mean 613 +eval/path length Std 0 +eval/path length Max 613 +eval/path length Min 613 +eval/Rewards Mean 0.00163132 +eval/Rewards Std 0.0403567 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0305077 +eval/Actions Std 0.724796 +eval/Actions Max 0.999833 +eval/Actions Min -0.999666 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.11179e-06 +time/evaluation sampling (s) 4.49641 +time/exploration sampling (s) 6.94865 +time/logging (s) 0.0100844 +time/saving (s) 0.0166833 +time/training (s) 18.6866 +time/epoch (s) 30.1584 +time/total (s) 8530.3 +Epoch -603 +------------------------------ ---------------- +2022-05-15 20:25:08.108936 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -602 finished +------------------------------ ---------------- +epoch -602 +replay_buffer/size 999047 +trainer/num train calls 399000 +trainer/QF1 Loss 0.405924 +trainer/QF2 Loss 0.335877 +trainer/Policy Loss 10.214 +trainer/Q1 Predictions Mean -72.5688 +trainer/Q1 Predictions Std 17.4755 +trainer/Q1 Predictions Max -8.13104 +trainer/Q1 Predictions Min -86.476 +trainer/Q2 Predictions Mean -72.5209 +trainer/Q2 Predictions Std 17.4613 +trainer/Q2 Predictions Max -7.07462 +trainer/Q2 Predictions Min -86.4731 +trainer/Q Targets Mean -72.6239 +trainer/Q Targets Std 17.5248 +trainer/Q Targets Max -7.04601 +trainer/Q Targets Min -86.5523 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0139442 +trainer/policy/mean Std 0.70477 +trainer/policy/mean Max 0.997479 +trainer/policy/mean Min -0.998718 +trainer/policy/std Mean 0.423626 +trainer/policy/std Std 0.0214641 +trainer/policy/std Max 0.447084 +trainer/policy/std Min 0.386531 +trainer/Advantage Weights Mean 2.10026 +trainer/Advantage Weights Std 11.2863 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.47448e-18 +trainer/Advantage Score Mean -0.426232 +trainer/Advantage Score Std 0.581323 +trainer/Advantage Score Max 1.00059 +trainer/Advantage Score Min -4.05405 +trainer/V1 Predictions Mean -72.3424 +trainer/V1 Predictions Std 17.6425 +trainer/V1 Predictions Max -3.75473 +trainer/V1 Predictions Min -86.432 +trainer/VF Loss 0.0626065 +expl/num steps total 399000 +expl/num paths total 469 +expl/path length Mean 500 +expl/path length Std 446 +expl/path length Max 946 +expl/path length Min 54 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0328948 +expl/Actions Std 0.828518 +expl/Actions Max 2.19984 +expl/Actions Min -2.78744 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 379435 +eval/num paths total 399 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.138138 +eval/Actions Std 0.60253 +eval/Actions Max 0.999441 +eval/Actions Min -0.999166 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.2937e-05 +time/evaluation sampling (s) 5.12222 +time/exploration sampling (s) 6.65045 +time/logging (s) 0.0102996 +time/saving (s) 0.0142474 +time/training (s) 19.2201 +time/epoch (s) 31.0173 +time/total (s) 8561.33 +Epoch -602 +------------------------------ ---------------- +2022-05-15 20:25:37.953754 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -601 finished +------------------------------ ---------------- +epoch -601 +replay_buffer/size 999047 +trainer/num train calls 400000 +trainer/QF1 Loss 0.678083 +trainer/QF2 Loss 0.571584 +trainer/Policy Loss 7.38718 +trainer/Q1 Predictions Mean -72.286 +trainer/Q1 Predictions Std 18.887 +trainer/Q1 Predictions Max -1.62562 +trainer/Q1 Predictions Min -86.8833 +trainer/Q2 Predictions Mean -72.2566 +trainer/Q2 Predictions Std 18.9077 +trainer/Q2 Predictions Max -1.73041 +trainer/Q2 Predictions Min -86.8486 +trainer/Q Targets Mean -72.1024 +trainer/Q Targets Std 18.6928 +trainer/Q Targets Max -1.71173 +trainer/Q Targets Min -86.3227 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00955449 +trainer/policy/mean Std 0.722353 +trainer/policy/mean Max 0.998392 +trainer/policy/mean Min -0.999815 +trainer/policy/std Mean 0.424408 +trainer/policy/std Std 0.0207686 +trainer/policy/std Max 0.45019 +trainer/policy/std Min 0.387921 +trainer/Advantage Weights Mean 2.04848 +trainer/Advantage Weights Std 11.3188 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.01331e-22 +trainer/Advantage Score Mean -0.455834 +trainer/Advantage Score Std 0.544112 +trainer/Advantage Score Max 0.618317 +trainer/Advantage Score Min -5.06437 +trainer/V1 Predictions Mean -71.7701 +trainer/V1 Predictions Std 18.9919 +trainer/V1 Predictions Max -1.48955 +trainer/V1 Predictions Min -86.2643 +trainer/VF Loss 0.0557262 +expl/num steps total 400000 +expl/num paths total 471 +expl/path length Mean 500 +expl/path length Std 261 +expl/path length Max 761 +expl/path length Min 239 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0630166 +expl/Actions Std 0.835907 +expl/Actions Max 2.36643 +expl/Actions Min -2.15902 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 379945 +eval/num paths total 400 +eval/path length Mean 510 +eval/path length Std 0 +eval/path length Max 510 +eval/path length Min 510 +eval/Rewards Mean 0.00196078 +eval/Rewards Std 0.0442373 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0100684 +eval/Actions Std 0.728326 +eval/Actions Max 0.999477 +eval/Actions Min -0.999334 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.2869e-05 +time/evaluation sampling (s) 4.74412 +time/exploration sampling (s) 6.64657 +time/logging (s) 0.0067816 +time/saving (s) 0.0155345 +time/training (s) 18.4177 +time/epoch (s) 29.8308 +time/total (s) 8591.16 +Epoch -601 +------------------------------ ---------------- +2022-05-15 20:26:07.996931 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -600 finished +------------------------------ ---------------- +epoch -600 +replay_buffer/size 999047 +trainer/num train calls 401000 +trainer/QF1 Loss 1.18219 +trainer/QF2 Loss 1.20245 +trainer/Policy Loss 28.1872 +trainer/Q1 Predictions Mean -74.0032 +trainer/Q1 Predictions Std 16.8968 +trainer/Q1 Predictions Max -0.555609 +trainer/Q1 Predictions Min -86.7584 +trainer/Q2 Predictions Mean -73.9509 +trainer/Q2 Predictions Std 16.8789 +trainer/Q2 Predictions Max -0.542987 +trainer/Q2 Predictions Min -86.8194 +trainer/Q Targets Mean -74.3329 +trainer/Q Targets Std 16.7747 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0645 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00547546 +trainer/policy/mean Std 0.713842 +trainer/policy/mean Max 0.998618 +trainer/policy/mean Min -0.998486 +trainer/policy/std Mean 0.42495 +trainer/policy/std Std 0.0202661 +trainer/policy/std Max 0.449264 +trainer/policy/std Min 0.393463 +trainer/Advantage Weights Mean 7.13674 +trainer/Advantage Weights Std 19.3235 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.85813e-14 +trainer/Advantage Score Mean -0.0989523 +trainer/Advantage Score Std 0.483098 +trainer/Advantage Score Max 2.09344 +trainer/Advantage Score Min -3.00549 +trainer/V1 Predictions Mean -74.0944 +trainer/V1 Predictions Std 16.9673 +trainer/V1 Predictions Max -0.39263 +trainer/V1 Predictions Min -86.9812 +trainer/VF Loss 0.05838 +expl/num steps total 401000 +expl/num paths total 472 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0452211 +expl/Actions Std 0.852639 +expl/Actions Max 2.57032 +expl/Actions Min -2.22049 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 380945 +eval/num paths total 401 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.103068 +eval/Actions Std 0.777269 +eval/Actions Max 0.999818 +eval/Actions Min -0.999116 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.86102e-06 +time/evaluation sampling (s) 5.01258 +time/exploration sampling (s) 6.45271 +time/logging (s) 0.0120685 +time/saving (s) 0.0272063 +time/training (s) 18.5331 +time/epoch (s) 30.0376 +time/total (s) 8621.21 +Epoch -600 +------------------------------ ---------------- +2022-05-15 20:26:39.102024 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -599 finished +------------------------------ ---------------- +epoch -599 +replay_buffer/size 999047 +trainer/num train calls 402000 +trainer/QF1 Loss 1.32486 +trainer/QF2 Loss 1.34926 +trainer/Policy Loss 5.2228 +trainer/Q1 Predictions Mean -73.9045 +trainer/Q1 Predictions Std 17.1682 +trainer/Q1 Predictions Max -1.16947 +trainer/Q1 Predictions Min -87.3994 +trainer/Q2 Predictions Mean -73.9027 +trainer/Q2 Predictions Std 17.2033 +trainer/Q2 Predictions Max -0.983254 +trainer/Q2 Predictions Min -87.275 +trainer/Q Targets Mean -73.2626 +trainer/Q Targets Std 17.4726 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3051 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0218539 +trainer/policy/mean Std 0.712358 +trainer/policy/mean Max 0.999048 +trainer/policy/mean Min -0.997675 +trainer/policy/std Mean 0.424356 +trainer/policy/std Std 0.0205972 +trainer/policy/std Max 0.444955 +trainer/policy/std Min 0.389961 +trainer/Advantage Weights Mean 1.23911 +trainer/Advantage Weights Std 8.05069 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.52793e-17 +trainer/Advantage Score Mean -0.648828 +trainer/Advantage Score Std 0.673621 +trainer/Advantage Score Max 0.474152 +trainer/Advantage Score Min -3.72679 +trainer/V1 Predictions Mean -72.9638 +trainer/V1 Predictions Std 17.6578 +trainer/V1 Predictions Max 1.18514 +trainer/V1 Predictions Min -86.2271 +trainer/VF Loss 0.0904134 +expl/num steps total 402000 +expl/num paths total 474 +expl/path length Mean 500 +expl/path length Std 442 +expl/path length Max 942 +expl/path length Min 58 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00721603 +expl/Actions Std 0.820017 +expl/Actions Max 2.65392 +expl/Actions Min -2.58015 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 381501 +eval/num paths total 402 +eval/path length Mean 556 +eval/path length Std 0 +eval/path length Max 556 +eval/path length Min 556 +eval/Rewards Mean 0.00179856 +eval/Rewards Std 0.0423713 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0626119 +eval/Actions Std 0.71283 +eval/Actions Max 0.999496 +eval/Actions Min -0.9986 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.04392e-05 +time/evaluation sampling (s) 5.13039 +time/exploration sampling (s) 6.99175 +time/logging (s) 0.0105648 +time/saving (s) 0.0188418 +time/training (s) 18.9373 +time/epoch (s) 31.0889 +time/total (s) 8652.3 +Epoch -599 +------------------------------ ---------------- +2022-05-15 20:27:09.510308 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -598 finished +------------------------------ ---------------- +epoch -598 +replay_buffer/size 999047 +trainer/num train calls 403000 +trainer/QF1 Loss 0.878911 +trainer/QF2 Loss 0.931375 +trainer/Policy Loss 8.80824 +trainer/Q1 Predictions Mean -73.0256 +trainer/Q1 Predictions Std 18.4495 +trainer/Q1 Predictions Max -0.794783 +trainer/Q1 Predictions Min -87.1578 +trainer/Q2 Predictions Mean -73.0292 +trainer/Q2 Predictions Std 18.3579 +trainer/Q2 Predictions Max -1.34049 +trainer/Q2 Predictions Min -87.0636 +trainer/Q Targets Mean -72.6493 +trainer/Q Targets Std 18.6369 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7285 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0198417 +trainer/policy/mean Std 0.722622 +trainer/policy/mean Max 0.999646 +trainer/policy/mean Min -0.999771 +trainer/policy/std Mean 0.422885 +trainer/policy/std Std 0.0211104 +trainer/policy/std Max 0.444537 +trainer/policy/std Min 0.386548 +trainer/Advantage Weights Mean 2.50015 +trainer/Advantage Weights Std 14.0293 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.68195e-17 +trainer/Advantage Score Mean -0.501092 +trainer/Advantage Score Std 0.563499 +trainer/Advantage Score Max 1.5168 +trainer/Advantage Score Min -3.72445 +trainer/V1 Predictions Mean -72.3854 +trainer/V1 Predictions Std 18.8514 +trainer/V1 Predictions Max 0.0924399 +trainer/V1 Predictions Min -86.6491 +trainer/VF Loss 0.0696303 +expl/num steps total 403000 +expl/num paths total 475 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.147146 +expl/Actions Std 0.794517 +expl/Actions Max 2.22341 +expl/Actions Min -2.1651 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 382247 +eval/num paths total 403 +eval/path length Mean 746 +eval/path length Std 0 +eval/path length Max 746 +eval/path length Min 746 +eval/Rewards Mean 0.00134048 +eval/Rewards Std 0.0365881 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00535699 +eval/Actions Std 0.734805 +eval/Actions Max 0.999269 +eval/Actions Min -0.997639 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.12713e-05 +time/evaluation sampling (s) 4.88992 +time/exploration sampling (s) 6.3747 +time/logging (s) 0.00755847 +time/saving (s) 0.0117052 +time/training (s) 19.1074 +time/epoch (s) 30.3913 +time/total (s) 8682.7 +Epoch -598 +------------------------------ ---------------- +2022-05-15 20:27:39.497885 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -597 finished +------------------------------ ---------------- +epoch -597 +replay_buffer/size 999047 +trainer/num train calls 404000 +trainer/QF1 Loss 0.705002 +trainer/QF2 Loss 0.701609 +trainer/Policy Loss 17.1014 +trainer/Q1 Predictions Mean -73.3013 +trainer/Q1 Predictions Std 16.345 +trainer/Q1 Predictions Max -2.06859 +trainer/Q1 Predictions Min -86.298 +trainer/Q2 Predictions Mean -73.3257 +trainer/Q2 Predictions Std 16.4312 +trainer/Q2 Predictions Max -1.68756 +trainer/Q2 Predictions Min -86.2854 +trainer/Q Targets Mean -73.444 +trainer/Q Targets Std 16.3735 +trainer/Q Targets Max -4.07869 +trainer/Q Targets Min -86.1672 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.030406 +trainer/policy/mean Std 0.711725 +trainer/policy/mean Max 0.998546 +trainer/policy/mean Min -0.999278 +trainer/policy/std Mean 0.424483 +trainer/policy/std Std 0.0207605 +trainer/policy/std Max 0.44703 +trainer/policy/std Min 0.389357 +trainer/Advantage Weights Mean 5.83866 +trainer/Advantage Weights Std 18.9912 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.39299e-33 +trainer/Advantage Score Mean -0.354606 +trainer/Advantage Score Std 0.820395 +trainer/Advantage Score Max 1.39888 +trainer/Advantage Score Min -7.56539 +trainer/V1 Predictions Mean -73.1652 +trainer/V1 Predictions Std 16.5275 +trainer/V1 Predictions Max -2.00981 +trainer/V1 Predictions Min -86.2017 +trainer/VF Loss 0.103625 +expl/num steps total 404000 +expl/num paths total 476 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.171565 +expl/Actions Std 0.835957 +expl/Actions Max 2.40128 +expl/Actions Min -2.46209 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 383247 +eval/num paths total 404 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.27738 +eval/Actions Std 0.624322 +eval/Actions Max 0.999694 +eval/Actions Min -0.997713 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.11807e-06 +time/evaluation sampling (s) 4.75429 +time/exploration sampling (s) 6.35356 +time/logging (s) 0.0113838 +time/saving (s) 0.0164091 +time/training (s) 18.8456 +time/epoch (s) 29.9813 +time/total (s) 8712.69 +Epoch -597 +------------------------------ ---------------- +2022-05-15 20:28:08.931828 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -596 finished +------------------------------ ---------------- +epoch -596 +replay_buffer/size 999047 +trainer/num train calls 405000 +trainer/QF1 Loss 0.61819 +trainer/QF2 Loss 0.562515 +trainer/Policy Loss 11.1851 +trainer/Q1 Predictions Mean -72.6506 +trainer/Q1 Predictions Std 18.3123 +trainer/Q1 Predictions Max -0.321625 +trainer/Q1 Predictions Min -87.0572 +trainer/Q2 Predictions Mean -72.5644 +trainer/Q2 Predictions Std 18.3769 +trainer/Q2 Predictions Max -0.489847 +trainer/Q2 Predictions Min -87.1073 +trainer/Q Targets Mean -72.5043 +trainer/Q Targets Std 18.3499 +trainer/Q Targets Max 1.47676 +trainer/Q Targets Min -86.9832 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0025951 +trainer/policy/mean Std 0.70901 +trainer/policy/mean Max 0.997927 +trainer/policy/mean Min -0.998294 +trainer/policy/std Mean 0.422294 +trainer/policy/std Std 0.0203412 +trainer/policy/std Max 0.4457 +trainer/policy/std Min 0.388089 +trainer/Advantage Weights Mean 4.06763 +trainer/Advantage Weights Std 18.3515 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.38109e-18 +trainer/Advantage Score Mean -0.505539 +trainer/Advantage Score Std 0.581031 +trainer/Advantage Score Max 2.70214 +trainer/Advantage Score Min -3.97636 +trainer/V1 Predictions Mean -72.2251 +trainer/V1 Predictions Std 18.3564 +trainer/V1 Predictions Max 2.09344 +trainer/V1 Predictions Min -86.7538 +trainer/VF Loss 0.0940861 +expl/num steps total 405000 +expl/num paths total 478 +expl/path length Mean 500 +expl/path length Std 218 +expl/path length Max 718 +expl/path length Min 282 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.026194 +expl/Actions Std 0.821687 +expl/Actions Max 2.1908 +expl/Actions Min -2.15385 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 383765 +eval/num paths total 405 +eval/path length Mean 518 +eval/path length Std 0 +eval/path length Max 518 +eval/path length Min 518 +eval/Rewards Mean 0.0019305 +eval/Rewards Std 0.043895 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0237269 +eval/Actions Std 0.739121 +eval/Actions Max 0.999527 +eval/Actions Min -0.999102 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.91597e-06 +time/evaluation sampling (s) 4.68283 +time/exploration sampling (s) 6.23561 +time/logging (s) 0.00793766 +time/saving (s) 0.0145396 +time/training (s) 18.4753 +time/epoch (s) 29.4162 +time/total (s) 8742.11 +Epoch -596 +------------------------------ ---------------- +2022-05-15 20:28:39.942213 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -595 finished +------------------------------ ---------------- +epoch -595 +replay_buffer/size 999047 +trainer/num train calls 406000 +trainer/QF1 Loss 0.833562 +trainer/QF2 Loss 0.834364 +trainer/Policy Loss 11.4368 +trainer/Q1 Predictions Mean -72.9746 +trainer/Q1 Predictions Std 17.5973 +trainer/Q1 Predictions Max -2.98203 +trainer/Q1 Predictions Min -87.1574 +trainer/Q2 Predictions Mean -72.9901 +trainer/Q2 Predictions Std 17.5112 +trainer/Q2 Predictions Max -3.39932 +trainer/Q2 Predictions Min -87.0818 +trainer/Q Targets Mean -72.6074 +trainer/Q Targets Std 17.3199 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9603 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00260045 +trainer/policy/mean Std 0.718105 +trainer/policy/mean Max 0.998858 +trainer/policy/mean Min -0.998038 +trainer/policy/std Mean 0.422848 +trainer/policy/std Std 0.0207795 +trainer/policy/std Max 0.446293 +trainer/policy/std Min 0.38692 +trainer/Advantage Weights Mean 2.64435 +trainer/Advantage Weights Std 13.7446 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.61287e-19 +trainer/Advantage Score Mean -0.527618 +trainer/Advantage Score Std 0.509089 +trainer/Advantage Score Max 1.54201 +trainer/Advantage Score Min -4.17193 +trainer/V1 Predictions Mean -72.3041 +trainer/V1 Predictions Std 17.5569 +trainer/V1 Predictions Max -2.30611 +trainer/V1 Predictions Min -86.8305 +trainer/VF Loss 0.0683322 +expl/num steps total 406000 +expl/num paths total 479 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0990211 +expl/Actions Std 0.895657 +expl/Actions Max 2.45922 +expl/Actions Min -2.37575 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 384765 +eval/num paths total 406 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0898389 +eval/Actions Std 0.724502 +eval/Actions Max 0.998546 +eval/Actions Min -0.999182 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.02424e-06 +time/evaluation sampling (s) 5.00718 +time/exploration sampling (s) 7.46912 +time/logging (s) 0.0142201 +time/saving (s) 0.0198161 +time/training (s) 18.4934 +time/epoch (s) 31.0038 +time/total (s) 8773.12 +Epoch -595 +------------------------------ ---------------- +2022-05-15 20:29:11.315414 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -594 finished +------------------------------ ---------------- +epoch -594 +replay_buffer/size 999047 +trainer/num train calls 407000 +trainer/QF1 Loss 0.730431 +trainer/QF2 Loss 0.88878 +trainer/Policy Loss 10.6377 +trainer/Q1 Predictions Mean -72.7319 +trainer/Q1 Predictions Std 17.986 +trainer/Q1 Predictions Max -0.516511 +trainer/Q1 Predictions Min -86.2596 +trainer/Q2 Predictions Mean -72.7378 +trainer/Q2 Predictions Std 17.9056 +trainer/Q2 Predictions Max -0.74924 +trainer/Q2 Predictions Min -86.3394 +trainer/Q Targets Mean -72.6731 +trainer/Q Targets Std 18.2194 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0523 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0174886 +trainer/policy/mean Std 0.698125 +trainer/policy/mean Max 0.999751 +trainer/policy/mean Min -0.994846 +trainer/policy/std Mean 0.42271 +trainer/policy/std Std 0.0206446 +trainer/policy/std Max 0.447144 +trainer/policy/std Min 0.389495 +trainer/Advantage Weights Mean 2.33141 +trainer/Advantage Weights Std 12.7442 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.51663e-17 +trainer/Advantage Score Mean -0.522773 +trainer/Advantage Score Std 0.600709 +trainer/Advantage Score Max 0.786686 +trainer/Advantage Score Min -3.78864 +trainer/V1 Predictions Mean -72.4687 +trainer/V1 Predictions Std 18.165 +trainer/V1 Predictions Max 1.10338 +trainer/V1 Predictions Min -85.9579 +trainer/VF Loss 0.0707753 +expl/num steps total 407000 +expl/num paths total 481 +expl/path length Mean 500 +expl/path length Std 23 +expl/path length Max 523 +expl/path length Min 477 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.007229 +expl/Actions Std 0.827696 +expl/Actions Max 2.4135 +expl/Actions Min -2.46979 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 385765 +eval/num paths total 407 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0514966 +eval/Actions Std 0.725278 +eval/Actions Max 0.999391 +eval/Actions Min -0.999393 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.21002e-05 +time/evaluation sampling (s) 5.67886 +time/exploration sampling (s) 7.00393 +time/logging (s) 0.00943934 +time/saving (s) 0.0147244 +time/training (s) 18.6455 +time/epoch (s) 31.3525 +time/total (s) 8804.48 +Epoch -594 +------------------------------ ---------------- +2022-05-15 20:29:42.491340 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -593 finished +------------------------------ ---------------- +epoch -593 +replay_buffer/size 999047 +trainer/num train calls 408000 +trainer/QF1 Loss 1.70951 +trainer/QF2 Loss 1.96531 +trainer/Policy Loss 17.7145 +trainer/Q1 Predictions Mean -70.6549 +trainer/Q1 Predictions Std 21.6074 +trainer/Q1 Predictions Max -0.474221 +trainer/Q1 Predictions Min -86.6274 +trainer/Q2 Predictions Mean -70.5971 +trainer/Q2 Predictions Std 21.6853 +trainer/Q2 Predictions Max -0.266292 +trainer/Q2 Predictions Min -86.7249 +trainer/Q Targets Mean -70.9893 +trainer/Q Targets Std 21.7812 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4409 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0253927 +trainer/policy/mean Std 0.721219 +trainer/policy/mean Max 0.999882 +trainer/policy/mean Min -0.999426 +trainer/policy/std Mean 0.422788 +trainer/policy/std Std 0.0219993 +trainer/policy/std Max 0.447337 +trainer/policy/std Min 0.387383 +trainer/Advantage Weights Mean 4.11891 +trainer/Advantage Weights Std 17.4524 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.64282e-13 +trainer/Advantage Score Mean -0.305394 +trainer/Advantage Score Std 0.486263 +trainer/Advantage Score Max 1.17677 +trainer/Advantage Score Min -2.83983 +trainer/V1 Predictions Mean -70.8178 +trainer/V1 Predictions Std 21.602 +trainer/V1 Predictions Max -0.614911 +trainer/V1 Predictions Min -87.1966 +trainer/VF Loss 0.0481174 +expl/num steps total 408000 +expl/num paths total 483 +expl/path length Mean 500 +expl/path length Std 99 +expl/path length Max 599 +expl/path length Min 401 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0162094 +expl/Actions Std 0.835401 +expl/Actions Max 2.38472 +expl/Actions Min -2.26967 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 386765 +eval/num paths total 408 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.145276 +eval/Actions Std 0.795998 +eval/Actions Max 0.999889 +eval/Actions Min -0.998321 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.19889e-05 +time/evaluation sampling (s) 5.0806 +time/exploration sampling (s) 6.83115 +time/logging (s) 0.011729 +time/saving (s) 0.014944 +time/training (s) 19.2297 +time/epoch (s) 31.1681 +time/total (s) 8835.66 +Epoch -593 +------------------------------ ---------------- +2022-05-15 20:30:12.751706 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -592 finished +------------------------------ ---------------- +epoch -592 +replay_buffer/size 999047 +trainer/num train calls 409000 +trainer/QF1 Loss 0.873855 +trainer/QF2 Loss 0.917886 +trainer/Policy Loss 11.7573 +trainer/Q1 Predictions Mean -71.7281 +trainer/Q1 Predictions Std 18.168 +trainer/Q1 Predictions Max -0.508852 +trainer/Q1 Predictions Min -86.5633 +trainer/Q2 Predictions Mean -71.6531 +trainer/Q2 Predictions Std 18.1925 +trainer/Q2 Predictions Max -0.459498 +trainer/Q2 Predictions Min -86.5991 +trainer/Q Targets Mean -71.5343 +trainer/Q Targets Std 18.3044 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5193 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0116422 +trainer/policy/mean Std 0.718673 +trainer/policy/mean Max 0.99934 +trainer/policy/mean Min -0.996779 +trainer/policy/std Mean 0.423123 +trainer/policy/std Std 0.0220003 +trainer/policy/std Max 0.447389 +trainer/policy/std Min 0.38683 +trainer/Advantage Weights Mean 3.04665 +trainer/Advantage Weights Std 15.4249 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.93042e-17 +trainer/Advantage Score Mean -0.553392 +trainer/Advantage Score Std 0.653087 +trainer/Advantage Score Max 1.41365 +trainer/Advantage Score Min -3.75485 +trainer/V1 Predictions Mean -71.3545 +trainer/V1 Predictions Std 18.2868 +trainer/V1 Predictions Max 0.33344 +trainer/V1 Predictions Min -86.3853 +trainer/VF Loss 0.0879383 +expl/num steps total 409000 +expl/num paths total 485 +expl/path length Mean 500 +expl/path length Std 428 +expl/path length Max 928 +expl/path length Min 72 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0297129 +expl/Actions Std 0.834358 +expl/Actions Max 2.41092 +expl/Actions Min -2.21636 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 387765 +eval/num paths total 409 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.238342 +eval/Actions Std 0.700613 +eval/Actions Max 0.999871 +eval/Actions Min -0.999282 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.35279e-05 +time/evaluation sampling (s) 4.89439 +time/exploration sampling (s) 6.03201 +time/logging (s) 0.0128208 +time/saving (s) 0.0180662 +time/training (s) 19.2911 +time/epoch (s) 30.2484 +time/total (s) 8865.91 +Epoch -592 +------------------------------ ---------------- +2022-05-15 20:30:44.309822 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -591 finished +------------------------------ ---------------- +epoch -591 +replay_buffer/size 999047 +trainer/num train calls 410000 +trainer/QF1 Loss 0.524857 +trainer/QF2 Loss 0.65792 +trainer/Policy Loss 15.5473 +trainer/Q1 Predictions Mean -74.1815 +trainer/Q1 Predictions Std 17.295 +trainer/Q1 Predictions Max -1.14913 +trainer/Q1 Predictions Min -86.5844 +trainer/Q2 Predictions Mean -74.0662 +trainer/Q2 Predictions Std 17.1974 +trainer/Q2 Predictions Max -1.53758 +trainer/Q2 Predictions Min -86.4108 +trainer/Q Targets Mean -74.3564 +trainer/Q Targets Std 17.238 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3752 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0110257 +trainer/policy/mean Std 0.725738 +trainer/policy/mean Max 0.999373 +trainer/policy/mean Min -0.998885 +trainer/policy/std Mean 0.422535 +trainer/policy/std Std 0.0211116 +trainer/policy/std Max 0.449968 +trainer/policy/std Min 0.387359 +trainer/Advantage Weights Mean 3.45673 +trainer/Advantage Weights Std 14.563 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.25742e-15 +trainer/Advantage Score Mean -0.257814 +trainer/Advantage Score Std 0.486872 +trainer/Advantage Score Max 1.85585 +trainer/Advantage Score Min -3.37246 +trainer/V1 Predictions Mean -74.0829 +trainer/V1 Predictions Std 17.3803 +trainer/V1 Predictions Max -0.896828 +trainer/V1 Predictions Min -86.2241 +trainer/VF Loss 0.0576988 +expl/num steps total 410000 +expl/num paths total 487 +expl/path length Mean 500 +expl/path length Std 249 +expl/path length Max 749 +expl/path length Min 251 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0440952 +expl/Actions Std 0.838404 +expl/Actions Max 2.70995 +expl/Actions Min -2.30685 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 388765 +eval/num paths total 410 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.139418 +eval/Actions Std 0.638416 +eval/Actions Max 0.99955 +eval/Actions Min -0.99855 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.05449e-05 +time/evaluation sampling (s) 5.45017 +time/exploration sampling (s) 6.56009 +time/logging (s) 0.0123552 +time/saving (s) 0.0161727 +time/training (s) 19.5036 +time/epoch (s) 31.5424 +time/total (s) 8897.46 +Epoch -591 +------------------------------ ---------------- +2022-05-15 20:31:14.663779 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -590 finished +------------------------------ ---------------- +epoch -590 +replay_buffer/size 999047 +trainer/num train calls 411000 +trainer/QF1 Loss 1.51776 +trainer/QF2 Loss 1.44028 +trainer/Policy Loss 5.74196 +trainer/Q1 Predictions Mean -73.3823 +trainer/Q1 Predictions Std 17.4448 +trainer/Q1 Predictions Max -0.522953 +trainer/Q1 Predictions Min -86.5386 +trainer/Q2 Predictions Mean -73.3484 +trainer/Q2 Predictions Std 17.4688 +trainer/Q2 Predictions Max -0.629763 +trainer/Q2 Predictions Min -86.4 +trainer/Q Targets Mean -72.9372 +trainer/Q Targets Std 17.6595 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.324 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00775618 +trainer/policy/mean Std 0.724496 +trainer/policy/mean Max 0.997825 +trainer/policy/mean Min -0.998669 +trainer/policy/std Mean 0.424253 +trainer/policy/std Std 0.0209651 +trainer/policy/std Max 0.448436 +trainer/policy/std Min 0.389253 +trainer/Advantage Weights Mean 1.76322 +trainer/Advantage Weights Std 10.0081 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.72217e-19 +trainer/Advantage Score Mean -0.544572 +trainer/Advantage Score Std 0.558049 +trainer/Advantage Score Max 0.98663 +trainer/Advantage Score Min -4.27477 +trainer/V1 Predictions Mean -72.7934 +trainer/V1 Predictions Std 17.5537 +trainer/V1 Predictions Max 0.21746 +trainer/V1 Predictions Min -86.2471 +trainer/VF Loss 0.0678631 +expl/num steps total 411000 +expl/num paths total 488 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.22498 +expl/Actions Std 0.843686 +expl/Actions Max 2.45234 +expl/Actions Min -2.27954 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 389765 +eval/num paths total 411 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.099038 +eval/Actions Std 0.693073 +eval/Actions Max 0.999907 +eval/Actions Min -0.99758 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.21896e-05 +time/evaluation sampling (s) 4.62312 +time/exploration sampling (s) 6.82031 +time/logging (s) 0.00837536 +time/saving (s) 0.0159188 +time/training (s) 18.8692 +time/epoch (s) 30.337 +time/total (s) 8927.8 +Epoch -590 +------------------------------ ---------------- +2022-05-15 20:31:45.299245 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -589 finished +------------------------------ ---------------- +epoch -589 +replay_buffer/size 999047 +trainer/num train calls 412000 +trainer/QF1 Loss 0.979723 +trainer/QF2 Loss 1.02405 +trainer/Policy Loss 9.25679 +trainer/Q1 Predictions Mean -72.5406 +trainer/Q1 Predictions Std 18.781 +trainer/Q1 Predictions Max -0.466006 +trainer/Q1 Predictions Min -87.4117 +trainer/Q2 Predictions Mean -72.6256 +trainer/Q2 Predictions Std 18.7263 +trainer/Q2 Predictions Max 0.365027 +trainer/Q2 Predictions Min -87.1703 +trainer/Q Targets Mean -72.1628 +trainer/Q Targets Std 18.7357 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7476 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00635821 +trainer/policy/mean Std 0.713488 +trainer/policy/mean Max 0.997845 +trainer/policy/mean Min -0.997698 +trainer/policy/std Mean 0.424857 +trainer/policy/std Std 0.0216186 +trainer/policy/std Max 0.447792 +trainer/policy/std Min 0.38651 +trainer/Advantage Weights Mean 1.69415 +trainer/Advantage Weights Std 9.72397 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.01017e-19 +trainer/Advantage Score Mean -0.518939 +trainer/Advantage Score Std 0.561863 +trainer/Advantage Score Max 0.471985 +trainer/Advantage Score Min -4.30509 +trainer/V1 Predictions Mean -71.8915 +trainer/V1 Predictions Std 18.8448 +trainer/V1 Predictions Max 0.0194687 +trainer/V1 Predictions Min -86.4237 +trainer/VF Loss 0.062549 +expl/num steps total 412000 +expl/num paths total 489 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0323316 +expl/Actions Std 0.839079 +expl/Actions Max 2.44674 +expl/Actions Min -2.25332 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 390765 +eval/num paths total 412 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.169457 +eval/Actions Std 0.717772 +eval/Actions Max 0.999723 +eval/Actions Min -0.998232 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.20541e-05 +time/evaluation sampling (s) 4.7626 +time/exploration sampling (s) 6.71513 +time/logging (s) 0.0125134 +time/saving (s) 0.0151817 +time/training (s) 19.1235 +time/epoch (s) 30.629 +time/total (s) 8958.44 +Epoch -589 +------------------------------ ---------------- +2022-05-15 20:32:16.539028 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -588 finished +------------------------------ ---------------- +epoch -588 +replay_buffer/size 999047 +trainer/num train calls 413000 +trainer/QF1 Loss 0.934399 +trainer/QF2 Loss 0.87237 +trainer/Policy Loss 16.7276 +trainer/Q1 Predictions Mean -72.9288 +trainer/Q1 Predictions Std 17.679 +trainer/Q1 Predictions Max -2.98929 +trainer/Q1 Predictions Min -86.5763 +trainer/Q2 Predictions Mean -72.8677 +trainer/Q2 Predictions Std 17.7304 +trainer/Q2 Predictions Max -3.46615 +trainer/Q2 Predictions Min -86.7685 +trainer/Q Targets Mean -72.6809 +trainer/Q Targets Std 18.2194 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8812 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000684492 +trainer/policy/mean Std 0.717027 +trainer/policy/mean Max 0.999969 +trainer/policy/mean Min -0.998547 +trainer/policy/std Mean 0.424366 +trainer/policy/std Std 0.0218096 +trainer/policy/std Max 0.448916 +trainer/policy/std Min 0.387047 +trainer/Advantage Weights Mean 2.84209 +trainer/Advantage Weights Std 12.1248 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.20956e-20 +trainer/Advantage Score Mean -0.423482 +trainer/Advantage Score Std 0.682781 +trainer/Advantage Score Max 0.996315 +trainer/Advantage Score Min -4.58614 +trainer/V1 Predictions Mean -72.412 +trainer/V1 Predictions Std 18.4019 +trainer/V1 Predictions Max -2.21099 +trainer/V1 Predictions Min -86.9531 +trainer/VF Loss 0.0759606 +expl/num steps total 413000 +expl/num paths total 491 +expl/path length Mean 500 +expl/path length Std 343 +expl/path length Max 843 +expl/path length Min 157 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00723946 +expl/Actions Std 0.841309 +expl/Actions Max 2.4527 +expl/Actions Min -2.33096 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 391765 +eval/num paths total 413 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0201408 +eval/Actions Std 0.711463 +eval/Actions Max 0.999703 +eval/Actions Min -0.999476 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.08532e-05 +time/evaluation sampling (s) 4.87336 +time/exploration sampling (s) 7.25026 +time/logging (s) 0.0123699 +time/saving (s) 0.0177249 +time/training (s) 19.0732 +time/epoch (s) 31.2269 +time/total (s) 8989.67 +Epoch -588 +------------------------------ ---------------- +2022-05-15 20:32:47.433865 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -587 finished +------------------------------ ---------------- +epoch -587 +replay_buffer/size 999047 +trainer/num train calls 414000 +trainer/QF1 Loss 0.734784 +trainer/QF2 Loss 0.768041 +trainer/Policy Loss 58.6815 +trainer/Q1 Predictions Mean -73.4478 +trainer/Q1 Predictions Std 17.7033 +trainer/Q1 Predictions Max -0.918685 +trainer/Q1 Predictions Min -86.9996 +trainer/Q2 Predictions Mean -73.4353 +trainer/Q2 Predictions Std 17.6888 +trainer/Q2 Predictions Max -1.1298 +trainer/Q2 Predictions Min -86.5828 +trainer/Q Targets Mean -73.6443 +trainer/Q Targets Std 17.6293 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5442 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0149654 +trainer/policy/mean Std 0.727827 +trainer/policy/mean Max 0.997413 +trainer/policy/mean Min -0.999177 +trainer/policy/std Mean 0.424125 +trainer/policy/std Std 0.0208194 +trainer/policy/std Max 0.447124 +trainer/policy/std Min 0.388703 +trainer/Advantage Weights Mean 12.1382 +trainer/Advantage Weights Std 28.8028 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.35857e-19 +trainer/Advantage Score Mean -0.104696 +trainer/Advantage Score Std 0.576878 +trainer/Advantage Score Max 2.14195 +trainer/Advantage Score Min -4.20704 +trainer/V1 Predictions Mean -73.4349 +trainer/V1 Predictions Std 17.6671 +trainer/V1 Predictions Max -1.23193 +trainer/V1 Predictions Min -86.5062 +trainer/VF Loss 0.0959943 +expl/num steps total 414000 +expl/num paths total 493 +expl/path length Mean 500 +expl/path length Std 227 +expl/path length Max 727 +expl/path length Min 273 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00103445 +expl/Actions Std 0.850947 +expl/Actions Max 2.37517 +expl/Actions Min -2.35073 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 392436 +eval/num paths total 414 +eval/path length Mean 671 +eval/path length Std 0 +eval/path length Max 671 +eval/path length Min 671 +eval/Rewards Mean 0.00149031 +eval/Rewards Std 0.0385758 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0493828 +eval/Actions Std 0.720233 +eval/Actions Max 0.999463 +eval/Actions Min -0.999105 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 7.86502e-06 +time/evaluation sampling (s) 5.12707 +time/exploration sampling (s) 6.92271 +time/logging (s) 0.00693757 +time/saving (s) 0.0133297 +time/training (s) 18.8052 +time/epoch (s) 30.8752 +time/total (s) 9020.55 +Epoch -587 +------------------------------ ---------------- +2022-05-15 20:33:17.622194 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -586 finished +------------------------------ ---------------- +epoch -586 +replay_buffer/size 999047 +trainer/num train calls 415000 +trainer/QF1 Loss 0.873304 +trainer/QF2 Loss 0.73835 +trainer/Policy Loss 12.9773 +trainer/Q1 Predictions Mean -72.4174 +trainer/Q1 Predictions Std 18.088 +trainer/Q1 Predictions Max -0.548159 +trainer/Q1 Predictions Min -86.4148 +trainer/Q2 Predictions Mean -72.3024 +trainer/Q2 Predictions Std 18.0802 +trainer/Q2 Predictions Max -1.15365 +trainer/Q2 Predictions Min -86.1931 +trainer/Q Targets Mean -72.2802 +trainer/Q Targets Std 17.7062 +trainer/Q Targets Max 0 +trainer/Q Targets Min -85.9832 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0229508 +trainer/policy/mean Std 0.7193 +trainer/policy/mean Max 0.999327 +trainer/policy/mean Min -0.997503 +trainer/policy/std Mean 0.423574 +trainer/policy/std Std 0.0199544 +trainer/policy/std Max 0.445393 +trainer/policy/std Min 0.389493 +trainer/Advantage Weights Mean 3.65842 +trainer/Advantage Weights Std 17.6004 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.56409e-13 +trainer/Advantage Score Mean -0.549997 +trainer/Advantage Score Std 0.559388 +trainer/Advantage Score Max 1.77897 +trainer/Advantage Score Min -2.76756 +trainer/V1 Predictions Mean -72.0185 +trainer/V1 Predictions Std 17.8036 +trainer/V1 Predictions Max 0.0255445 +trainer/V1 Predictions Min -85.8583 +trainer/VF Loss 0.100733 +expl/num steps total 415000 +expl/num paths total 494 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.111762 +expl/Actions Std 0.81453 +expl/Actions Max 2.23485 +expl/Actions Min -2.45396 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 393436 +eval/num paths total 415 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.27944 +eval/Actions Std 0.63565 +eval/Actions Max 0.998727 +eval/Actions Min -0.997351 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.56884e-06 +time/evaluation sampling (s) 5.2241 +time/exploration sampling (s) 6.4062 +time/logging (s) 0.00891859 +time/saving (s) 0.0151197 +time/training (s) 18.5265 +time/epoch (s) 30.1808 +time/total (s) 9050.74 +Epoch -586 +------------------------------ ---------------- +2022-05-15 20:33:49.114485 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -585 finished +------------------------------ ---------------- +epoch -585 +replay_buffer/size 999047 +trainer/num train calls 416000 +trainer/QF1 Loss 0.792677 +trainer/QF2 Loss 0.889678 +trainer/Policy Loss 40.4935 +trainer/Q1 Predictions Mean -72.6149 +trainer/Q1 Predictions Std 19.3701 +trainer/Q1 Predictions Max -0.473792 +trainer/Q1 Predictions Min -86.8132 +trainer/Q2 Predictions Mean -72.6034 +trainer/Q2 Predictions Std 19.3562 +trainer/Q2 Predictions Max -0.562068 +trainer/Q2 Predictions Min -86.8447 +trainer/Q Targets Mean -72.3768 +trainer/Q Targets Std 19.258 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.1978 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0122442 +trainer/policy/mean Std 0.718886 +trainer/policy/mean Max 0.99763 +trainer/policy/mean Min -0.99903 +trainer/policy/std Mean 0.423759 +trainer/policy/std Std 0.0205192 +trainer/policy/std Max 0.44517 +trainer/policy/std Min 0.389549 +trainer/Advantage Weights Mean 8.14254 +trainer/Advantage Weights Std 25.9486 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.80264e-10 +trainer/Advantage Score Mean -0.31729 +trainer/Advantage Score Std 0.529428 +trainer/Advantage Score Max 2.21739 +trainer/Advantage Score Min -2.24366 +trainer/V1 Predictions Mean -72.2472 +trainer/V1 Predictions Std 19.0489 +trainer/V1 Predictions Max -0.458268 +trainer/V1 Predictions Min -86.0723 +trainer/VF Loss 0.117442 +expl/num steps total 416000 +expl/num paths total 496 +expl/path length Mean 500 +expl/path length Std 343 +expl/path length Max 843 +expl/path length Min 157 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0240746 +expl/Actions Std 0.828583 +expl/Actions Max 2.23516 +expl/Actions Min -2.38127 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 394062 +eval/num paths total 416 +eval/path length Mean 626 +eval/path length Std 0 +eval/path length Max 626 +eval/path length Min 626 +eval/Rewards Mean 0.00159744 +eval/Rewards Std 0.0399361 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0234177 +eval/Actions Std 0.737025 +eval/Actions Max 0.999412 +eval/Actions Min -0.998207 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.03428e-05 +time/evaluation sampling (s) 4.98192 +time/exploration sampling (s) 6.93829 +time/logging (s) 0.0106236 +time/saving (s) 0.0173932 +time/training (s) 19.5371 +time/epoch (s) 31.4853 +time/total (s) 9082.23 +Epoch -585 +------------------------------ ---------------- +2022-05-15 20:34:19.376594 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -584 finished +------------------------------ ---------------- +epoch -584 +replay_buffer/size 999047 +trainer/num train calls 417000 +trainer/QF1 Loss 0.761475 +trainer/QF2 Loss 0.640047 +trainer/Policy Loss 7.73831 +trainer/Q1 Predictions Mean -73.3076 +trainer/Q1 Predictions Std 17.445 +trainer/Q1 Predictions Max -0.743533 +trainer/Q1 Predictions Min -87.1602 +trainer/Q2 Predictions Mean -73.3312 +trainer/Q2 Predictions Std 17.4831 +trainer/Q2 Predictions Max -0.590737 +trainer/Q2 Predictions Min -87.0764 +trainer/Q Targets Mean -72.9719 +trainer/Q Targets Std 17.6312 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6639 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00370564 +trainer/policy/mean Std 0.721262 +trainer/policy/mean Max 0.999061 +trainer/policy/mean Min -0.996192 +trainer/policy/std Mean 0.424336 +trainer/policy/std Std 0.0208711 +trainer/policy/std Max 0.444711 +trainer/policy/std Min 0.389952 +trainer/Advantage Weights Mean 2.28248 +trainer/Advantage Weights Std 13.9618 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.29609e-11 +trainer/Advantage Score Mean -0.726522 +trainer/Advantage Score Std 0.48934 +trainer/Advantage Score Max 0.699936 +trainer/Advantage Score Min -2.50691 +trainer/V1 Predictions Mean -72.802 +trainer/V1 Predictions Std 17.5678 +trainer/V1 Predictions Max 1.04907 +trainer/V1 Predictions Min -86.6383 +trainer/VF Loss 0.0828721 +expl/num steps total 417000 +expl/num paths total 498 +expl/path length Mean 500 +expl/path length Std 400 +expl/path length Max 900 +expl/path length Min 100 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0326976 +expl/Actions Std 0.848159 +expl/Actions Max 2.29927 +expl/Actions Min -2.36258 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 395062 +eval/num paths total 417 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.324911 +eval/Actions Std 0.792119 +eval/Actions Max 0.999615 +eval/Actions Min -0.999072 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.03782e-05 +time/evaluation sampling (s) 5.13961 +time/exploration sampling (s) 6.27283 +time/logging (s) 0.0121081 +time/saving (s) 0.016105 +time/training (s) 18.8042 +time/epoch (s) 30.2449 +time/total (s) 9112.48 +Epoch -584 +------------------------------ ---------------- +2022-05-15 20:34:49.491254 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -583 finished +------------------------------ ---------------- +epoch -583 +replay_buffer/size 999047 +trainer/num train calls 418000 +trainer/QF1 Loss 0.765885 +trainer/QF2 Loss 0.681263 +trainer/Policy Loss 18.8436 +trainer/Q1 Predictions Mean -72.8025 +trainer/Q1 Predictions Std 19.4429 +trainer/Q1 Predictions Max -0.747778 +trainer/Q1 Predictions Min -87.0651 +trainer/Q2 Predictions Mean -72.7576 +trainer/Q2 Predictions Std 19.4303 +trainer/Q2 Predictions Max -0.129057 +trainer/Q2 Predictions Min -87.1121 +trainer/Q Targets Mean -72.6418 +trainer/Q Targets Std 19.4412 +trainer/Q Targets Max 0.268602 +trainer/Q Targets Min -86.4645 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.026052 +trainer/policy/mean Std 0.716013 +trainer/policy/mean Max 0.996966 +trainer/policy/mean Min -0.997462 +trainer/policy/std Mean 0.422383 +trainer/policy/std Std 0.0203261 +trainer/policy/std Max 0.443394 +trainer/policy/std Min 0.389191 +trainer/Advantage Weights Mean 2.08277 +trainer/Advantage Weights Std 12.9659 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.98943e-13 +trainer/Advantage Score Mean -0.600434 +trainer/Advantage Score Std 0.576991 +trainer/Advantage Score Max 1.08531 +trainer/Advantage Score Min -2.855 +trainer/V1 Predictions Mean -72.3211 +trainer/V1 Predictions Std 19.6542 +trainer/V1 Predictions Max 0.850127 +trainer/V1 Predictions Min -86.5701 +trainer/VF Loss 0.0789656 +expl/num steps total 418000 +expl/num paths total 499 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.163588 +expl/Actions Std 0.807534 +expl/Actions Max 2.78519 +expl/Actions Min -2.2664 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 396062 +eval/num paths total 418 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.10634 +eval/Actions Std 0.735114 +eval/Actions Max 0.999931 +eval/Actions Min -0.998983 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.35563e-05 +time/evaluation sampling (s) 4.71333 +time/exploration sampling (s) 6.35034 +time/logging (s) 0.0117596 +time/saving (s) 0.0156167 +time/training (s) 19.0127 +time/epoch (s) 30.1037 +time/total (s) 9142.59 +Epoch -583 +------------------------------ ---------------- +2022-05-15 20:35:19.655091 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -582 finished +------------------------------ ---------------- +epoch -582 +replay_buffer/size 999047 +trainer/num train calls 419000 +trainer/QF1 Loss 3.68672 +trainer/QF2 Loss 3.63278 +trainer/Policy Loss 13.3314 +trainer/Q1 Predictions Mean -74.0572 +trainer/Q1 Predictions Std 17.2204 +trainer/Q1 Predictions Max -0.887894 +trainer/Q1 Predictions Min -86.3387 +trainer/Q2 Predictions Mean -74.0288 +trainer/Q2 Predictions Std 17.2618 +trainer/Q2 Predictions Max -0.718768 +trainer/Q2 Predictions Min -86.3551 +trainer/Q Targets Mean -73.832 +trainer/Q Targets Std 17.1555 +trainer/Q Targets Max -1.50126 +trainer/Q Targets Min -86.4881 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00244106 +trainer/policy/mean Std 0.713559 +trainer/policy/mean Max 0.998875 +trainer/policy/mean Min -0.997306 +trainer/policy/std Mean 0.423406 +trainer/policy/std Std 0.0192226 +trainer/policy/std Max 0.44277 +trainer/policy/std Min 0.390675 +trainer/Advantage Weights Mean 4.03516 +trainer/Advantage Weights Std 17.7919 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.69711e-13 +trainer/Advantage Score Mean -0.392398 +trainer/Advantage Score Std 0.518119 +trainer/Advantage Score Max 1.59006 +trainer/Advantage Score Min -2.94047 +trainer/V1 Predictions Mean -73.6843 +trainer/V1 Predictions Std 17.2967 +trainer/V1 Predictions Max 0.444501 +trainer/V1 Predictions Min -86.2693 +trainer/VF Loss 0.071419 +expl/num steps total 419000 +expl/num paths total 500 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0750644 +expl/Actions Std 0.871232 +expl/Actions Max 2.48968 +expl/Actions Min -2.38325 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 397057 +eval/num paths total 419 +eval/path length Mean 995 +eval/path length Std 0 +eval/path length Max 995 +eval/path length Min 995 +eval/Rewards Mean 0.00100503 +eval/Rewards Std 0.0316862 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0384049 +eval/Actions Std 0.750707 +eval/Actions Max 0.999634 +eval/Actions Min -0.999883 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.12831e-06 +time/evaluation sampling (s) 4.90831 +time/exploration sampling (s) 6.89374 +time/logging (s) 0.00788907 +time/saving (s) 0.0115504 +time/training (s) 18.3285 +time/epoch (s) 30.15 +time/total (s) 9172.74 +Epoch -582 +------------------------------ ---------------- +2022-05-15 20:35:50.334824 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -581 finished +------------------------------ ---------------- +epoch -581 +replay_buffer/size 999047 +trainer/num train calls 420000 +trainer/QF1 Loss 0.692584 +trainer/QF2 Loss 0.86394 +trainer/Policy Loss 10.2777 +trainer/Q1 Predictions Mean -71.3411 +trainer/Q1 Predictions Std 20.2048 +trainer/Q1 Predictions Max -0.507972 +trainer/Q1 Predictions Min -86.2468 +trainer/Q2 Predictions Mean -71.3683 +trainer/Q2 Predictions Std 20.2769 +trainer/Q2 Predictions Max -0.195653 +trainer/Q2 Predictions Min -86.3581 +trainer/Q Targets Mean -71.5832 +trainer/Q Targets Std 20.1254 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7245 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0116959 +trainer/policy/mean Std 0.72665 +trainer/policy/mean Max 0.999482 +trainer/policy/mean Min -0.999246 +trainer/policy/std Mean 0.422478 +trainer/policy/std Std 0.0199911 +trainer/policy/std Max 0.443362 +trainer/policy/std Min 0.387824 +trainer/Advantage Weights Mean 3.04012 +trainer/Advantage Weights Std 14.5438 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.24648e-14 +trainer/Advantage Score Mean -0.428353 +trainer/Advantage Score Std 0.555513 +trainer/Advantage Score Max 1.7082 +trainer/Advantage Score Min -3.07901 +trainer/V1 Predictions Mean -71.3411 +trainer/V1 Predictions Std 20.3296 +trainer/V1 Predictions Max 0.247374 +trainer/V1 Predictions Min -86.5624 +trainer/VF Loss 0.0673996 +expl/num steps total 420000 +expl/num paths total 501 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0296551 +expl/Actions Std 0.844484 +expl/Actions Max 2.55002 +expl/Actions Min -2.30074 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 397782 +eval/num paths total 420 +eval/path length Mean 725 +eval/path length Std 0 +eval/path length Max 725 +eval/path length Min 725 +eval/Rewards Mean 0.00137931 +eval/Rewards Std 0.0371134 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00943522 +eval/Actions Std 0.751303 +eval/Actions Max 0.999925 +eval/Actions Min -0.999566 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.17929e-05 +time/evaluation sampling (s) 4.5829 +time/exploration sampling (s) 6.98931 +time/logging (s) 0.00981135 +time/saving (s) 0.0151839 +time/training (s) 19.073 +time/epoch (s) 30.6703 +time/total (s) 9203.42 +Epoch -581 +------------------------------ ---------------- +2022-05-15 20:36:21.311171 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -580 finished +------------------------------ ---------------- +epoch -580 +replay_buffer/size 999047 +trainer/num train calls 421000 +trainer/QF1 Loss 0.67976 +trainer/QF2 Loss 0.677744 +trainer/Policy Loss 32.4643 +trainer/Q1 Predictions Mean -72.2554 +trainer/Q1 Predictions Std 17.9978 +trainer/Q1 Predictions Max -1.56552 +trainer/Q1 Predictions Min -86.6176 +trainer/Q2 Predictions Mean -72.2625 +trainer/Q2 Predictions Std 17.9168 +trainer/Q2 Predictions Max -1.08415 +trainer/Q2 Predictions Min -86.5643 +trainer/Q Targets Mean -72.4875 +trainer/Q Targets Std 18.1235 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7874 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00345005 +trainer/policy/mean Std 0.711246 +trainer/policy/mean Max 0.998538 +trainer/policy/mean Min -0.99747 +trainer/policy/std Mean 0.4229 +trainer/policy/std Std 0.0214563 +trainer/policy/std Max 0.444555 +trainer/policy/std Min 0.385249 +trainer/Advantage Weights Mean 7.40257 +trainer/Advantage Weights Std 22.7934 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.14899e-17 +trainer/Advantage Score Mean -0.222583 +trainer/Advantage Score Std 0.549231 +trainer/Advantage Score Max 2.27852 +trainer/Advantage Score Min -3.77211 +trainer/V1 Predictions Mean -72.1992 +trainer/V1 Predictions Std 18.2281 +trainer/V1 Predictions Max -0.278778 +trainer/V1 Predictions Min -86.6543 +trainer/VF Loss 0.0740636 +expl/num steps total 421000 +expl/num paths total 503 +expl/path length Mean 500 +expl/path length Std 41 +expl/path length Max 541 +expl/path length Min 459 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0899883 +expl/Actions Std 0.845411 +expl/Actions Max 2.25578 +expl/Actions Min -2.50173 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 398285 +eval/num paths total 421 +eval/path length Mean 503 +eval/path length Std 0 +eval/path length Max 503 +eval/path length Min 503 +eval/Rewards Mean 0.00198807 +eval/Rewards Std 0.0445435 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -7.30637e-05 +eval/Actions Std 0.732776 +eval/Actions Max 0.99839 +eval/Actions Min -0.998109 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.03433e-05 +time/evaluation sampling (s) 5.00175 +time/exploration sampling (s) 7.30074 +time/logging (s) 0.00833521 +time/saving (s) 0.0132452 +time/training (s) 18.6415 +time/epoch (s) 30.9656 +time/total (s) 9234.39 +Epoch -580 +------------------------------ ---------------- +2022-05-15 20:36:51.338367 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -579 finished +------------------------------ ---------------- +epoch -579 +replay_buffer/size 999047 +trainer/num train calls 422000 +trainer/QF1 Loss 0.67577 +trainer/QF2 Loss 0.631729 +trainer/Policy Loss 4.60785 +trainer/Q1 Predictions Mean -72.4287 +trainer/Q1 Predictions Std 17.8842 +trainer/Q1 Predictions Max -1.46914 +trainer/Q1 Predictions Min -86.7207 +trainer/Q2 Predictions Mean -72.3364 +trainer/Q2 Predictions Std 17.992 +trainer/Q2 Predictions Max -1.51013 +trainer/Q2 Predictions Min -86.8014 +trainer/Q Targets Mean -72.013 +trainer/Q Targets Std 17.887 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3569 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0106966 +trainer/policy/mean Std 0.713556 +trainer/policy/mean Max 0.998376 +trainer/policy/mean Min -0.998332 +trainer/policy/std Mean 0.423032 +trainer/policy/std Std 0.0213119 +trainer/policy/std Max 0.445517 +trainer/policy/std Min 0.385153 +trainer/Advantage Weights Mean 1.4253 +trainer/Advantage Weights Std 10.9227 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.39554e-13 +trainer/Advantage Score Mean -0.590283 +trainer/Advantage Score Std 0.496145 +trainer/Advantage Score Max 1.73821 +trainer/Advantage Score Min -2.8453 +trainer/V1 Predictions Mean -71.7318 +trainer/V1 Predictions Std 17.9333 +trainer/V1 Predictions Max -1.21131 +trainer/V1 Predictions Min -86.1505 +trainer/VF Loss 0.0750682 +expl/num steps total 422000 +expl/num paths total 504 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.251319 +expl/Actions Std 0.819844 +expl/Actions Max 2.26939 +expl/Actions Min -2.39431 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 399285 +eval/num paths total 422 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.163841 +eval/Actions Std 0.742402 +eval/Actions Max 0.999831 +eval/Actions Min -0.998324 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.05123e-05 +time/evaluation sampling (s) 4.75826 +time/exploration sampling (s) 6.83893 +time/logging (s) 0.0119682 +time/saving (s) 0.0148266 +time/training (s) 18.3949 +time/epoch (s) 30.0189 +time/total (s) 9264.41 +Epoch -579 +------------------------------ ---------------- +2022-05-15 20:37:21.791229 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -578 finished +------------------------------ ---------------- +epoch -578 +replay_buffer/size 999047 +trainer/num train calls 423000 +trainer/QF1 Loss 0.931565 +trainer/QF2 Loss 0.97615 +trainer/Policy Loss 10.7786 +trainer/Q1 Predictions Mean -72.7462 +trainer/Q1 Predictions Std 19.7668 +trainer/Q1 Predictions Max -0.85174 +trainer/Q1 Predictions Min -86.6793 +trainer/Q2 Predictions Mean -72.7858 +trainer/Q2 Predictions Std 19.784 +trainer/Q2 Predictions Max -1.22569 +trainer/Q2 Predictions Min -86.7013 +trainer/Q Targets Mean -72.689 +trainer/Q Targets Std 19.7471 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9651 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0115666 +trainer/policy/mean Std 0.716475 +trainer/policy/mean Max 0.999773 +trainer/policy/mean Min -0.996934 +trainer/policy/std Mean 0.42289 +trainer/policy/std Std 0.0208208 +trainer/policy/std Max 0.445318 +trainer/policy/std Min 0.385052 +trainer/Advantage Weights Mean 3.54052 +trainer/Advantage Weights Std 15.325 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.65243e-18 +trainer/Advantage Score Mean -0.39524 +trainer/Advantage Score Std 0.536838 +trainer/Advantage Score Max 1.86011 +trainer/Advantage Score Min -4.01511 +trainer/V1 Predictions Mean -72.4611 +trainer/V1 Predictions Std 19.7472 +trainer/V1 Predictions Max -0.209367 +trainer/V1 Predictions Min -86.6948 +trainer/VF Loss 0.0653216 +expl/num steps total 423000 +expl/num paths total 506 +expl/path length Mean 500 +expl/path length Std 235 +expl/path length Max 735 +expl/path length Min 265 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0298922 +expl/Actions Std 0.831732 +expl/Actions Max 2.30383 +expl/Actions Min -2.43383 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 400251 +eval/num paths total 423 +eval/path length Mean 966 +eval/path length Std 0 +eval/path length Max 966 +eval/path length Min 966 +eval/Rewards Mean 0.0010352 +eval/Rewards Std 0.0321578 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0423277 +eval/Actions Std 0.743383 +eval/Actions Max 0.999762 +eval/Actions Min -0.9998 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.04671e-05 +time/evaluation sampling (s) 5.00929 +time/exploration sampling (s) 7.18056 +time/logging (s) 0.0124584 +time/saving (s) 0.0179341 +time/training (s) 18.2241 +time/epoch (s) 30.4444 +time/total (s) 9294.86 +Epoch -578 +------------------------------ ---------------- +2022-05-15 20:37:53.187781 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -577 finished +------------------------------ ---------------- +epoch -577 +replay_buffer/size 999047 +trainer/num train calls 424000 +trainer/QF1 Loss 1.04298 +trainer/QF2 Loss 0.907139 +trainer/Policy Loss 7.11271 +trainer/Q1 Predictions Mean -72.289 +trainer/Q1 Predictions Std 20.25 +trainer/Q1 Predictions Max -0.681159 +trainer/Q1 Predictions Min -86.7155 +trainer/Q2 Predictions Mean -72.2242 +trainer/Q2 Predictions Std 20.2218 +trainer/Q2 Predictions Max -0.636405 +trainer/Q2 Predictions Min -86.555 +trainer/Q Targets Mean -71.8486 +trainer/Q Targets Std 20.2746 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2284 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0205592 +trainer/policy/mean Std 0.714847 +trainer/policy/mean Max 0.999514 +trainer/policy/mean Min -0.998171 +trainer/policy/std Mean 0.422479 +trainer/policy/std Std 0.0208427 +trainer/policy/std Max 0.446129 +trainer/policy/std Min 0.38468 +trainer/Advantage Weights Mean 2.11904 +trainer/Advantage Weights Std 12.5717 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.8326e-17 +trainer/Advantage Score Mean -0.485891 +trainer/Advantage Score Std 0.538975 +trainer/Advantage Score Max 1.17441 +trainer/Advantage Score Min -3.81027 +trainer/V1 Predictions Mean -71.6039 +trainer/V1 Predictions Std 20.4374 +trainer/V1 Predictions Max 0.0387211 +trainer/V1 Predictions Min -86.0258 +trainer/VF Loss 0.0626027 +expl/num steps total 424000 +expl/num paths total 507 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0773068 +expl/Actions Std 0.820769 +expl/Actions Max 2.46817 +expl/Actions Min -2.51174 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 401251 +eval/num paths total 424 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.146961 +eval/Actions Std 0.684941 +eval/Actions Max 0.999144 +eval/Actions Min -0.998697 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.10087e-05 +time/evaluation sampling (s) 5.33081 +time/exploration sampling (s) 7.10966 +time/logging (s) 0.00909566 +time/saving (s) 0.015927 +time/training (s) 18.9134 +time/epoch (s) 31.3789 +time/total (s) 9326.25 +Epoch -577 +------------------------------ ---------------- +2022-05-15 20:38:24.272444 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -576 finished +------------------------------ ---------------- +epoch -576 +replay_buffer/size 999047 +trainer/num train calls 425000 +trainer/QF1 Loss 0.75228 +trainer/QF2 Loss 0.782864 +trainer/Policy Loss 17.0287 +trainer/Q1 Predictions Mean -72.8193 +trainer/Q1 Predictions Std 17.7544 +trainer/Q1 Predictions Max -0.790558 +trainer/Q1 Predictions Min -86.0683 +trainer/Q2 Predictions Mean -72.9097 +trainer/Q2 Predictions Std 17.6865 +trainer/Q2 Predictions Max -1.14721 +trainer/Q2 Predictions Min -86.347 +trainer/Q Targets Mean -72.8769 +trainer/Q Targets Std 17.9299 +trainer/Q Targets Max -3.68676 +trainer/Q Targets Min -86.2618 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.020848 +trainer/policy/mean Std 0.72119 +trainer/policy/mean Max 0.999035 +trainer/policy/mean Min -0.998492 +trainer/policy/std Mean 0.422853 +trainer/policy/std Std 0.0222847 +trainer/policy/std Max 0.448983 +trainer/policy/std Min 0.385083 +trainer/Advantage Weights Mean 2.56605 +trainer/Advantage Weights Std 13.9966 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.11662e-13 +trainer/Advantage Score Mean -0.469233 +trainer/Advantage Score Std 0.484609 +trainer/Advantage Score Max 0.941631 +trainer/Advantage Score Min -2.81226 +trainer/V1 Predictions Mean -72.6475 +trainer/V1 Predictions Std 17.9166 +trainer/V1 Predictions Max 0.0594719 +trainer/V1 Predictions Min -86.0678 +trainer/VF Loss 0.0535062 +expl/num steps total 425000 +expl/num paths total 508 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.352024 +expl/Actions Std 0.864371 +expl/Actions Max 2.64102 +expl/Actions Min -2.59768 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 402251 +eval/num paths total 425 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0149519 +eval/Actions Std 0.758297 +eval/Actions Max 0.998823 +eval/Actions Min -0.996682 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.02471e-06 +time/evaluation sampling (s) 5.38 +time/exploration sampling (s) 6.26942 +time/logging (s) 0.0124165 +time/saving (s) 0.0169101 +time/training (s) 19.3919 +time/epoch (s) 31.0706 +time/total (s) 9357.33 +Epoch -576 +------------------------------ ---------------- +2022-05-15 20:38:54.688542 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -575 finished +------------------------------ ---------------- +epoch -575 +replay_buffer/size 999047 +trainer/num train calls 426000 +trainer/QF1 Loss 1.10368 +trainer/QF2 Loss 1.21713 +trainer/Policy Loss 1.91459 +trainer/Q1 Predictions Mean -71.6782 +trainer/Q1 Predictions Std 19.2666 +trainer/Q1 Predictions Max -0.351332 +trainer/Q1 Predictions Min -87.1596 +trainer/Q2 Predictions Mean -71.743 +trainer/Q2 Predictions Std 19.2673 +trainer/Q2 Predictions Max -0.392297 +trainer/Q2 Predictions Min -87.009 +trainer/Q Targets Mean -71.1474 +trainer/Q Targets Std 19.4376 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6754 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0254461 +trainer/policy/mean Std 0.720157 +trainer/policy/mean Max 0.998743 +trainer/policy/mean Min -0.998349 +trainer/policy/std Mean 0.424897 +trainer/policy/std Std 0.0211033 +trainer/policy/std Max 0.448187 +trainer/policy/std Min 0.389278 +trainer/Advantage Weights Mean 0.366343 +trainer/Advantage Weights Std 3.74807 +trainer/Advantage Weights Max 58.9928 +trainer/Advantage Weights Min 6.82703e-18 +trainer/Advantage Score Mean -0.783274 +trainer/Advantage Score Std 0.707111 +trainer/Advantage Score Max 0.407742 +trainer/Advantage Score Min -3.95256 +trainer/V1 Predictions Mean -70.9229 +trainer/V1 Predictions Std 19.5699 +trainer/V1 Predictions Max 1.28071 +trainer/V1 Predictions Min -86.4081 +trainer/VF Loss 0.112229 +expl/num steps total 426000 +expl/num paths total 509 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.128755 +expl/Actions Std 0.844187 +expl/Actions Max 2.61936 +expl/Actions Min -2.3117 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 403251 +eval/num paths total 426 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.127987 +eval/Actions Std 0.690222 +eval/Actions Max 0.998587 +eval/Actions Min -0.998881 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.95626e-06 +time/evaluation sampling (s) 4.87951 +time/exploration sampling (s) 6.77394 +time/logging (s) 0.00972067 +time/saving (s) 0.0163324 +time/training (s) 18.7191 +time/epoch (s) 30.3987 +time/total (s) 9387.74 +Epoch -575 +------------------------------ ---------------- +2022-05-15 20:39:24.626952 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -574 finished +------------------------------ ---------------- +epoch -574 +replay_buffer/size 999047 +trainer/num train calls 427000 +trainer/QF1 Loss 0.728046 +trainer/QF2 Loss 0.783039 +trainer/Policy Loss 7.70297 +trainer/Q1 Predictions Mean -74.6734 +trainer/Q1 Predictions Std 15.8708 +trainer/Q1 Predictions Max -2.25154 +trainer/Q1 Predictions Min -86.7515 +trainer/Q2 Predictions Mean -74.5972 +trainer/Q2 Predictions Std 15.9296 +trainer/Q2 Predictions Max -0.778565 +trainer/Q2 Predictions Min -86.7776 +trainer/Q Targets Mean -74.2709 +trainer/Q Targets Std 15.7336 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5788 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0183921 +trainer/policy/mean Std 0.727199 +trainer/policy/mean Max 0.999055 +trainer/policy/mean Min -0.997241 +trainer/policy/std Mean 0.423119 +trainer/policy/std Std 0.0198655 +trainer/policy/std Max 0.445051 +trainer/policy/std Min 0.390303 +trainer/Advantage Weights Mean 1.98981 +trainer/Advantage Weights Std 12.524 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.66408e-16 +trainer/Advantage Score Mean -0.466673 +trainer/Advantage Score Std 0.476816 +trainer/Advantage Score Max 0.658485 +trainer/Advantage Score Min -3.49446 +trainer/V1 Predictions Mean -73.9714 +trainer/V1 Predictions Std 15.9815 +trainer/V1 Predictions Max -0.140392 +trainer/V1 Predictions Min -86.5665 +trainer/VF Loss 0.0498075 +expl/num steps total 427000 +expl/num paths total 511 +expl/path length Mean 500 +expl/path length Std 446 +expl/path length Max 946 +expl/path length Min 54 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.032449 +expl/Actions Std 0.832699 +expl/Actions Max 2.46157 +expl/Actions Min -2.19417 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 403835 +eval/num paths total 427 +eval/path length Mean 584 +eval/path length Std 0 +eval/path length Max 584 +eval/path length Min 584 +eval/Rewards Mean 0.00171233 +eval/Rewards Std 0.0413449 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0122376 +eval/Actions Std 0.752899 +eval/Actions Max 0.9999 +eval/Actions Min -0.998926 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 6.06012e-06 +time/evaluation sampling (s) 4.48759 +time/exploration sampling (s) 6.73309 +time/logging (s) 0.0103375 +time/saving (s) 0.0147733 +time/training (s) 18.6845 +time/epoch (s) 29.9303 +time/total (s) 9417.67 +Epoch -574 +------------------------------ ---------------- +2022-05-15 20:39:55.021069 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -573 finished +------------------------------ ---------------- +epoch -573 +replay_buffer/size 999047 +trainer/num train calls 428000 +trainer/QF1 Loss 0.693689 +trainer/QF2 Loss 0.6623 +trainer/Policy Loss 16.2181 +trainer/Q1 Predictions Mean -72.0228 +trainer/Q1 Predictions Std 20.1233 +trainer/Q1 Predictions Max -0.670879 +trainer/Q1 Predictions Min -87.5984 +trainer/Q2 Predictions Mean -72.049 +trainer/Q2 Predictions Std 20.1759 +trainer/Q2 Predictions Max -0.10073 +trainer/Q2 Predictions Min -87.876 +trainer/Q Targets Mean -72.1498 +trainer/Q Targets Std 19.9472 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9324 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00897483 +trainer/policy/mean Std 0.730274 +trainer/policy/mean Max 0.999384 +trainer/policy/mean Min -0.999544 +trainer/policy/std Mean 0.419453 +trainer/policy/std Std 0.0221251 +trainer/policy/std Max 0.44314 +trainer/policy/std Min 0.38067 +trainer/Advantage Weights Mean 4.43196 +trainer/Advantage Weights Std 18.2159 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.7453e-13 +trainer/Advantage Score Mean -0.263917 +trainer/Advantage Score Std 0.461661 +trainer/Advantage Score Max 1.951 +trainer/Advantage Score Min -2.93767 +trainer/V1 Predictions Mean -71.8624 +trainer/V1 Predictions Std 20.0609 +trainer/V1 Predictions Max -0.406782 +trainer/V1 Predictions Min -86.9586 +trainer/VF Loss 0.0651942 +expl/num steps total 428000 +expl/num paths total 513 +expl/path length Mean 500 +expl/path length Std 11 +expl/path length Max 511 +expl/path length Min 489 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00353108 +expl/Actions Std 0.826283 +expl/Actions Max 2.19923 +expl/Actions Min -2.52651 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 404345 +eval/num paths total 428 +eval/path length Mean 510 +eval/path length Std 0 +eval/path length Max 510 +eval/path length Min 510 +eval/Rewards Mean 0.00196078 +eval/Rewards Std 0.0442373 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0277516 +eval/Actions Std 0.736693 +eval/Actions Max 0.999566 +eval/Actions Min -0.999065 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 5.13671e-06 +time/evaluation sampling (s) 4.91507 +time/exploration sampling (s) 6.71637 +time/logging (s) 0.0105627 +time/saving (s) 0.0189682 +time/training (s) 18.7199 +time/epoch (s) 30.3809 +time/total (s) 9448.06 +Epoch -573 +------------------------------ ---------------- +2022-05-15 20:40:25.535379 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -572 finished +------------------------------ ---------------- +epoch -572 +replay_buffer/size 999047 +trainer/num train calls 429000 +trainer/QF1 Loss 0.494694 +trainer/QF2 Loss 0.57483 +trainer/Policy Loss 21.6288 +trainer/Q1 Predictions Mean -71.493 +trainer/Q1 Predictions Std 19.3347 +trainer/Q1 Predictions Max -0.965559 +trainer/Q1 Predictions Min -86.4049 +trainer/Q2 Predictions Mean -71.441 +trainer/Q2 Predictions Std 19.2492 +trainer/Q2 Predictions Max -0.284149 +trainer/Q2 Predictions Min -86.284 +trainer/Q Targets Mean -71.4621 +trainer/Q Targets Std 19.2494 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5585 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0130379 +trainer/policy/mean Std 0.717817 +trainer/policy/mean Max 0.998655 +trainer/policy/mean Min -0.996875 +trainer/policy/std Mean 0.419791 +trainer/policy/std Std 0.020752 +trainer/policy/std Max 0.439468 +trainer/policy/std Min 0.382464 +trainer/Advantage Weights Mean 6.00465 +trainer/Advantage Weights Std 20.2099 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.70947e-18 +trainer/Advantage Score Mean -0.261303 +trainer/Advantage Score Std 0.643622 +trainer/Advantage Score Max 1.47398 +trainer/Advantage Score Min -4.09103 +trainer/V1 Predictions Mean -71.1594 +trainer/V1 Predictions Std 19.5176 +trainer/V1 Predictions Max 0.908258 +trainer/V1 Predictions Min -85.904 +trainer/VF Loss 0.0704754 +expl/num steps total 429000 +expl/num paths total 514 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0364677 +expl/Actions Std 0.823425 +expl/Actions Max 2.28952 +expl/Actions Min -2.25016 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 405345 +eval/num paths total 429 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.348427 +eval/Actions Std 0.72504 +eval/Actions Max 0.999673 +eval/Actions Min -0.998924 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.27504e-06 +time/evaluation sampling (s) 5.42073 +time/exploration sampling (s) 6.63664 +time/logging (s) 0.0127189 +time/saving (s) 0.0179098 +time/training (s) 18.4133 +time/epoch (s) 30.5013 +time/total (s) 9478.57 +Epoch -572 +------------------------------ ---------------- +2022-05-15 20:40:56.429767 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -571 finished +------------------------------ ---------------- +epoch -571 +replay_buffer/size 999047 +trainer/num train calls 430000 +trainer/QF1 Loss 0.610961 +trainer/QF2 Loss 0.64338 +trainer/Policy Loss 24.7237 +trainer/Q1 Predictions Mean -73.0789 +trainer/Q1 Predictions Std 18.9964 +trainer/Q1 Predictions Max -0.69994 +trainer/Q1 Predictions Min -87.1607 +trainer/Q2 Predictions Mean -73.0534 +trainer/Q2 Predictions Std 19.0222 +trainer/Q2 Predictions Max -1.15494 +trainer/Q2 Predictions Min -87.1904 +trainer/Q Targets Mean -73.2437 +trainer/Q Targets Std 18.9972 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3489 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0208533 +trainer/policy/mean Std 0.718571 +trainer/policy/mean Max 0.998169 +trainer/policy/mean Min -0.997361 +trainer/policy/std Mean 0.422214 +trainer/policy/std Std 0.0215972 +trainer/policy/std Max 0.444771 +trainer/policy/std Min 0.384338 +trainer/Advantage Weights Mean 3.9671 +trainer/Advantage Weights Std 17.5037 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.3272e-12 +trainer/Advantage Score Mean -0.3475 +trainer/Advantage Score Std 0.563307 +trainer/Advantage Score Max 2.43445 +trainer/Advantage Score Min -2.73479 +trainer/V1 Predictions Mean -72.9614 +trainer/V1 Predictions Std 19.0319 +trainer/V1 Predictions Max -0.492635 +trainer/V1 Predictions Min -87.182 +trainer/VF Loss 0.0844594 +expl/num steps total 430000 +expl/num paths total 515 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.288872 +expl/Actions Std 0.851308 +expl/Actions Max 2.16912 +expl/Actions Min -2.51805 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 406345 +eval/num paths total 430 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0848888 +eval/Actions Std 0.663446 +eval/Actions Max 0.999283 +eval/Actions Min -0.999329 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.02079e-06 +time/evaluation sampling (s) 5.17266 +time/exploration sampling (s) 6.90695 +time/logging (s) 0.0111992 +time/saving (s) 0.014865 +time/training (s) 18.7723 +time/epoch (s) 30.878 +time/total (s) 9509.45 +Epoch -571 +------------------------------ ---------------- +2022-05-15 20:41:26.909027 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -570 finished +------------------------------ ---------------- +epoch -570 +replay_buffer/size 999047 +trainer/num train calls 431000 +trainer/QF1 Loss 0.793601 +trainer/QF2 Loss 0.842193 +trainer/Policy Loss 38.2501 +trainer/Q1 Predictions Mean -73.7249 +trainer/Q1 Predictions Std 16.64 +trainer/Q1 Predictions Max -0.758315 +trainer/Q1 Predictions Min -86.0599 +trainer/Q2 Predictions Mean -73.6491 +trainer/Q2 Predictions Std 16.7142 +trainer/Q2 Predictions Max -0.691153 +trainer/Q2 Predictions Min -86.0706 +trainer/Q Targets Mean -74.0028 +trainer/Q Targets Std 16.5715 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6877 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00124681 +trainer/policy/mean Std 0.715878 +trainer/policy/mean Max 0.998626 +trainer/policy/mean Min -0.99837 +trainer/policy/std Mean 0.422297 +trainer/policy/std Std 0.0214847 +trainer/policy/std Max 0.444168 +trainer/policy/std Min 0.383304 +trainer/Advantage Weights Mean 6.29179 +trainer/Advantage Weights Std 19.1469 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.08282e-17 +trainer/Advantage Score Mean -0.244419 +trainer/Advantage Score Std 0.618514 +trainer/Advantage Score Max 1.31048 +trainer/Advantage Score Min -3.77372 +trainer/V1 Predictions Mean -73.6711 +trainer/V1 Predictions Std 16.8921 +trainer/V1 Predictions Max -0.94543 +trainer/V1 Predictions Min -86.3717 +trainer/VF Loss 0.0728583 +expl/num steps total 431000 +expl/num paths total 516 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.11898 +expl/Actions Std 0.87436 +expl/Actions Max 2.33656 +expl/Actions Min -2.52701 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 407201 +eval/num paths total 431 +eval/path length Mean 856 +eval/path length Std 0 +eval/path length Max 856 +eval/path length Min 856 +eval/Rewards Mean 0.00116822 +eval/Rewards Std 0.0341593 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0458291 +eval/Actions Std 0.755164 +eval/Actions Max 0.999723 +eval/Actions Min -0.998867 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.10596e-06 +time/evaluation sampling (s) 4.96485 +time/exploration sampling (s) 6.94774 +time/logging (s) 0.01079 +time/saving (s) 0.0155183 +time/training (s) 18.5296 +time/epoch (s) 30.4686 +time/total (s) 9539.93 +Epoch -570 +------------------------------ ---------------- +2022-05-15 20:41:58.141599 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -569 finished +------------------------------ ---------------- +epoch -569 +replay_buffer/size 999047 +trainer/num train calls 432000 +trainer/QF1 Loss 0.686404 +trainer/QF2 Loss 0.724586 +trainer/Policy Loss 11.4546 +trainer/Q1 Predictions Mean -73.1155 +trainer/Q1 Predictions Std 18.2531 +trainer/Q1 Predictions Max -0.445657 +trainer/Q1 Predictions Min -87.9672 +trainer/Q2 Predictions Mean -73.0489 +trainer/Q2 Predictions Std 18.1891 +trainer/Q2 Predictions Max -0.461046 +trainer/Q2 Predictions Min -87.6363 +trainer/Q Targets Mean -72.9103 +trainer/Q Targets Std 18.3117 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5065 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0188148 +trainer/policy/mean Std 0.706509 +trainer/policy/mean Max 0.999047 +trainer/policy/mean Min -0.997567 +trainer/policy/std Mean 0.42237 +trainer/policy/std Std 0.0213265 +trainer/policy/std Max 0.443779 +trainer/policy/std Min 0.38563 +trainer/Advantage Weights Mean 2.35875 +trainer/Advantage Weights Std 10.2228 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.73267e-12 +trainer/Advantage Score Mean -0.383841 +trainer/Advantage Score Std 0.529592 +trainer/Advantage Score Max 0.63848 +trainer/Advantage Score Min -2.63139 +trainer/V1 Predictions Mean -72.6161 +trainer/V1 Predictions Std 18.4412 +trainer/V1 Predictions Max 0.922618 +trainer/V1 Predictions Min -87.3889 +trainer/VF Loss 0.049715 +expl/num steps total 432000 +expl/num paths total 518 +expl/path length Mean 500 +expl/path length Std 272 +expl/path length Max 772 +expl/path length Min 228 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0425231 +expl/Actions Std 0.827473 +expl/Actions Max 2.14062 +expl/Actions Min -2.33048 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 408201 +eval/num paths total 432 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.139708 +eval/Actions Std 0.683868 +eval/Actions Max 0.999062 +eval/Actions Min -0.998808 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.29752e-05 +time/evaluation sampling (s) 5.48867 +time/exploration sampling (s) 6.54682 +time/logging (s) 0.01186 +time/saving (s) 0.0190624 +time/training (s) 19.1533 +time/epoch (s) 31.2197 +time/total (s) 9571.15 +Epoch -569 +------------------------------ ---------------- +2022-05-15 20:42:28.532506 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -568 finished +------------------------------ ---------------- +epoch -568 +replay_buffer/size 999047 +trainer/num train calls 433000 +trainer/QF1 Loss 0.982016 +trainer/QF2 Loss 0.957544 +trainer/Policy Loss 9.04684 +trainer/Q1 Predictions Mean -71.576 +trainer/Q1 Predictions Std 20.1094 +trainer/Q1 Predictions Max -1.42526 +trainer/Q1 Predictions Min -86.8038 +trainer/Q2 Predictions Mean -71.4954 +trainer/Q2 Predictions Std 20.1118 +trainer/Q2 Predictions Max -1.79701 +trainer/Q2 Predictions Min -86.7542 +trainer/Q Targets Mean -71.4336 +trainer/Q Targets Std 19.8453 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7392 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0265715 +trainer/policy/mean Std 0.719221 +trainer/policy/mean Max 0.998785 +trainer/policy/mean Min -0.998392 +trainer/policy/std Mean 0.422037 +trainer/policy/std Std 0.0202115 +trainer/policy/std Max 0.442816 +trainer/policy/std Min 0.389155 +trainer/Advantage Weights Mean 2.96495 +trainer/Advantage Weights Std 14.6967 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.47439e-14 +trainer/Advantage Score Mean -0.425317 +trainer/Advantage Score Std 0.559588 +trainer/Advantage Score Max 1.42556 +trainer/Advantage Score Min -3.18479 +trainer/V1 Predictions Mean -71.1659 +trainer/V1 Predictions Std 19.9388 +trainer/V1 Predictions Max -1.72597 +trainer/V1 Predictions Min -86.6769 +trainer/VF Loss 0.0641783 +expl/num steps total 433000 +expl/num paths total 519 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0926413 +expl/Actions Std 0.864654 +expl/Actions Max 2.31019 +expl/Actions Min -2.30105 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 408700 +eval/num paths total 433 +eval/path length Mean 499 +eval/path length Std 0 +eval/path length Max 499 +eval/path length Min 499 +eval/Rewards Mean 0.00200401 +eval/Rewards Std 0.0447213 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00872914 +eval/Actions Std 0.732184 +eval/Actions Max 0.99897 +eval/Actions Min -0.999752 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.06483e-05 +time/evaluation sampling (s) 4.8725 +time/exploration sampling (s) 6.21211 +time/logging (s) 0.00918354 +time/saving (s) 0.0176314 +time/training (s) 19.2639 +time/epoch (s) 30.3753 +time/total (s) 9601.54 +Epoch -568 +------------------------------ ---------------- +2022-05-15 20:42:59.304316 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -567 finished +------------------------------ ---------------- +epoch -567 +replay_buffer/size 999047 +trainer/num train calls 434000 +trainer/QF1 Loss 0.619001 +trainer/QF2 Loss 0.584295 +trainer/Policy Loss 19.2146 +trainer/Q1 Predictions Mean -72.9002 +trainer/Q1 Predictions Std 19.5983 +trainer/Q1 Predictions Max -0.620847 +trainer/Q1 Predictions Min -86.8601 +trainer/Q2 Predictions Mean -72.8655 +trainer/Q2 Predictions Std 19.6145 +trainer/Q2 Predictions Max -0.382878 +trainer/Q2 Predictions Min -86.8549 +trainer/Q Targets Mean -73.0109 +trainer/Q Targets Std 19.5825 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.846 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0288049 +trainer/policy/mean Std 0.720019 +trainer/policy/mean Max 0.999352 +trainer/policy/mean Min -0.996955 +trainer/policy/std Mean 0.423136 +trainer/policy/std Std 0.0210487 +trainer/policy/std Max 0.445429 +trainer/policy/std Min 0.38638 +trainer/Advantage Weights Mean 3.52221 +trainer/Advantage Weights Std 13.8299 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.23267e-14 +trainer/Advantage Score Mean -0.333503 +trainer/Advantage Score Std 0.495676 +trainer/Advantage Score Max 0.905815 +trainer/Advantage Score Min -3.2027 +trainer/V1 Predictions Mean -72.7685 +trainer/V1 Predictions Std 19.744 +trainer/V1 Predictions Max 0.0671248 +trainer/V1 Predictions Min -86.8595 +trainer/VF Loss 0.0470493 +expl/num steps total 434000 +expl/num paths total 520 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0582112 +expl/Actions Std 0.893378 +expl/Actions Max 2.40382 +expl/Actions Min -2.32592 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 409537 +eval/num paths total 434 +eval/path length Mean 837 +eval/path length Std 0 +eval/path length Max 837 +eval/path length Min 837 +eval/Rewards Mean 0.00119474 +eval/Rewards Std 0.0345444 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0231561 +eval/Actions Std 0.723242 +eval/Actions Max 0.999714 +eval/Actions Min -0.999344 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 6.41495e-06 +time/evaluation sampling (s) 4.74817 +time/exploration sampling (s) 6.94769 +time/logging (s) 0.0103851 +time/saving (s) 0.0157783 +time/training (s) 19.0317 +time/epoch (s) 30.7537 +time/total (s) 9632.3 +Epoch -567 +------------------------------ ---------------- +2022-05-15 20:43:29.159380 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -566 finished +------------------------------ ---------------- +epoch -566 +replay_buffer/size 999047 +trainer/num train calls 435000 +trainer/QF1 Loss 0.944456 +trainer/QF2 Loss 0.85915 +trainer/Policy Loss 4.52218 +trainer/Q1 Predictions Mean -74.0941 +trainer/Q1 Predictions Std 15.7417 +trainer/Q1 Predictions Max -0.417661 +trainer/Q1 Predictions Min -86.9922 +trainer/Q2 Predictions Mean -74.0767 +trainer/Q2 Predictions Std 15.7306 +trainer/Q2 Predictions Max -0.424778 +trainer/Q2 Predictions Min -87.0917 +trainer/Q Targets Mean -73.5995 +trainer/Q Targets Std 16.0561 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6995 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0156256 +trainer/policy/mean Std 0.733311 +trainer/policy/mean Max 0.999596 +trainer/policy/mean Min -0.998753 +trainer/policy/std Mean 0.422529 +trainer/policy/std Std 0.021422 +trainer/policy/std Max 0.446982 +trainer/policy/std Min 0.385526 +trainer/Advantage Weights Mean 0.602819 +trainer/Advantage Weights Std 6.65342 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.25278e-14 +trainer/Advantage Score Mean -0.686713 +trainer/Advantage Score Std 0.524759 +trainer/Advantage Score Max 1.02782 +trainer/Advantage Score Min -3.1424 +trainer/V1 Predictions Mean -73.3705 +trainer/V1 Predictions Std 16.137 +trainer/V1 Predictions Max 2.15691 +trainer/V1 Predictions Min -86.5072 +trainer/VF Loss 0.07845 +expl/num steps total 435000 +expl/num paths total 522 +expl/path length Mean 500 +expl/path length Std 8 +expl/path length Max 508 +expl/path length Min 492 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0115477 +expl/Actions Std 0.83105 +expl/Actions Max 2.55962 +expl/Actions Min -2.3255 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 410537 +eval/num paths total 435 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.227707 +eval/Actions Std 0.619094 +eval/Actions Max 0.998971 +eval/Actions Min -0.99871 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.11458e-06 +time/evaluation sampling (s) 4.76649 +time/exploration sampling (s) 6.62367 +time/logging (s) 0.0111643 +time/saving (s) 0.0157063 +time/training (s) 18.4252 +time/epoch (s) 29.8423 +time/total (s) 9662.14 +Epoch -566 +------------------------------ ---------------- +2022-05-15 20:44:00.412747 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -565 finished +------------------------------ ---------------- +epoch -565 +replay_buffer/size 999047 +trainer/num train calls 436000 +trainer/QF1 Loss 1.05327 +trainer/QF2 Loss 1.09366 +trainer/Policy Loss 17.4608 +trainer/Q1 Predictions Mean -73.5509 +trainer/Q1 Predictions Std 17.4333 +trainer/Q1 Predictions Max -0.253437 +trainer/Q1 Predictions Min -87.8008 +trainer/Q2 Predictions Mean -73.5265 +trainer/Q2 Predictions Std 17.4515 +trainer/Q2 Predictions Max -0.37711 +trainer/Q2 Predictions Min -87.5768 +trainer/Q Targets Mean -73.2844 +trainer/Q Targets Std 17.4935 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4793 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0205655 +trainer/policy/mean Std 0.713296 +trainer/policy/mean Max 0.998138 +trainer/policy/mean Min -0.997653 +trainer/policy/std Mean 0.425345 +trainer/policy/std Std 0.0211384 +trainer/policy/std Max 0.449644 +trainer/policy/std Min 0.391688 +trainer/Advantage Weights Mean 3.49506 +trainer/Advantage Weights Std 16.3163 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.89296e-16 +trainer/Advantage Score Mean -0.3615 +trainer/Advantage Score Std 0.507249 +trainer/Advantage Score Max 1.29464 +trainer/Advantage Score Min -3.47754 +trainer/V1 Predictions Mean -73.0805 +trainer/V1 Predictions Std 17.3929 +trainer/V1 Predictions Max 0.685432 +trainer/V1 Predictions Min -87.2518 +trainer/VF Loss 0.0507834 +expl/num steps total 436000 +expl/num paths total 524 +expl/path length Mean 500 +expl/path length Std 423 +expl/path length Max 923 +expl/path length Min 77 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0328965 +expl/Actions Std 0.832614 +expl/Actions Max 2.32236 +expl/Actions Min -2.49645 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 411537 +eval/num paths total 436 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.557287 +eval/Actions Std 0.482367 +eval/Actions Max 0.999146 +eval/Actions Min -0.99967 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.29817e-05 +time/evaluation sampling (s) 5.10096 +time/exploration sampling (s) 7.39717 +time/logging (s) 0.0119597 +time/saving (s) 0.0191416 +time/training (s) 18.7117 +time/epoch (s) 31.2409 +time/total (s) 9693.39 +Epoch -565 +------------------------------ ---------------- +2022-05-15 20:44:31.358228 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -564 finished +------------------------------ ---------------- +epoch -564 +replay_buffer/size 999047 +trainer/num train calls 437000 +trainer/QF1 Loss 1.48842 +trainer/QF2 Loss 1.13383 +trainer/Policy Loss 43.615 +trainer/Q1 Predictions Mean -72.2268 +trainer/Q1 Predictions Std 20.8625 +trainer/Q1 Predictions Max -0.624772 +trainer/Q1 Predictions Min -86.6489 +trainer/Q2 Predictions Mean -72.3374 +trainer/Q2 Predictions Std 20.866 +trainer/Q2 Predictions Max -0.500913 +trainer/Q2 Predictions Min -86.7695 +trainer/Q Targets Mean -72.1871 +trainer/Q Targets Std 20.9253 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8182 +trainer/rewards Mean -0.976562 +trainer/rewards Std 0.151288 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0234375 +trainer/terminals Std 0.151288 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0253443 +trainer/policy/mean Std 0.720029 +trainer/policy/mean Max 0.999702 +trainer/policy/mean Min -0.998722 +trainer/policy/std Mean 0.422015 +trainer/policy/std Std 0.0214471 +trainer/policy/std Max 0.444729 +trainer/policy/std Min 0.386291 +trainer/Advantage Weights Mean 9.48992 +trainer/Advantage Weights Std 24.6062 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.34289e-16 +trainer/Advantage Score Mean -0.153001 +trainer/Advantage Score Std 0.585411 +trainer/Advantage Score Max 1.30851 +trainer/Advantage Score Min -3.65465 +trainer/V1 Predictions Mean -71.961 +trainer/V1 Predictions Std 20.8848 +trainer/V1 Predictions Max 0.596723 +trainer/V1 Predictions Min -86.7713 +trainer/VF Loss 0.0779662 +expl/num steps total 437000 +expl/num paths total 525 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.217487 +expl/Actions Std 0.80213 +expl/Actions Max 2.4529 +expl/Actions Min -2.30641 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 412537 +eval/num paths total 437 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0696356 +eval/Actions Std 0.725977 +eval/Actions Max 0.999834 +eval/Actions Min -0.999497 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.06678e-05 +time/evaluation sampling (s) 5.03826 +time/exploration sampling (s) 7.29665 +time/logging (s) 0.0112822 +time/saving (s) 0.0154319 +time/training (s) 18.571 +time/epoch (s) 30.9326 +time/total (s) 9724.33 +Epoch -564 +------------------------------ ---------------- +2022-05-15 20:45:00.718456 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -563 finished +------------------------------ ---------------- +epoch -563 +replay_buffer/size 999047 +trainer/num train calls 438000 +trainer/QF1 Loss 0.899438 +trainer/QF2 Loss 0.835109 +trainer/Policy Loss 19.0119 +trainer/Q1 Predictions Mean -71.9761 +trainer/Q1 Predictions Std 19.1943 +trainer/Q1 Predictions Max -0.419822 +trainer/Q1 Predictions Min -87.2017 +trainer/Q2 Predictions Mean -71.9746 +trainer/Q2 Predictions Std 19.2314 +trainer/Q2 Predictions Max -0.386669 +trainer/Q2 Predictions Min -87.0742 +trainer/Q Targets Mean -71.783 +trainer/Q Targets Std 19.247 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6094 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0247187 +trainer/policy/mean Std 0.726671 +trainer/policy/mean Max 0.99927 +trainer/policy/mean Min -0.997898 +trainer/policy/std Mean 0.420003 +trainer/policy/std Std 0.0212287 +trainer/policy/std Max 0.442802 +trainer/policy/std Min 0.384803 +trainer/Advantage Weights Mean 4.54646 +trainer/Advantage Weights Std 18.1991 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.54292e-17 +trainer/Advantage Score Mean -0.404883 +trainer/Advantage Score Std 0.597937 +trainer/Advantage Score Max 1.48063 +trainer/Advantage Score Min -3.71233 +trainer/V1 Predictions Mean -71.5323 +trainer/V1 Predictions Std 19.2982 +trainer/V1 Predictions Max 0.262803 +trainer/V1 Predictions Min -87.4824 +trainer/VF Loss 0.0776525 +expl/num steps total 438000 +expl/num paths total 527 +expl/path length Mean 500 +expl/path length Std 425 +expl/path length Max 925 +expl/path length Min 75 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0138212 +expl/Actions Std 0.836389 +expl/Actions Max 2.36125 +expl/Actions Min -2.32032 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 413537 +eval/num paths total 438 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.106192 +eval/Actions Std 0.656591 +eval/Actions Max 0.999687 +eval/Actions Min -0.999691 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.15068e-06 +time/evaluation sampling (s) 4.70197 +time/exploration sampling (s) 6.27319 +time/logging (s) 0.0109475 +time/saving (s) 0.0148255 +time/training (s) 18.3451 +time/epoch (s) 29.346 +time/total (s) 9753.68 +Epoch -563 +------------------------------ ---------------- +2022-05-15 20:45:31.133546 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -562 finished +------------------------------ ---------------- +epoch -562 +replay_buffer/size 999047 +trainer/num train calls 439000 +trainer/QF1 Loss 0.644604 +trainer/QF2 Loss 0.642542 +trainer/Policy Loss 11.7809 +trainer/Q1 Predictions Mean -72.9038 +trainer/Q1 Predictions Std 18.0846 +trainer/Q1 Predictions Max -1.12888 +trainer/Q1 Predictions Min -86.8885 +trainer/Q2 Predictions Mean -72.841 +trainer/Q2 Predictions Std 18.0559 +trainer/Q2 Predictions Max 0.231088 +trainer/Q2 Predictions Min -86.8806 +trainer/Q Targets Mean -73.1289 +trainer/Q Targets Std 17.7411 +trainer/Q Targets Max -1.51871 +trainer/Q Targets Min -87.1945 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00792989 +trainer/policy/mean Std 0.725889 +trainer/policy/mean Max 0.999959 +trainer/policy/mean Min -0.999689 +trainer/policy/std Mean 0.42192 +trainer/policy/std Std 0.0206214 +trainer/policy/std Max 0.446439 +trainer/policy/std Min 0.387288 +trainer/Advantage Weights Mean 2.74292 +trainer/Advantage Weights Std 14.0593 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.82224e-14 +trainer/Advantage Score Mean -0.394821 +trainer/Advantage Score Std 0.516423 +trainer/Advantage Score Max 1.97729 +trainer/Advantage Score Min -3.16361 +trainer/V1 Predictions Mean -72.7569 +trainer/V1 Predictions Std 18.1299 +trainer/V1 Predictions Max -0.35687 +trainer/V1 Predictions Min -86.9178 +trainer/VF Loss 0.0689188 +expl/num steps total 439000 +expl/num paths total 528 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0547039 +expl/Actions Std 0.80042 +expl/Actions Max 2.50813 +expl/Actions Min -2.10892 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 414497 +eval/num paths total 439 +eval/path length Mean 960 +eval/path length Std 0 +eval/path length Max 960 +eval/path length Min 960 +eval/Rewards Mean 0.00104167 +eval/Rewards Std 0.032258 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0296313 +eval/Actions Std 0.723725 +eval/Actions Max 0.999328 +eval/Actions Min -0.999559 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.64123e-06 +time/evaluation sampling (s) 5.24732 +time/exploration sampling (s) 6.7891 +time/logging (s) 0.0130315 +time/saving (s) 0.0219617 +time/training (s) 18.3323 +time/epoch (s) 30.4037 +time/total (s) 9784.09 +Epoch -562 +------------------------------ ---------------- +2022-05-15 20:46:01.753766 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -561 finished +------------------------------ ---------------- +epoch -561 +replay_buffer/size 999047 +trainer/num train calls 440000 +trainer/QF1 Loss 5.43721 +trainer/QF2 Loss 5.48661 +trainer/Policy Loss 6.90223 +trainer/Q1 Predictions Mean -74.7349 +trainer/Q1 Predictions Std 15.597 +trainer/Q1 Predictions Max -3.39048 +trainer/Q1 Predictions Min -86.5863 +trainer/Q2 Predictions Mean -74.7637 +trainer/Q2 Predictions Std 15.6619 +trainer/Q2 Predictions Max -2.20581 +trainer/Q2 Predictions Min -86.7136 +trainer/Q Targets Mean -74.6444 +trainer/Q Targets Std 15.6255 +trainer/Q Targets Max -4.04889 +trainer/Q Targets Min -86.7776 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0280871 +trainer/policy/mean Std 0.718546 +trainer/policy/mean Max 0.999013 +trainer/policy/mean Min -0.996987 +trainer/policy/std Mean 0.421522 +trainer/policy/std Std 0.020652 +trainer/policy/std Max 0.443962 +trainer/policy/std Min 0.385709 +trainer/Advantage Weights Mean 1.22305 +trainer/Advantage Weights Std 8.93842 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.11899e-23 +trainer/Advantage Score Mean -0.463035 +trainer/Advantage Score Std 0.519249 +trainer/Advantage Score Max 0.580921 +trainer/Advantage Score Min -5.22085 +trainer/V1 Predictions Mean -74.4796 +trainer/V1 Predictions Std 15.8332 +trainer/V1 Predictions Max -2.05398 +trainer/V1 Predictions Min -86.8146 +trainer/VF Loss 0.0514541 +expl/num steps total 440000 +expl/num paths total 529 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.430277 +expl/Actions Std 0.726047 +expl/Actions Max 2.28186 +expl/Actions Min -2.60659 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 415497 +eval/num paths total 440 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0323696 +eval/Actions Std 0.741546 +eval/Actions Max 0.999244 +eval/Actions Min -0.999134 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.19309e-06 +time/evaluation sampling (s) 4.86084 +time/exploration sampling (s) 6.3595 +time/logging (s) 0.0122805 +time/saving (s) 0.0183823 +time/training (s) 19.3543 +time/epoch (s) 30.6053 +time/total (s) 9814.71 +Epoch -561 +------------------------------ ---------------- +2022-05-15 20:46:31.319042 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -560 finished +------------------------------ ---------------- +epoch -560 +replay_buffer/size 999047 +trainer/num train calls 441000 +trainer/QF1 Loss 1.18122 +trainer/QF2 Loss 0.924058 +trainer/Policy Loss 1.58227 +trainer/Q1 Predictions Mean -73.6847 +trainer/Q1 Predictions Std 18.0992 +trainer/Q1 Predictions Max -0.27953 +trainer/Q1 Predictions Min -87.0761 +trainer/Q2 Predictions Mean -73.5619 +trainer/Q2 Predictions Std 18.0918 +trainer/Q2 Predictions Max -0.398651 +trainer/Q2 Predictions Min -87.0038 +trainer/Q Targets Mean -73.0094 +trainer/Q Targets Std 18.0882 +trainer/Q Targets Max 0.0294048 +trainer/Q Targets Min -86.6954 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00309712 +trainer/policy/mean Std 0.728677 +trainer/policy/mean Max 0.999713 +trainer/policy/mean Min -0.999247 +trainer/policy/std Mean 0.422777 +trainer/policy/std Std 0.0196522 +trainer/policy/std Max 0.444552 +trainer/policy/std Min 0.38892 +trainer/Advantage Weights Mean 0.432779 +trainer/Advantage Weights Std 6.23813 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.3523e-16 +trainer/Advantage Score Mean -0.767418 +trainer/Advantage Score Std 0.489426 +trainer/Advantage Score Max 0.800446 +trainer/Advantage Score Min -3.56317 +trainer/V1 Predictions Mean -72.7252 +trainer/V1 Predictions Std 18.2574 +trainer/V1 Predictions Max 0.872907 +trainer/V1 Predictions Min -86.5812 +trainer/VF Loss 0.0848702 +expl/num steps total 441000 +expl/num paths total 531 +expl/path length Mean 500 +expl/path length Std 473 +expl/path length Max 973 +expl/path length Min 27 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0381475 +expl/Actions Std 0.83167 +expl/Actions Max 2.31046 +expl/Actions Min -2.36687 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 416273 +eval/num paths total 441 +eval/path length Mean 776 +eval/path length Std 0 +eval/path length Max 776 +eval/path length Min 776 +eval/Rewards Mean 0.00128866 +eval/Rewards Std 0.0358748 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0343378 +eval/Actions Std 0.740389 +eval/Actions Max 0.99992 +eval/Actions Min -0.999872 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.09351e-05 +time/evaluation sampling (s) 4.59335 +time/exploration sampling (s) 6.41843 +time/logging (s) 0.0116509 +time/saving (s) 0.0179161 +time/training (s) 18.508 +time/epoch (s) 29.5494 +time/total (s) 9844.26 +Epoch -560 +------------------------------ ---------------- +2022-05-15 20:47:01.650028 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -559 finished +------------------------------ ---------------- +epoch -559 +replay_buffer/size 999047 +trainer/num train calls 442000 +trainer/QF1 Loss 0.714001 +trainer/QF2 Loss 0.719 +trainer/Policy Loss 15.4137 +trainer/Q1 Predictions Mean -73.9895 +trainer/Q1 Predictions Std 16.5221 +trainer/Q1 Predictions Max -0.295815 +trainer/Q1 Predictions Min -86.3248 +trainer/Q2 Predictions Mean -73.9232 +trainer/Q2 Predictions Std 16.5135 +trainer/Q2 Predictions Max -0.388648 +trainer/Q2 Predictions Min -86.0207 +trainer/Q Targets Mean -74.1703 +trainer/Q Targets Std 16.4571 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.0946 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00778588 +trainer/policy/mean Std 0.738555 +trainer/policy/mean Max 0.999552 +trainer/policy/mean Min -0.998585 +trainer/policy/std Mean 0.42071 +trainer/policy/std Std 0.0206012 +trainer/policy/std Max 0.44151 +trainer/policy/std Min 0.38863 +trainer/Advantage Weights Mean 2.03668 +trainer/Advantage Weights Std 10.3514 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.63505e-17 +trainer/Advantage Score Mean -0.404572 +trainer/Advantage Score Std 0.511134 +trainer/Advantage Score Max 1.40713 +trainer/Advantage Score Min -3.74149 +trainer/V1 Predictions Mean -73.8796 +trainer/V1 Predictions Std 16.7177 +trainer/V1 Predictions Max 0.45352 +trainer/V1 Predictions Min -86.1771 +trainer/VF Loss 0.054052 +expl/num steps total 442000 +expl/num paths total 533 +expl/path length Mean 500 +expl/path length Std 398 +expl/path length Max 898 +expl/path length Min 102 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0281599 +expl/Actions Std 0.829052 +expl/Actions Max 2.2587 +expl/Actions Min -2.20992 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 417273 +eval/num paths total 442 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.049057 +eval/Actions Std 0.655723 +eval/Actions Max 0.999585 +eval/Actions Min -0.99961 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.96999e-06 +time/evaluation sampling (s) 4.72452 +time/exploration sampling (s) 6.94293 +time/logging (s) 0.0112036 +time/saving (s) 0.0160141 +time/training (s) 18.6208 +time/epoch (s) 30.3155 +time/total (s) 9874.59 +Epoch -559 +------------------------------ ---------------- +2022-05-15 20:47:31.364808 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -558 finished +------------------------------ ---------------- +epoch -558 +replay_buffer/size 999047 +trainer/num train calls 443000 +trainer/QF1 Loss 0.942479 +trainer/QF2 Loss 0.802065 +trainer/Policy Loss 24.8945 +trainer/Q1 Predictions Mean -72.2647 +trainer/Q1 Predictions Std 16.8377 +trainer/Q1 Predictions Max -1.34166 +trainer/Q1 Predictions Min -86.5911 +trainer/Q2 Predictions Mean -72.3028 +trainer/Q2 Predictions Std 16.7881 +trainer/Q2 Predictions Max -0.776423 +trainer/Q2 Predictions Min -87.0638 +trainer/Q Targets Mean -72.6806 +trainer/Q Targets Std 16.6325 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4734 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00980007 +trainer/policy/mean Std 0.72485 +trainer/policy/mean Max 0.999069 +trainer/policy/mean Min -0.998398 +trainer/policy/std Mean 0.421247 +trainer/policy/std Std 0.0205424 +trainer/policy/std Max 0.442827 +trainer/policy/std Min 0.386987 +trainer/Advantage Weights Mean 7.45792 +trainer/Advantage Weights Std 21.8509 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.19834e-18 +trainer/Advantage Score Mean -0.228925 +trainer/Advantage Score Std 0.58502 +trainer/Advantage Score Max 1.30849 +trainer/Advantage Score Min -3.96223 +trainer/V1 Predictions Mean -72.3686 +trainer/V1 Predictions Std 16.9444 +trainer/V1 Predictions Max -1.3764 +trainer/V1 Predictions Min -87.5046 +trainer/VF Loss 0.0668753 +expl/num steps total 443000 +expl/num paths total 535 +expl/path length Mean 500 +expl/path length Std 139 +expl/path length Max 639 +expl/path length Min 361 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.024341 +expl/Actions Std 0.837305 +expl/Actions Max 2.31802 +expl/Actions Min -2.53499 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 418273 +eval/num paths total 443 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.131435 +eval/Actions Std 0.694071 +eval/Actions Max 0.999484 +eval/Actions Min -0.999762 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.62704e-06 +time/evaluation sampling (s) 4.72204 +time/exploration sampling (s) 6.54497 +time/logging (s) 0.0127596 +time/saving (s) 0.0181482 +time/training (s) 18.4055 +time/epoch (s) 29.7034 +time/total (s) 9904.29 +Epoch -558 +------------------------------ ---------------- +2022-05-15 20:48:01.998982 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -557 finished +------------------------------ ---------------- +epoch -557 +replay_buffer/size 999047 +trainer/num train calls 444000 +trainer/QF1 Loss 0.516348 +trainer/QF2 Loss 0.505436 +trainer/Policy Loss 18.0492 +trainer/Q1 Predictions Mean -73.2643 +trainer/Q1 Predictions Std 18.4066 +trainer/Q1 Predictions Max -1.03633 +trainer/Q1 Predictions Min -87.0034 +trainer/Q2 Predictions Mean -73.2537 +trainer/Q2 Predictions Std 18.4527 +trainer/Q2 Predictions Max -0.632139 +trainer/Q2 Predictions Min -86.9254 +trainer/Q Targets Mean -73.2906 +trainer/Q Targets Std 18.4014 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9087 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0392933 +trainer/policy/mean Std 0.730734 +trainer/policy/mean Max 0.999836 +trainer/policy/mean Min -0.999462 +trainer/policy/std Mean 0.419878 +trainer/policy/std Std 0.019221 +trainer/policy/std Max 0.442642 +trainer/policy/std Min 0.388501 +trainer/Advantage Weights Mean 4.50646 +trainer/Advantage Weights Std 16.753 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.06219e-19 +trainer/Advantage Score Mean -0.282792 +trainer/Advantage Score Std 0.561476 +trainer/Advantage Score Max 0.879017 +trainer/Advantage Score Min -4.36888 +trainer/V1 Predictions Mean -73.032 +trainer/V1 Predictions Std 18.6446 +trainer/V1 Predictions Max 0.346818 +trainer/V1 Predictions Min -86.8756 +trainer/VF Loss 0.0519976 +expl/num steps total 444000 +expl/num paths total 536 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00759482 +expl/Actions Std 0.852446 +expl/Actions Max 2.63072 +expl/Actions Min -2.35523 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 419273 +eval/num paths total 444 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.167952 +eval/Actions Std 0.778956 +eval/Actions Max 0.999878 +eval/Actions Min -0.999463 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.44403e-06 +time/evaluation sampling (s) 5.13995 +time/exploration sampling (s) 6.53917 +time/logging (s) 0.0145203 +time/saving (s) 0.0201338 +time/training (s) 18.9071 +time/epoch (s) 30.6208 +time/total (s) 9934.92 +Epoch -557 +------------------------------ ---------------- +2022-05-15 20:48:32.609146 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -556 finished +------------------------------ ---------------- +epoch -556 +replay_buffer/size 999047 +trainer/num train calls 445000 +trainer/QF1 Loss 0.579188 +trainer/QF2 Loss 0.741738 +trainer/Policy Loss 12.133 +trainer/Q1 Predictions Mean -74.5228 +trainer/Q1 Predictions Std 16.0291 +trainer/Q1 Predictions Max -2.4806 +trainer/Q1 Predictions Min -87.5929 +trainer/Q2 Predictions Mean -74.4381 +trainer/Q2 Predictions Std 16.1186 +trainer/Q2 Predictions Max -3.09896 +trainer/Q2 Predictions Min -87.2019 +trainer/Q Targets Mean -74.4821 +trainer/Q Targets Std 15.7406 +trainer/Q Targets Max -2.53235 +trainer/Q Targets Min -87.2067 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00536941 +trainer/policy/mean Std 0.724933 +trainer/policy/mean Max 0.999191 +trainer/policy/mean Min -0.997749 +trainer/policy/std Mean 0.421381 +trainer/policy/std Std 0.0199694 +trainer/policy/std Max 0.440889 +trainer/policy/std Min 0.389366 +trainer/Advantage Weights Mean 2.00139 +trainer/Advantage Weights Std 11.9557 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.32671e-16 +trainer/Advantage Score Mean -0.415239 +trainer/Advantage Score Std 0.54967 +trainer/Advantage Score Max 1.44291 +trainer/Advantage Score Min -3.48498 +trainer/V1 Predictions Mean -74.1999 +trainer/V1 Predictions Std 16.0563 +trainer/V1 Predictions Max -0.85102 +trainer/V1 Predictions Min -87.074 +trainer/VF Loss 0.0602769 +expl/num steps total 445000 +expl/num paths total 537 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.173458 +expl/Actions Std 0.826357 +expl/Actions Max 2.54384 +expl/Actions Min -2.24824 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 420273 +eval/num paths total 445 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.15262 +eval/Actions Std 0.64112 +eval/Actions Max 0.998848 +eval/Actions Min -0.99945 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.11179e-06 +time/evaluation sampling (s) 5.15349 +time/exploration sampling (s) 6.77901 +time/logging (s) 0.0113995 +time/saving (s) 0.0152129 +time/training (s) 18.6314 +time/epoch (s) 30.5905 +time/total (s) 9965.52 +Epoch -556 +------------------------------ ---------------- +2022-05-15 20:49:02.808177 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -555 finished +------------------------------ ---------------- +epoch -555 +replay_buffer/size 999047 +trainer/num train calls 446000 +trainer/QF1 Loss 2.66134 +trainer/QF2 Loss 2.68457 +trainer/Policy Loss 12.1326 +trainer/Q1 Predictions Mean -72.5809 +trainer/Q1 Predictions Std 18.5341 +trainer/Q1 Predictions Max -0.58924 +trainer/Q1 Predictions Min -86.7094 +trainer/Q2 Predictions Mean -72.5846 +trainer/Q2 Predictions Std 18.5203 +trainer/Q2 Predictions Max -0.396722 +trainer/Q2 Predictions Min -86.8061 +trainer/Q Targets Mean -72.7129 +trainer/Q Targets Std 18.3917 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6204 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0115851 +trainer/policy/mean Std 0.725499 +trainer/policy/mean Max 0.998684 +trainer/policy/mean Min -0.998046 +trainer/policy/std Mean 0.421769 +trainer/policy/std Std 0.0202906 +trainer/policy/std Max 0.44379 +trainer/policy/std Min 0.387377 +trainer/Advantage Weights Mean 2.87413 +trainer/Advantage Weights Std 14.3359 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.13273e-19 +trainer/Advantage Score Mean -0.382699 +trainer/Advantage Score Std 0.62282 +trainer/Advantage Score Max 3.19178 +trainer/Advantage Score Min -4.29917 +trainer/V1 Predictions Mean -72.5133 +trainer/V1 Predictions Std 18.5681 +trainer/V1 Predictions Max -0.661713 +trainer/V1 Predictions Min -86.4181 +trainer/VF Loss 0.101027 +expl/num steps total 446000 +expl/num paths total 539 +expl/path length Mean 500 +expl/path length Std 353 +expl/path length Max 853 +expl/path length Min 147 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0211046 +expl/Actions Std 0.805186 +expl/Actions Max 2.27172 +expl/Actions Min -2.35959 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 420824 +eval/num paths total 446 +eval/path length Mean 551 +eval/path length Std 0 +eval/path length Max 551 +eval/path length Min 551 +eval/Rewards Mean 0.00181488 +eval/Rewards Std 0.0425628 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0331888 +eval/Actions Std 0.732085 +eval/Actions Max 0.999702 +eval/Actions Min -0.999023 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.56998e-05 +time/evaluation sampling (s) 4.97198 +time/exploration sampling (s) 6.08236 +time/logging (s) 0.0111919 +time/saving (s) 0.0165299 +time/training (s) 19.1029 +time/epoch (s) 30.185 +time/total (s) 9995.71 +Epoch -555 +------------------------------ ---------------- +2022-05-15 20:49:34.252459 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -554 finished +------------------------------ ---------------- +epoch -554 +replay_buffer/size 999047 +trainer/num train calls 447000 +trainer/QF1 Loss 0.797828 +trainer/QF2 Loss 0.800165 +trainer/Policy Loss 4.54008 +trainer/Q1 Predictions Mean -73.4692 +trainer/Q1 Predictions Std 17.7396 +trainer/Q1 Predictions Max -0.973138 +trainer/Q1 Predictions Min -87.2261 +trainer/Q2 Predictions Mean -73.4737 +trainer/Q2 Predictions Std 17.7236 +trainer/Q2 Predictions Max -0.869491 +trainer/Q2 Predictions Min -87.8622 +trainer/Q Targets Mean -73.2348 +trainer/Q Targets Std 18.0607 +trainer/Q Targets Max 0.458197 +trainer/Q Targets Min -87.427 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0341952 +trainer/policy/mean Std 0.719847 +trainer/policy/mean Max 0.998565 +trainer/policy/mean Min -0.998887 +trainer/policy/std Mean 0.420748 +trainer/policy/std Std 0.0189907 +trainer/policy/std Max 0.440999 +trainer/policy/std Min 0.388745 +trainer/Advantage Weights Mean 1.15625 +trainer/Advantage Weights Std 7.0183 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.28949e-25 +trainer/Advantage Score Mean -0.584707 +trainer/Advantage Score Std 0.64336 +trainer/Advantage Score Max 0.960125 +trainer/Advantage Score Min -5.55782 +trainer/V1 Predictions Mean -72.9164 +trainer/V1 Predictions Std 18.3304 +trainer/V1 Predictions Max 0.950126 +trainer/V1 Predictions Min -87.5793 +trainer/VF Loss 0.0810164 +expl/num steps total 447000 +expl/num paths total 541 +expl/path length Mean 500 +expl/path length Std 0 +expl/path length Max 500 +expl/path length Min 500 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0235027 +expl/Actions Std 0.833913 +expl/Actions Max 2.22469 +expl/Actions Min -2.42682 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 421477 +eval/num paths total 447 +eval/path length Mean 653 +eval/path length Std 0 +eval/path length Max 653 +eval/path length Min 653 +eval/Rewards Mean 0.00153139 +eval/Rewards Std 0.039103 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0129696 +eval/Actions Std 0.725672 +eval/Actions Max 0.999501 +eval/Actions Min -0.999317 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.2537e-05 +time/evaluation sampling (s) 4.72729 +time/exploration sampling (s) 7.14889 +time/logging (s) 0.0111081 +time/saving (s) 0.0187722 +time/training (s) 19.5233 +time/epoch (s) 31.4294 +time/total (s) 10027.2 +Epoch -554 +------------------------------ ---------------- +2022-05-15 20:50:05.293857 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -553 finished +------------------------------ ---------------- +epoch -553 +replay_buffer/size 999047 +trainer/num train calls 448000 +trainer/QF1 Loss 0.528676 +trainer/QF2 Loss 0.579482 +trainer/Policy Loss 14.4171 +trainer/Q1 Predictions Mean -73.2743 +trainer/Q1 Predictions Std 17.7079 +trainer/Q1 Predictions Max -0.87938 +trainer/Q1 Predictions Min -86.1467 +trainer/Q2 Predictions Mean -73.233 +trainer/Q2 Predictions Std 17.6709 +trainer/Q2 Predictions Max -0.667467 +trainer/Q2 Predictions Min -86.0442 +trainer/Q Targets Mean -73.1281 +trainer/Q Targets Std 17.7325 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2337 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00270356 +trainer/policy/mean Std 0.713773 +trainer/policy/mean Max 0.999892 +trainer/policy/mean Min -0.999556 +trainer/policy/std Mean 0.421635 +trainer/policy/std Std 0.0213578 +trainer/policy/std Max 0.446065 +trainer/policy/std Min 0.386065 +trainer/Advantage Weights Mean 1.54275 +trainer/Advantage Weights Std 10.7831 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.5096e-15 +trainer/Advantage Score Mean -0.522777 +trainer/Advantage Score Std 0.563948 +trainer/Advantage Score Max 1.04512 +trainer/Advantage Score Min -3.41269 +trainer/V1 Predictions Mean -72.8108 +trainer/V1 Predictions Std 17.9161 +trainer/V1 Predictions Max -0.0667667 +trainer/V1 Predictions Min -86.1616 +trainer/VF Loss 0.0663723 +expl/num steps total 448000 +expl/num paths total 543 +expl/path length Mean 500 +expl/path length Std 195 +expl/path length Max 695 +expl/path length Min 305 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0192814 +expl/Actions Std 0.818608 +expl/Actions Max 2.54006 +expl/Actions Min -2.43311 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 422477 +eval/num paths total 448 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0718583 +eval/Actions Std 0.731571 +eval/Actions Max 0.999634 +eval/Actions Min -0.998885 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.09388e-05 +time/evaluation sampling (s) 4.83056 +time/exploration sampling (s) 6.7304 +time/logging (s) 0.00934923 +time/saving (s) 0.0150854 +time/training (s) 19.4387 +time/epoch (s) 31.0241 +time/total (s) 10058.2 +Epoch -553 +------------------------------ ---------------- +2022-05-15 20:50:36.971194 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -552 finished +------------------------------ ---------------- +epoch -552 +replay_buffer/size 999047 +trainer/num train calls 449000 +trainer/QF1 Loss 0.655601 +trainer/QF2 Loss 0.552049 +trainer/Policy Loss 4.69987 +trainer/Q1 Predictions Mean -74.5964 +trainer/Q1 Predictions Std 15.3193 +trainer/Q1 Predictions Max -0.647837 +trainer/Q1 Predictions Min -87.2637 +trainer/Q2 Predictions Mean -74.5841 +trainer/Q2 Predictions Std 15.2359 +trainer/Q2 Predictions Max -0.753679 +trainer/Q2 Predictions Min -87.2619 +trainer/Q Targets Mean -74.3292 +trainer/Q Targets Std 15.1895 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7984 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0046281 +trainer/policy/mean Std 0.727138 +trainer/policy/mean Max 0.99901 +trainer/policy/mean Min -0.997814 +trainer/policy/std Mean 0.422095 +trainer/policy/std Std 0.021254 +trainer/policy/std Max 0.444034 +trainer/policy/std Min 0.386317 +trainer/Advantage Weights Mean 1.66819 +trainer/Advantage Weights Std 11.6033 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.67397e-15 +trainer/Advantage Score Mean -0.54086 +trainer/Advantage Score Std 0.518226 +trainer/Advantage Score Max 1.15341 +trainer/Advantage Score Min -3.35552 +trainer/V1 Predictions Mean -74.0674 +trainer/V1 Predictions Std 15.2857 +trainer/V1 Predictions Max -0.420024 +trainer/V1 Predictions Min -86.4927 +trainer/VF Loss 0.0665618 +expl/num steps total 449000 +expl/num paths total 544 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00500978 +expl/Actions Std 0.850753 +expl/Actions Max 2.42066 +expl/Actions Min -2.3657 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 423477 +eval/num paths total 449 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00495896 +eval/Actions Std 0.803241 +eval/Actions Max 0.999716 +eval/Actions Min -0.999074 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.28001e-05 +time/evaluation sampling (s) 5.36741 +time/exploration sampling (s) 6.97278 +time/logging (s) 0.0113904 +time/saving (s) 0.0162131 +time/training (s) 19.2986 +time/epoch (s) 31.6664 +time/total (s) 10089.9 +Epoch -552 +------------------------------ ---------------- +2022-05-15 20:51:07.284296 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -551 finished +------------------------------ ---------------- +epoch -551 +replay_buffer/size 999047 +trainer/num train calls 450000 +trainer/QF1 Loss 1.05411 +trainer/QF2 Loss 0.972421 +trainer/Policy Loss 44.9858 +trainer/Q1 Predictions Mean -71.9475 +trainer/Q1 Predictions Std 20.1697 +trainer/Q1 Predictions Max -1.11545 +trainer/Q1 Predictions Min -86.807 +trainer/Q2 Predictions Mean -72.006 +trainer/Q2 Predictions Std 20.1482 +trainer/Q2 Predictions Max -1.23567 +trainer/Q2 Predictions Min -86.9498 +trainer/Q Targets Mean -72.0412 +trainer/Q Targets Std 19.9505 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7048 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0343793 +trainer/policy/mean Std 0.724034 +trainer/policy/mean Max 0.999627 +trainer/policy/mean Min -0.998019 +trainer/policy/std Mean 0.421678 +trainer/policy/std Std 0.0205504 +trainer/policy/std Max 0.440872 +trainer/policy/std Min 0.387088 +trainer/Advantage Weights Mean 9.23329 +trainer/Advantage Weights Std 25.337 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.24072e-16 +trainer/Advantage Score Mean -0.312569 +trainer/Advantage Score Std 0.627828 +trainer/Advantage Score Max 1.42992 +trainer/Advantage Score Min -3.48616 +trainer/V1 Predictions Mean -71.7949 +trainer/V1 Predictions Std 20.1177 +trainer/V1 Predictions Max 0.0092802 +trainer/V1 Predictions Min -86.3821 +trainer/VF Loss 0.0917099 +expl/num steps total 450000 +expl/num paths total 545 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0392633 +expl/Actions Std 0.836693 +expl/Actions Max 2.4167 +expl/Actions Min -2.20782 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 424477 +eval/num paths total 450 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0608691 +eval/Actions Std 0.742389 +eval/Actions Max 0.99979 +eval/Actions Min -0.99878 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.75976e-06 +time/evaluation sampling (s) 4.50118 +time/exploration sampling (s) 6.80923 +time/logging (s) 0.0136726 +time/saving (s) 0.0124468 +time/training (s) 18.9646 +time/epoch (s) 30.3012 +time/total (s) 10120.2 +Epoch -551 +------------------------------ ---------------- +2022-05-15 20:51:37.817469 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -550 finished +------------------------------ ---------------- +epoch -550 +replay_buffer/size 999047 +trainer/num train calls 451000 +trainer/QF1 Loss 1.05 +trainer/QF2 Loss 1.06011 +trainer/Policy Loss 73.475 +trainer/Q1 Predictions Mean -70.4853 +trainer/Q1 Predictions Std 21.4925 +trainer/Q1 Predictions Max -0.813768 +trainer/Q1 Predictions Min -86.6369 +trainer/Q2 Predictions Mean -70.5036 +trainer/Q2 Predictions Std 21.4465 +trainer/Q2 Predictions Max -0.460835 +trainer/Q2 Predictions Min -86.6441 +trainer/Q Targets Mean -70.957 +trainer/Q Targets Std 21.73 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9972 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0228019 +trainer/policy/mean Std 0.734192 +trainer/policy/mean Max 0.999304 +trainer/policy/mean Min -0.998709 +trainer/policy/std Mean 0.421702 +trainer/policy/std Std 0.0201051 +trainer/policy/std Max 0.441084 +trainer/policy/std Min 0.388831 +trainer/Advantage Weights Mean 18.398 +trainer/Advantage Weights Std 32.1485 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.29841e-16 +trainer/Advantage Score Mean -0.0138718 +trainer/Advantage Score Std 0.594534 +trainer/Advantage Score Max 3.00903 +trainer/Advantage Score Min -3.65802 +trainer/V1 Predictions Mean -70.7919 +trainer/V1 Predictions Std 21.737 +trainer/V1 Predictions Max -1.23503 +trainer/V1 Predictions Min -87.0201 +trainer/VF Loss 0.121804 +expl/num steps total 451000 +expl/num paths total 547 +expl/path length Mean 500 +expl/path length Std 197 +expl/path length Max 697 +expl/path length Min 303 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0143832 +expl/Actions Std 0.831265 +expl/Actions Max 2.39946 +expl/Actions Min -2.35524 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 425477 +eval/num paths total 451 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0601562 +eval/Actions Std 0.725388 +eval/Actions Max 0.999044 +eval/Actions Min -0.999616 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.93488e-06 +time/evaluation sampling (s) 4.5268 +time/exploration sampling (s) 6.87966 +time/logging (s) 0.011561 +time/saving (s) 0.0279375 +time/training (s) 19.0727 +time/epoch (s) 30.5187 +time/total (s) 10150.7 +Epoch -550 +------------------------------ ---------------- +2022-05-15 20:52:09.488942 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -549 finished +------------------------------ ---------------- +epoch -549 +replay_buffer/size 999047 +trainer/num train calls 452000 +trainer/QF1 Loss 0.507677 +trainer/QF2 Loss 0.555891 +trainer/Policy Loss 34.4619 +trainer/Q1 Predictions Mean -73.3828 +trainer/Q1 Predictions Std 17.5946 +trainer/Q1 Predictions Max -4.33592 +trainer/Q1 Predictions Min -86.3839 +trainer/Q2 Predictions Mean -73.4466 +trainer/Q2 Predictions Std 17.5621 +trainer/Q2 Predictions Max -3.88553 +trainer/Q2 Predictions Min -86.3657 +trainer/Q Targets Mean -73.6225 +trainer/Q Targets Std 17.6082 +trainer/Q Targets Max -4.62117 +trainer/Q Targets Min -86.8533 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00602151 +trainer/policy/mean Std 0.715369 +trainer/policy/mean Max 0.999498 +trainer/policy/mean Min -0.997864 +trainer/policy/std Mean 0.421442 +trainer/policy/std Std 0.0208257 +trainer/policy/std Max 0.444293 +trainer/policy/std Min 0.386355 +trainer/Advantage Weights Mean 6.6392 +trainer/Advantage Weights Std 19.3796 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.69771e-17 +trainer/Advantage Score Mean -0.238112 +trainer/Advantage Score Std 0.614672 +trainer/Advantage Score Max 2.68762 +trainer/Advantage Score Min -3.81515 +trainer/V1 Predictions Mean -73.3944 +trainer/V1 Predictions Std 17.7003 +trainer/V1 Predictions Max -3.84652 +trainer/V1 Predictions Min -86.7128 +trainer/VF Loss 0.0923848 +expl/num steps total 452000 +expl/num paths total 548 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.28402 +expl/Actions Std 0.775347 +expl/Actions Max 2.48449 +expl/Actions Min -2.3517 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 426477 +eval/num paths total 452 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.088906 +eval/Actions Std 0.7129 +eval/Actions Max 0.99901 +eval/Actions Min -0.998444 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 6.29667e-06 +time/evaluation sampling (s) 5.08345 +time/exploration sampling (s) 7.77941 +time/logging (s) 0.0122836 +time/saving (s) 0.0235736 +time/training (s) 18.7597 +time/epoch (s) 31.6585 +time/total (s) 10182.4 +Epoch -549 +------------------------------ ---------------- +2022-05-15 20:52:39.039430 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -548 finished +------------------------------ ---------------- +epoch -548 +replay_buffer/size 999047 +trainer/num train calls 453000 +trainer/QF1 Loss 0.620264 +trainer/QF2 Loss 0.745254 +trainer/Policy Loss 20.3509 +trainer/Q1 Predictions Mean -73.1133 +trainer/Q1 Predictions Std 18.6718 +trainer/Q1 Predictions Max -0.552277 +trainer/Q1 Predictions Min -86.5897 +trainer/Q2 Predictions Mean -73.0797 +trainer/Q2 Predictions Std 18.6957 +trainer/Q2 Predictions Max 0.480385 +trainer/Q2 Predictions Min -86.6938 +trainer/Q Targets Mean -73.1198 +trainer/Q Targets Std 18.5713 +trainer/Q Targets Max -0.44181 +trainer/Q Targets Min -86.5998 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0340215 +trainer/policy/mean Std 0.718096 +trainer/policy/mean Max 0.999393 +trainer/policy/mean Min -0.998044 +trainer/policy/std Mean 0.419174 +trainer/policy/std Std 0.0192874 +trainer/policy/std Max 0.438102 +trainer/policy/std Min 0.387806 +trainer/Advantage Weights Mean 4.41047 +trainer/Advantage Weights Std 17.8586 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.94279e-20 +trainer/Advantage Score Mean -0.382594 +trainer/Advantage Score Std 0.698219 +trainer/Advantage Score Max 2.5256 +trainer/Advantage Score Min -4.46798 +trainer/V1 Predictions Mean -72.8591 +trainer/V1 Predictions Std 18.7155 +trainer/V1 Predictions Max 0.225589 +trainer/V1 Predictions Min -86.4689 +trainer/VF Loss 0.111759 +expl/num steps total 453000 +expl/num paths total 550 +expl/path length Mean 500 +expl/path length Std 203 +expl/path length Max 703 +expl/path length Min 297 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0172511 +expl/Actions Std 0.822497 +expl/Actions Max 2.17681 +expl/Actions Min -2.38554 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 427477 +eval/num paths total 453 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0892404 +eval/Actions Std 0.708232 +eval/Actions Max 0.999483 +eval/Actions Min -0.998737 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.04448e-05 +time/evaluation sampling (s) 4.78098 +time/exploration sampling (s) 6.0865 +time/logging (s) 0.0165886 +time/saving (s) 0.01804 +time/training (s) 18.6375 +time/epoch (s) 29.5397 +time/total (s) 10211.9 +Epoch -548 +------------------------------ ---------------- +2022-05-15 20:53:08.923043 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -547 finished +------------------------------ ---------------- +epoch -547 +replay_buffer/size 999047 +trainer/num train calls 454000 +trainer/QF1 Loss 1.35131 +trainer/QF2 Loss 1.15602 +trainer/Policy Loss 29.5104 +trainer/Q1 Predictions Mean -72.8366 +trainer/Q1 Predictions Std 19.1485 +trainer/Q1 Predictions Max -0.886329 +trainer/Q1 Predictions Min -87.7926 +trainer/Q2 Predictions Mean -72.7553 +trainer/Q2 Predictions Std 19.0682 +trainer/Q2 Predictions Max -0.079824 +trainer/Q2 Predictions Min -87.5812 +trainer/Q Targets Mean -72.9365 +trainer/Q Targets Std 18.6677 +trainer/Q Targets Max -0.611522 +trainer/Q Targets Min -87.9843 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0332201 +trainer/policy/mean Std 0.716799 +trainer/policy/mean Max 0.999648 +trainer/policy/mean Min -0.997048 +trainer/policy/std Mean 0.420438 +trainer/policy/std Std 0.0206148 +trainer/policy/std Max 0.441452 +trainer/policy/std Min 0.390462 +trainer/Advantage Weights Mean 7.03178 +trainer/Advantage Weights Std 21.9586 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.57557e-15 +trainer/Advantage Score Mean -0.207888 +trainer/Advantage Score Std 0.540283 +trainer/Advantage Score Max 1.36609 +trainer/Advantage Score Min -3.25138 +trainer/V1 Predictions Mean -72.6479 +trainer/V1 Predictions Std 18.9223 +trainer/V1 Predictions Max -0.100359 +trainer/V1 Predictions Min -87.8638 +trainer/VF Loss 0.075155 +expl/num steps total 454000 +expl/num paths total 552 +expl/path length Mean 500 +expl/path length Std 55 +expl/path length Max 555 +expl/path length Min 445 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.037474 +expl/Actions Std 0.847009 +expl/Actions Max 2.46261 +expl/Actions Min -2.70585 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 428477 +eval/num paths total 454 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00717779 +eval/Actions Std 0.65357 +eval/Actions Max 0.999385 +eval/Actions Min -0.997768 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.30902e-05 +time/evaluation sampling (s) 4.65966 +time/exploration sampling (s) 6.39278 +time/logging (s) 0.00916672 +time/saving (s) 0.013091 +time/training (s) 18.7863 +time/epoch (s) 29.861 +time/total (s) 10241.8 +Epoch -547 +------------------------------ ---------------- +2022-05-15 20:53:39.676123 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -546 finished +------------------------------ ---------------- +epoch -546 +replay_buffer/size 999047 +trainer/num train calls 455000 +trainer/QF1 Loss 1.35106 +trainer/QF2 Loss 1.22426 +trainer/Policy Loss 23.0213 +trainer/Q1 Predictions Mean -74.4255 +trainer/Q1 Predictions Std 16.1544 +trainer/Q1 Predictions Max -0.733258 +trainer/Q1 Predictions Min -87.1455 +trainer/Q2 Predictions Mean -74.4423 +trainer/Q2 Predictions Std 16.1336 +trainer/Q2 Predictions Max -0.686094 +trainer/Q2 Predictions Min -86.8706 +trainer/Q Targets Mean -74.6665 +trainer/Q Targets Std 15.8697 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6683 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00440558 +trainer/policy/mean Std 0.711748 +trainer/policy/mean Max 0.997573 +trainer/policy/mean Min -0.999092 +trainer/policy/std Mean 0.422491 +trainer/policy/std Std 0.020208 +trainer/policy/std Max 0.443463 +trainer/policy/std Min 0.390561 +trainer/Advantage Weights Mean 6.25439 +trainer/Advantage Weights Std 22.0914 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.85524e-21 +trainer/Advantage Score Mean -0.283708 +trainer/Advantage Score Std 0.556717 +trainer/Advantage Score Max 1.61539 +trainer/Advantage Score Min -4.60663 +trainer/V1 Predictions Mean -74.4347 +trainer/V1 Predictions Std 16.0484 +trainer/V1 Predictions Max -0.705384 +trainer/V1 Predictions Min -86.6199 +trainer/VF Loss 0.0756781 +expl/num steps total 455000 +expl/num paths total 554 +expl/path length Mean 500 +expl/path length Std 120 +expl/path length Max 620 +expl/path length Min 380 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0319825 +expl/Actions Std 0.838173 +expl/Actions Max 2.23023 +expl/Actions Min -2.19464 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 429477 +eval/num paths total 455 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0667746 +eval/Actions Std 0.650222 +eval/Actions Max 0.999547 +eval/Actions Min -0.999219 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.32821e-05 +time/evaluation sampling (s) 4.71404 +time/exploration sampling (s) 7.06064 +time/logging (s) 0.0127963 +time/saving (s) 0.0184265 +time/training (s) 18.9386 +time/epoch (s) 30.7445 +time/total (s) 10272.5 +Epoch -546 +------------------------------ ---------------- +2022-05-15 20:54:10.608061 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -545 finished +------------------------------ ---------------- +epoch -545 +replay_buffer/size 999047 +trainer/num train calls 456000 +trainer/QF1 Loss 1.62169 +trainer/QF2 Loss 1.431 +trainer/Policy Loss 21.3464 +trainer/Q1 Predictions Mean -73.1793 +trainer/Q1 Predictions Std 17.7343 +trainer/Q1 Predictions Max -5.0127 +trainer/Q1 Predictions Min -87.4206 +trainer/Q2 Predictions Mean -73.1621 +trainer/Q2 Predictions Std 17.6899 +trainer/Q2 Predictions Max -4.7608 +trainer/Q2 Predictions Min -87.3743 +trainer/Q Targets Mean -72.9767 +trainer/Q Targets Std 18.0262 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7855 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0183196 +trainer/policy/mean Std 0.709085 +trainer/policy/mean Max 0.999671 +trainer/policy/mean Min -0.999231 +trainer/policy/std Mean 0.423524 +trainer/policy/std Std 0.019956 +trainer/policy/std Max 0.44312 +trainer/policy/std Min 0.392242 +trainer/Advantage Weights Mean 4.13698 +trainer/Advantage Weights Std 16.2021 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.68806e-19 +trainer/Advantage Score Mean -0.326062 +trainer/Advantage Score Std 0.605132 +trainer/Advantage Score Max 1.52737 +trainer/Advantage Score Min -4.27603 +trainer/V1 Predictions Mean -72.805 +trainer/V1 Predictions Std 17.7039 +trainer/V1 Predictions Max -4.43321 +trainer/V1 Predictions Min -86.6616 +trainer/VF Loss 0.0672592 +expl/num steps total 456000 +expl/num paths total 556 +expl/path length Mean 500 +expl/path length Std 344 +expl/path length Max 844 +expl/path length Min 156 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00691321 +expl/Actions Std 0.829592 +expl/Actions Max 2.29708 +expl/Actions Min -2.23752 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 430477 +eval/num paths total 456 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.235564 +eval/Actions Std 0.692373 +eval/Actions Max 0.99965 +eval/Actions Min -0.998054 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.91016e-06 +time/evaluation sampling (s) 4.82524 +time/exploration sampling (s) 6.72849 +time/logging (s) 0.00829253 +time/saving (s) 0.0143424 +time/training (s) 19.3361 +time/epoch (s) 30.9125 +time/total (s) 10303.4 +Epoch -545 +------------------------------ ---------------- +2022-05-15 20:54:40.920644 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -544 finished +------------------------------ ---------------- +epoch -544 +replay_buffer/size 999047 +trainer/num train calls 457000 +trainer/QF1 Loss 0.63074 +trainer/QF2 Loss 0.578712 +trainer/Policy Loss 22.6446 +trainer/Q1 Predictions Mean -72.0815 +trainer/Q1 Predictions Std 18.8923 +trainer/Q1 Predictions Max -0.250647 +trainer/Q1 Predictions Min -86.2037 +trainer/Q2 Predictions Mean -72.1197 +trainer/Q2 Predictions Std 18.9568 +trainer/Q2 Predictions Max -0.287511 +trainer/Q2 Predictions Min -86.194 +trainer/Q Targets Mean -72.2188 +trainer/Q Targets Std 19.055 +trainer/Q Targets Max 0.058338 +trainer/Q Targets Min -86.55 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00637714 +trainer/policy/mean Std 0.720214 +trainer/policy/mean Max 0.999472 +trainer/policy/mean Min -0.996497 +trainer/policy/std Mean 0.421605 +trainer/policy/std Std 0.0208679 +trainer/policy/std Max 0.443306 +trainer/policy/std Min 0.390408 +trainer/Advantage Weights Mean 6.70271 +trainer/Advantage Weights Std 20.7566 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.82752e-18 +trainer/Advantage Score Mean -0.39361 +trainer/Advantage Score Std 0.666519 +trainer/Advantage Score Max 1.99826 +trainer/Advantage Score Min -3.93889 +trainer/V1 Predictions Mean -71.9163 +trainer/V1 Predictions Std 19.2215 +trainer/V1 Predictions Max 1.11004 +trainer/V1 Predictions Min -86.1801 +trainer/VF Loss 0.10157 +expl/num steps total 457000 +expl/num paths total 558 +expl/path length Mean 500 +expl/path length Std 138 +expl/path length Max 638 +expl/path length Min 362 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0333931 +expl/Actions Std 0.821327 +expl/Actions Max 2.48401 +expl/Actions Min -2.2595 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 431477 +eval/num paths total 457 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0443967 +eval/Actions Std 0.791286 +eval/Actions Max 0.999871 +eval/Actions Min -0.999314 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.35793e-06 +time/evaluation sampling (s) 4.67459 +time/exploration sampling (s) 6.65488 +time/logging (s) 0.0122576 +time/saving (s) 0.0136729 +time/training (s) 18.9517 +time/epoch (s) 30.3071 +time/total (s) 10333.7 +Epoch -544 +------------------------------ ---------------- +2022-05-15 20:55:11.824451 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -543 finished +------------------------------ ---------------- +epoch -543 +replay_buffer/size 999047 +trainer/num train calls 458000 +trainer/QF1 Loss 0.782699 +trainer/QF2 Loss 0.727284 +trainer/Policy Loss 9.32403 +trainer/Q1 Predictions Mean -71.2901 +trainer/Q1 Predictions Std 20.2063 +trainer/Q1 Predictions Max -0.524622 +trainer/Q1 Predictions Min -86.6672 +trainer/Q2 Predictions Mean -71.2789 +trainer/Q2 Predictions Std 20.2436 +trainer/Q2 Predictions Max -0.827581 +trainer/Q2 Predictions Min -86.4428 +trainer/Q Targets Mean -71.0252 +trainer/Q Targets Std 20.4485 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.153 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00809784 +trainer/policy/mean Std 0.718068 +trainer/policy/mean Max 0.99937 +trainer/policy/mean Min -0.998977 +trainer/policy/std Mean 0.423206 +trainer/policy/std Std 0.0198343 +trainer/policy/std Max 0.442818 +trainer/policy/std Min 0.388977 +trainer/Advantage Weights Mean 2.05 +trainer/Advantage Weights Std 10.4549 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.1557e-15 +trainer/Advantage Score Mean -0.414108 +trainer/Advantage Score Std 0.563977 +trainer/Advantage Score Max 1.87767 +trainer/Advantage Score Min -3.43941 +trainer/V1 Predictions Mean -70.8245 +trainer/V1 Predictions Std 20.5414 +trainer/V1 Predictions Max -0.0394717 +trainer/V1 Predictions Min -85.982 +trainer/VF Loss 0.0639771 +expl/num steps total 458000 +expl/num paths total 560 +expl/path length Mean 500 +expl/path length Std 261 +expl/path length Max 761 +expl/path length Min 239 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00998017 +expl/Actions Std 0.815984 +expl/Actions Max 2.5575 +expl/Actions Min -2.43149 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 432477 +eval/num paths total 458 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.192648 +eval/Actions Std 0.701198 +eval/Actions Max 0.999729 +eval/Actions Min -0.999856 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.03386e-05 +time/evaluation sampling (s) 4.758 +time/exploration sampling (s) 7.25154 +time/logging (s) 0.0104283 +time/saving (s) 0.0195788 +time/training (s) 18.8489 +time/epoch (s) 30.8885 +time/total (s) 10364.6 +Epoch -543 +------------------------------ ---------------- +2022-05-15 20:55:43.200487 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -542 finished +------------------------------ ---------------- +epoch -542 +replay_buffer/size 999047 +trainer/num train calls 459000 +trainer/QF1 Loss 0.820831 +trainer/QF2 Loss 0.777324 +trainer/Policy Loss 4.41546 +trainer/Q1 Predictions Mean -73.9574 +trainer/Q1 Predictions Std 17.4048 +trainer/Q1 Predictions Max -0.454195 +trainer/Q1 Predictions Min -86.779 +trainer/Q2 Predictions Mean -73.9502 +trainer/Q2 Predictions Std 17.4406 +trainer/Q2 Predictions Max -0.386441 +trainer/Q2 Predictions Min -86.6252 +trainer/Q Targets Mean -73.8145 +trainer/Q Targets Std 17.4508 +trainer/Q Targets Max 3.87453 +trainer/Q Targets Min -85.9859 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0432953 +trainer/policy/mean Std 0.729445 +trainer/policy/mean Max 0.999138 +trainer/policy/mean Min -0.999528 +trainer/policy/std Mean 0.422577 +trainer/policy/std Std 0.0200918 +trainer/policy/std Max 0.443785 +trainer/policy/std Min 0.390143 +trainer/Advantage Weights Mean 1.39367 +trainer/Advantage Weights Std 9.32872 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.43883e-21 +trainer/Advantage Score Mean -0.574471 +trainer/Advantage Score Std 0.553521 +trainer/Advantage Score Max 0.747604 +trainer/Advantage Score Min -4.62214 +trainer/V1 Predictions Mean -73.5629 +trainer/V1 Predictions Std 17.564 +trainer/V1 Predictions Max 4.16562 +trainer/V1 Predictions Min -85.8968 +trainer/VF Loss 0.0679721 +expl/num steps total 459000 +expl/num paths total 561 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00345696 +expl/Actions Std 0.886684 +expl/Actions Max 2.4539 +expl/Actions Min -2.27711 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 433263 +eval/num paths total 459 +eval/path length Mean 786 +eval/path length Std 0 +eval/path length Max 786 +eval/path length Min 786 +eval/Rewards Mean 0.00127226 +eval/Rewards Std 0.0356461 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0146372 +eval/Actions Std 0.739373 +eval/Actions Max 0.999843 +eval/Actions Min -0.998368 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.16499e-05 +time/evaluation sampling (s) 4.85284 +time/exploration sampling (s) 6.77667 +time/logging (s) 0.00787958 +time/saving (s) 0.0133925 +time/training (s) 19.7077 +time/epoch (s) 31.3585 +time/total (s) 10396 +Epoch -542 +------------------------------ ---------------- +2022-05-15 20:56:13.395624 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -541 finished +------------------------------ ---------------- +epoch -541 +replay_buffer/size 999047 +trainer/num train calls 460000 +trainer/QF1 Loss 0.503357 +trainer/QF2 Loss 0.571789 +trainer/Policy Loss 5.88437 +trainer/Q1 Predictions Mean -74.3277 +trainer/Q1 Predictions Std 16.5803 +trainer/Q1 Predictions Max -0.559661 +trainer/Q1 Predictions Min -86.1723 +trainer/Q2 Predictions Mean -74.2873 +trainer/Q2 Predictions Std 16.5811 +trainer/Q2 Predictions Max 1.04616 +trainer/Q2 Predictions Min -86.1287 +trainer/Q Targets Mean -74.7557 +trainer/Q Targets Std 16.5675 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6552 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0250861 +trainer/policy/mean Std 0.719101 +trainer/policy/mean Max 0.998805 +trainer/policy/mean Min -0.995443 +trainer/policy/std Mean 0.421208 +trainer/policy/std Std 0.0201135 +trainer/policy/std Max 0.441622 +trainer/policy/std Min 0.385203 +trainer/Advantage Weights Mean 1.97222 +trainer/Advantage Weights Std 11.5174 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.84118e-13 +trainer/Advantage Score Mean -0.358811 +trainer/Advantage Score Std 0.412225 +trainer/Advantage Score Max 0.854576 +trainer/Advantage Score Min -2.83564 +trainer/V1 Predictions Mean -74.5233 +trainer/V1 Predictions Std 16.6173 +trainer/V1 Predictions Max -0.372378 +trainer/V1 Predictions Min -86.4922 +trainer/VF Loss 0.0367214 +expl/num steps total 460000 +expl/num paths total 562 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.231745 +expl/Actions Std 0.878653 +expl/Actions Max 2.29577 +expl/Actions Min -2.68641 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 433806 +eval/num paths total 460 +eval/path length Mean 543 +eval/path length Std 0 +eval/path length Max 543 +eval/path length Min 543 +eval/Rewards Mean 0.00184162 +eval/Rewards Std 0.0428746 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.0165965 +eval/Actions Std 0.740948 +eval/Actions Max 0.998737 +eval/Actions Min -0.999657 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.30571e-05 +time/evaluation sampling (s) 4.6106 +time/exploration sampling (s) 6.75836 +time/logging (s) 0.00955926 +time/saving (s) 0.0150537 +time/training (s) 18.788 +time/epoch (s) 30.1816 +time/total (s) 10426.2 +Epoch -541 +------------------------------ ---------------- +2022-05-15 20:56:43.200057 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -540 finished +------------------------------ ---------------- +epoch -540 +replay_buffer/size 999047 +trainer/num train calls 461000 +trainer/QF1 Loss 0.625466 +trainer/QF2 Loss 0.53266 +trainer/Policy Loss 8.80952 +trainer/Q1 Predictions Mean -73.4785 +trainer/Q1 Predictions Std 16.7286 +trainer/Q1 Predictions Max -0.345538 +trainer/Q1 Predictions Min -86.739 +trainer/Q2 Predictions Mean -73.4982 +trainer/Q2 Predictions Std 16.6652 +trainer/Q2 Predictions Max -0.0668445 +trainer/Q2 Predictions Min -86.7803 +trainer/Q Targets Mean -73.4527 +trainer/Q Targets Std 16.8448 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5364 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00775938 +trainer/policy/mean Std 0.725888 +trainer/policy/mean Max 0.998591 +trainer/policy/mean Min -0.999666 +trainer/policy/std Mean 0.421463 +trainer/policy/std Std 0.0215146 +trainer/policy/std Max 0.444553 +trainer/policy/std Min 0.385537 +trainer/Advantage Weights Mean 2.04709 +trainer/Advantage Weights Std 11.1956 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.14378e-14 +trainer/Advantage Score Mean -0.407708 +trainer/Advantage Score Std 0.51794 +trainer/Advantage Score Max 0.674191 +trainer/Advantage Score Min -3.04208 +trainer/V1 Predictions Mean -73.1432 +trainer/V1 Predictions Std 17.0288 +trainer/V1 Predictions Max 0.841588 +trainer/V1 Predictions Min -86.3074 +trainer/VF Loss 0.0489212 +expl/num steps total 461000 +expl/num paths total 564 +expl/path length Mean 500 +expl/path length Std 123 +expl/path length Max 623 +expl/path length Min 377 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0352834 +expl/Actions Std 0.818858 +expl/Actions Max 2.65803 +expl/Actions Min -2.22738 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 434321 +eval/num paths total 461 +eval/path length Mean 515 +eval/path length Std 0 +eval/path length Max 515 +eval/path length Min 515 +eval/Rewards Mean 0.00194175 +eval/Rewards Std 0.0440225 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0239542 +eval/Actions Std 0.733693 +eval/Actions Max 0.999277 +eval/Actions Min -0.998609 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.28897e-06 +time/evaluation sampling (s) 4.62659 +time/exploration sampling (s) 6.81948 +time/logging (s) 0.0107549 +time/saving (s) 0.0171945 +time/training (s) 18.3195 +time/epoch (s) 29.7935 +time/total (s) 10456 +Epoch -540 +------------------------------ ---------------- +2022-05-15 20:57:12.775677 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -539 finished +------------------------------ ---------------- +epoch -539 +replay_buffer/size 999047 +trainer/num train calls 462000 +trainer/QF1 Loss 0.853472 +trainer/QF2 Loss 0.802138 +trainer/Policy Loss 14.8084 +trainer/Q1 Predictions Mean -72.1274 +trainer/Q1 Predictions Std 19.0582 +trainer/Q1 Predictions Max -0.450877 +trainer/Q1 Predictions Min -86.7217 +trainer/Q2 Predictions Mean -72.1599 +trainer/Q2 Predictions Std 19.075 +trainer/Q2 Predictions Max 0.283859 +trainer/Q2 Predictions Min -86.3012 +trainer/Q Targets Mean -72.7284 +trainer/Q Targets Std 19.0195 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8768 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0146518 +trainer/policy/mean Std 0.73288 +trainer/policy/mean Max 0.999251 +trainer/policy/mean Min -0.997795 +trainer/policy/std Mean 0.42079 +trainer/policy/std Std 0.0191999 +trainer/policy/std Max 0.442306 +trainer/policy/std Min 0.38778 +trainer/Advantage Weights Mean 3.638 +trainer/Advantage Weights Std 16.0024 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.23305e-15 +trainer/Advantage Score Mean -0.351285 +trainer/Advantage Score Std 0.482002 +trainer/Advantage Score Max 0.717247 +trainer/Advantage Score Min -3.2316 +trainer/V1 Predictions Mean -72.4332 +trainer/V1 Predictions Std 19.2148 +trainer/V1 Predictions Max -0.278041 +trainer/V1 Predictions Min -86.8297 +trainer/VF Loss 0.0467321 +expl/num steps total 462000 +expl/num paths total 565 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0294415 +expl/Actions Std 0.839683 +expl/Actions Max 2.35608 +expl/Actions Min -2.28827 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 435185 +eval/num paths total 462 +eval/path length Mean 864 +eval/path length Std 0 +eval/path length Max 864 +eval/path length Min 864 +eval/Rewards Mean 0.00115741 +eval/Rewards Std 0.034001 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0572493 +eval/Actions Std 0.73783 +eval/Actions Max 0.999757 +eval/Actions Min -0.99919 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.00252e-05 +time/evaluation sampling (s) 4.74013 +time/exploration sampling (s) 5.94595 +time/logging (s) 0.0111563 +time/saving (s) 0.0163411 +time/training (s) 18.8466 +time/epoch (s) 29.5602 +time/total (s) 10485.6 +Epoch -539 +------------------------------ ---------------- +2022-05-15 20:57:43.241960 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -538 finished +------------------------------ ---------------- +epoch -538 +replay_buffer/size 999047 +trainer/num train calls 463000 +trainer/QF1 Loss 0.72655 +trainer/QF2 Loss 0.721421 +trainer/Policy Loss 22.8289 +trainer/Q1 Predictions Mean -72.5858 +trainer/Q1 Predictions Std 17.3542 +trainer/Q1 Predictions Max -0.456277 +trainer/Q1 Predictions Min -86.2563 +trainer/Q2 Predictions Mean -72.5052 +trainer/Q2 Predictions Std 17.2626 +trainer/Q2 Predictions Max -1.04515 +trainer/Q2 Predictions Min -86.116 +trainer/Q Targets Mean -72.7771 +trainer/Q Targets Std 17.2913 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.285 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00484183 +trainer/policy/mean Std 0.73577 +trainer/policy/mean Max 0.996425 +trainer/policy/mean Min -0.998238 +trainer/policy/std Mean 0.419719 +trainer/policy/std Std 0.021063 +trainer/policy/std Max 0.441925 +trainer/policy/std Min 0.383323 +trainer/Advantage Weights Mean 5.68639 +trainer/Advantage Weights Std 21.1204 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.92841e-15 +trainer/Advantage Score Mean -0.413115 +trainer/Advantage Score Std 0.546412 +trainer/Advantage Score Max 1.3409 +trainer/Advantage Score Min -3.34643 +trainer/V1 Predictions Mean -72.5661 +trainer/V1 Predictions Std 17.1916 +trainer/V1 Predictions Max -0.732347 +trainer/V1 Predictions Min -86.2048 +trainer/VF Loss 0.0716274 +expl/num steps total 463000 +expl/num paths total 567 +expl/path length Mean 500 +expl/path length Std 429 +expl/path length Max 929 +expl/path length Min 71 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0530556 +expl/Actions Std 0.814823 +expl/Actions Max 2.21298 +expl/Actions Min -2.39688 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 435782 +eval/num paths total 463 +eval/path length Mean 597 +eval/path length Std 0 +eval/path length Max 597 +eval/path length Min 597 +eval/Rewards Mean 0.00167504 +eval/Rewards Std 0.040893 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.031183 +eval/Actions Std 0.735645 +eval/Actions Max 0.999939 +eval/Actions Min -0.999587 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.28788e-05 +time/evaluation sampling (s) 4.89145 +time/exploration sampling (s) 6.32087 +time/logging (s) 0.00875887 +time/saving (s) 0.0178278 +time/training (s) 19.2112 +time/epoch (s) 30.4501 +time/total (s) 10516 +Epoch -538 +------------------------------ ---------------- +2022-05-15 20:58:14.322285 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -537 finished +------------------------------ ---------------- +epoch -537 +replay_buffer/size 999047 +trainer/num train calls 464000 +trainer/QF1 Loss 4.40822 +trainer/QF2 Loss 4.38202 +trainer/Policy Loss 24.2499 +trainer/Q1 Predictions Mean -74.4527 +trainer/Q1 Predictions Std 16.254 +trainer/Q1 Predictions Max -0.249197 +trainer/Q1 Predictions Min -86.4963 +trainer/Q2 Predictions Mean -74.4003 +trainer/Q2 Predictions Std 16.2769 +trainer/Q2 Predictions Max -0.387211 +trainer/Q2 Predictions Min -85.982 +trainer/Q Targets Mean -74.5964 +trainer/Q Targets Std 16.2772 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.1989 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0136294 +trainer/policy/mean Std 0.720271 +trainer/policy/mean Max 0.999137 +trainer/policy/mean Min -0.998614 +trainer/policy/std Mean 0.419692 +trainer/policy/std Std 0.021395 +trainer/policy/std Max 0.440972 +trainer/policy/std Min 0.381972 +trainer/Advantage Weights Mean 4.00951 +trainer/Advantage Weights Std 16.5145 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.87989e-12 +trainer/Advantage Score Mean -0.305257 +trainer/Advantage Score Std 0.507254 +trainer/Advantage Score Max 1.66483 +trainer/Advantage Score Min -2.62752 +trainer/V1 Predictions Mean -74.1993 +trainer/V1 Predictions Std 16.4836 +trainer/V1 Predictions Max 0.0516233 +trainer/V1 Predictions Min -86.0089 +trainer/VF Loss 0.0631421 +expl/num steps total 464000 +expl/num paths total 568 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0264284 +expl/Actions Std 0.816328 +expl/Actions Max 2.25109 +expl/Actions Min -2.17626 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 436782 +eval/num paths total 464 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.070418 +eval/Actions Std 0.683336 +eval/Actions Max 0.999815 +eval/Actions Min -0.999345 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.43104e-06 +time/evaluation sampling (s) 4.93428 +time/exploration sampling (s) 6.5758 +time/logging (s) 0.0112221 +time/saving (s) 0.0188192 +time/training (s) 19.5319 +time/epoch (s) 31.072 +time/total (s) 10547.1 +Epoch -537 +------------------------------ ---------------- +2022-05-15 20:58:44.738165 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -536 finished +------------------------------ ---------------- +epoch -536 +replay_buffer/size 999047 +trainer/num train calls 465000 +trainer/QF1 Loss 0.8136 +trainer/QF2 Loss 0.775852 +trainer/Policy Loss 16.8619 +trainer/Q1 Predictions Mean -75.8079 +trainer/Q1 Predictions Std 14.2548 +trainer/Q1 Predictions Max -1.1316 +trainer/Q1 Predictions Min -86.5691 +trainer/Q2 Predictions Mean -75.849 +trainer/Q2 Predictions Std 14.2613 +trainer/Q2 Predictions Max -1.57908 +trainer/Q2 Predictions Min -86.7424 +trainer/Q Targets Mean -76.2362 +trainer/Q Targets Std 14.0273 +trainer/Q Targets Max -0.953465 +trainer/Q Targets Min -87.0461 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00549601 +trainer/policy/mean Std 0.722655 +trainer/policy/mean Max 0.999752 +trainer/policy/mean Min -0.999868 +trainer/policy/std Mean 0.42037 +trainer/policy/std Std 0.0202932 +trainer/policy/std Max 0.441028 +trainer/policy/std Min 0.388971 +trainer/Advantage Weights Mean 2.9804 +trainer/Advantage Weights Std 13.1784 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.09781e-19 +trainer/Advantage Score Mean -0.271981 +trainer/Advantage Score Std 0.530329 +trainer/Advantage Score Max 2.06875 +trainer/Advantage Score Min -4.30082 +trainer/V1 Predictions Mean -75.9671 +trainer/V1 Predictions Std 14.3252 +trainer/V1 Predictions Max -1.13126 +trainer/V1 Predictions Min -86.9149 +trainer/VF Loss 0.062032 +expl/num steps total 465000 +expl/num paths total 570 +expl/path length Mean 500 +expl/path length Std 167 +expl/path length Max 667 +expl/path length Min 333 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0405445 +expl/Actions Std 0.829791 +expl/Actions Max 2.46292 +expl/Actions Min -2.33658 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 437782 +eval/num paths total 465 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0820773 +eval/Actions Std 0.756491 +eval/Actions Max 0.99978 +eval/Actions Min -0.999127 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.2584e-05 +time/evaluation sampling (s) 4.37696 +time/exploration sampling (s) 6.68695 +time/logging (s) 0.0125902 +time/saving (s) 0.0179699 +time/training (s) 19.3064 +time/epoch (s) 30.4009 +time/total (s) 10577.5 +Epoch -536 +------------------------------ ---------------- +2022-05-15 20:59:15.102310 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -535 finished +------------------------------ ---------------- +epoch -535 +replay_buffer/size 999047 +trainer/num train calls 466000 +trainer/QF1 Loss 0.503435 +trainer/QF2 Loss 0.55802 +trainer/Policy Loss 22.3058 +trainer/Q1 Predictions Mean -72.377 +trainer/Q1 Predictions Std 20.1193 +trainer/Q1 Predictions Max -0.743773 +trainer/Q1 Predictions Min -86.396 +trainer/Q2 Predictions Mean -72.2904 +trainer/Q2 Predictions Std 20.1078 +trainer/Q2 Predictions Max -0.795833 +trainer/Q2 Predictions Min -86.3116 +trainer/Q Targets Mean -72.5296 +trainer/Q Targets Std 20.0873 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5579 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0154122 +trainer/policy/mean Std 0.718912 +trainer/policy/mean Max 0.998782 +trainer/policy/mean Min -0.997627 +trainer/policy/std Mean 0.422456 +trainer/policy/std Std 0.0215624 +trainer/policy/std Max 0.447003 +trainer/policy/std Min 0.389711 +trainer/Advantage Weights Mean 5.99325 +trainer/Advantage Weights Std 21.0306 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.19075e-11 +trainer/Advantage Score Mean -0.269545 +trainer/Advantage Score Std 0.527169 +trainer/Advantage Score Max 1.98422 +trainer/Advantage Score Min -2.51539 +trainer/V1 Predictions Mean -72.3292 +trainer/V1 Predictions Std 20.1448 +trainer/V1 Predictions Max -0.139122 +trainer/V1 Predictions Min -86.4113 +trainer/VF Loss 0.0736398 +expl/num steps total 466000 +expl/num paths total 571 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00575276 +expl/Actions Std 0.830055 +expl/Actions Max 2.35764 +expl/Actions Min -2.40626 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 438782 +eval/num paths total 466 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0555462 +eval/Actions Std 0.699573 +eval/Actions Max 0.999473 +eval/Actions Min -0.998955 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.83705e-06 +time/evaluation sampling (s) 4.99351 +time/exploration sampling (s) 6.11033 +time/logging (s) 0.0124162 +time/saving (s) 0.0180161 +time/training (s) 19.2144 +time/epoch (s) 30.3487 +time/total (s) 10607.9 +Epoch -535 +------------------------------ ---------------- +2022-05-15 20:59:46.020592 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -534 finished +------------------------------ ---------------- +epoch -534 +replay_buffer/size 999047 +trainer/num train calls 467000 +trainer/QF1 Loss 0.541526 +trainer/QF2 Loss 0.564925 +trainer/Policy Loss 9.55954 +trainer/Q1 Predictions Mean -73.2187 +trainer/Q1 Predictions Std 19.162 +trainer/Q1 Predictions Max -0.268166 +trainer/Q1 Predictions Min -87.4527 +trainer/Q2 Predictions Mean -73.072 +trainer/Q2 Predictions Std 19.2798 +trainer/Q2 Predictions Max -0.56911 +trainer/Q2 Predictions Min -87.006 +trainer/Q Targets Mean -73.0489 +trainer/Q Targets Std 19.1637 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0624 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.014558 +trainer/policy/mean Std 0.724264 +trainer/policy/mean Max 0.998152 +trainer/policy/mean Min -0.998123 +trainer/policy/std Mean 0.420708 +trainer/policy/std Std 0.0198426 +trainer/policy/std Max 0.442292 +trainer/policy/std Min 0.391274 +trainer/Advantage Weights Mean 3.05564 +trainer/Advantage Weights Std 14.6611 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.67524e-18 +trainer/Advantage Score Mean -0.405137 +trainer/Advantage Score Std 0.561967 +trainer/Advantage Score Max 2.19542 +trainer/Advantage Score Min -3.99043 +trainer/V1 Predictions Mean -72.7947 +trainer/V1 Predictions Std 19.2265 +trainer/V1 Predictions Max 0.417301 +trainer/V1 Predictions Min -86.553 +trainer/VF Loss 0.0783562 +expl/num steps total 467000 +expl/num paths total 572 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.292438 +expl/Actions Std 0.844208 +expl/Actions Max 2.27178 +expl/Actions Min -2.61494 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 439353 +eval/num paths total 467 +eval/path length Mean 571 +eval/path length Std 0 +eval/path length Max 571 +eval/path length Min 571 +eval/Rewards Mean 0.00175131 +eval/Rewards Std 0.041812 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0336119 +eval/Actions Std 0.729687 +eval/Actions Max 0.999057 +eval/Actions Min -0.998796 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.78303e-06 +time/evaluation sampling (s) 4.81834 +time/exploration sampling (s) 6.44041 +time/logging (s) 0.00977206 +time/saving (s) 0.0279051 +time/training (s) 19.6042 +time/epoch (s) 30.9006 +time/total (s) 10638.8 +Epoch -534 +------------------------------ ---------------- +2022-05-15 21:00:15.971989 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -533 finished +------------------------------ ---------------- +epoch -533 +replay_buffer/size 999047 +trainer/num train calls 468000 +trainer/QF1 Loss 1.64323 +trainer/QF2 Loss 1.78835 +trainer/Policy Loss 33.4549 +trainer/Q1 Predictions Mean -72.4609 +trainer/Q1 Predictions Std 19.3361 +trainer/Q1 Predictions Max -0.78873 +trainer/Q1 Predictions Min -86.66 +trainer/Q2 Predictions Mean -72.4838 +trainer/Q2 Predictions Std 19.2964 +trainer/Q2 Predictions Max -0.947506 +trainer/Q2 Predictions Min -86.6144 +trainer/Q Targets Mean -72.8519 +trainer/Q Targets Std 18.8266 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.933 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00773571 +trainer/policy/mean Std 0.725578 +trainer/policy/mean Max 0.999512 +trainer/policy/mean Min -0.997102 +trainer/policy/std Mean 0.418694 +trainer/policy/std Std 0.0200588 +trainer/policy/std Max 0.439423 +trainer/policy/std Min 0.384138 +trainer/Advantage Weights Mean 8.53417 +trainer/Advantage Weights Std 25.5913 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.30681e-09 +trainer/Advantage Score Mean -0.204817 +trainer/Advantage Score Std 0.540299 +trainer/Advantage Score Max 2.67067 +trainer/Advantage Score Min -1.95273 +trainer/V1 Predictions Mean -72.5505 +trainer/V1 Predictions Std 18.9675 +trainer/V1 Predictions Max -3.02718 +trainer/V1 Predictions Min -86.7788 +trainer/VF Loss 0.123976 +expl/num steps total 468000 +expl/num paths total 574 +expl/path length Mean 500 +expl/path length Std 243 +expl/path length Max 743 +expl/path length Min 257 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0221089 +expl/Actions Std 0.822779 +expl/Actions Max 2.35635 +expl/Actions Min -2.20384 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 440353 +eval/num paths total 468 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.130075 +eval/Actions Std 0.73092 +eval/Actions Max 0.999764 +eval/Actions Min -0.99992 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.1947e-05 +time/evaluation sampling (s) 4.77948 +time/exploration sampling (s) 6.29041 +time/logging (s) 0.0119065 +time/saving (s) 0.0152987 +time/training (s) 18.8402 +time/epoch (s) 29.9373 +time/total (s) 10668.7 +Epoch -533 +------------------------------ ---------------- +2022-05-15 21:00:45.995403 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -532 finished +------------------------------ ---------------- +epoch -532 +replay_buffer/size 999047 +trainer/num train calls 469000 +trainer/QF1 Loss 1.11549 +trainer/QF2 Loss 1.01662 +trainer/Policy Loss 5.95778 +trainer/Q1 Predictions Mean -71.6449 +trainer/Q1 Predictions Std 19.2818 +trainer/Q1 Predictions Max -3.63195 +trainer/Q1 Predictions Min -86.5737 +trainer/Q2 Predictions Mean -71.5085 +trainer/Q2 Predictions Std 19.4305 +trainer/Q2 Predictions Max -2.52469 +trainer/Q2 Predictions Min -86.6116 +trainer/Q Targets Mean -71.1806 +trainer/Q Targets Std 19.6954 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7988 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00921578 +trainer/policy/mean Std 0.720513 +trainer/policy/mean Max 0.99884 +trainer/policy/mean Min -0.999558 +trainer/policy/std Mean 0.419544 +trainer/policy/std Std 0.0200087 +trainer/policy/std Max 0.443125 +trainer/policy/std Min 0.386588 +trainer/Advantage Weights Mean 1.24887 +trainer/Advantage Weights Std 9.26017 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.17729e-13 +trainer/Advantage Score Mean -0.561371 +trainer/Advantage Score Std 0.579429 +trainer/Advantage Score Max 0.806662 +trainer/Advantage Score Min -2.97704 +trainer/V1 Predictions Mean -70.9837 +trainer/V1 Predictions Std 19.6429 +trainer/V1 Predictions Max -2.28614 +trainer/V1 Predictions Min -86.6595 +trainer/VF Loss 0.0693412 +expl/num steps total 469000 +expl/num paths total 575 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.224195 +expl/Actions Std 0.795791 +expl/Actions Max 2.24199 +expl/Actions Min -2.64698 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 441104 +eval/num paths total 469 +eval/path length Mean 751 +eval/path length Std 0 +eval/path length Max 751 +eval/path length Min 751 +eval/Rewards Mean 0.00133156 +eval/Rewards Std 0.0364662 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0335104 +eval/Actions Std 0.740465 +eval/Actions Max 0.99902 +eval/Actions Min -0.998904 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.01028e-06 +time/evaluation sampling (s) 4.55888 +time/exploration sampling (s) 6.49936 +time/logging (s) 0.00774507 +time/saving (s) 0.0115909 +time/training (s) 18.9281 +time/epoch (s) 30.0056 +time/total (s) 10698.7 +Epoch -532 +------------------------------ ---------------- +2022-05-15 21:01:16.640801 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -531 finished +------------------------------ ---------------- +epoch -531 +replay_buffer/size 999047 +trainer/num train calls 470000 +trainer/QF1 Loss 0.592211 +trainer/QF2 Loss 0.621302 +trainer/Policy Loss 10.1627 +trainer/Q1 Predictions Mean -72.6913 +trainer/Q1 Predictions Std 19.212 +trainer/Q1 Predictions Max -0.554546 +trainer/Q1 Predictions Min -86.3765 +trainer/Q2 Predictions Mean -72.6384 +trainer/Q2 Predictions Std 19.2261 +trainer/Q2 Predictions Max -0.28913 +trainer/Q2 Predictions Min -86.4776 +trainer/Q Targets Mean -72.5225 +trainer/Q Targets Std 19.2871 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.1882 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0056068 +trainer/policy/mean Std 0.716009 +trainer/policy/mean Max 0.998647 +trainer/policy/mean Min -0.997665 +trainer/policy/std Mean 0.419657 +trainer/policy/std Std 0.0200654 +trainer/policy/std Max 0.441444 +trainer/policy/std Min 0.387894 +trainer/Advantage Weights Mean 2.45575 +trainer/Advantage Weights Std 13.6518 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.81056e-12 +trainer/Advantage Score Mean -0.450553 +trainer/Advantage Score Std 0.484029 +trainer/Advantage Score Max 0.956128 +trainer/Advantage Score Min -2.57125 +trainer/V1 Predictions Mean -72.2634 +trainer/V1 Predictions Std 19.3214 +trainer/V1 Predictions Max -0.704691 +trainer/V1 Predictions Min -86.029 +trainer/VF Loss 0.0538742 +expl/num steps total 470000 +expl/num paths total 577 +expl/path length Mean 500 +expl/path length Std 157 +expl/path length Max 657 +expl/path length Min 343 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0152383 +expl/Actions Std 0.822297 +expl/Actions Max 2.19846 +expl/Actions Min -2.40478 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 442104 +eval/num paths total 470 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.145201 +eval/Actions Std 0.676639 +eval/Actions Max 0.996076 +eval/Actions Min -0.99661 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.30511e-05 +time/evaluation sampling (s) 4.72504 +time/exploration sampling (s) 6.28525 +time/logging (s) 0.0128754 +time/saving (s) 0.0173687 +time/training (s) 19.5959 +time/epoch (s) 30.6365 +time/total (s) 10729.4 +Epoch -531 +------------------------------ ---------------- +2022-05-15 21:01:46.594473 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -530 finished +------------------------------ ---------------- +epoch -530 +replay_buffer/size 999047 +trainer/num train calls 471000 +trainer/QF1 Loss 1.12263 +trainer/QF2 Loss 1.20061 +trainer/Policy Loss 29.4411 +trainer/Q1 Predictions Mean -72.2497 +trainer/Q1 Predictions Std 18.5188 +trainer/Q1 Predictions Max -1.05838 +trainer/Q1 Predictions Min -86.7401 +trainer/Q2 Predictions Mean -72.2293 +trainer/Q2 Predictions Std 18.5042 +trainer/Q2 Predictions Max -0.749309 +trainer/Q2 Predictions Min -86.8967 +trainer/Q Targets Mean -72.2043 +trainer/Q Targets Std 19.0288 +trainer/Q Targets Max 0.322891 +trainer/Q Targets Min -87.3384 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0099755 +trainer/policy/mean Std 0.723702 +trainer/policy/mean Max 0.999456 +trainer/policy/mean Min -0.999526 +trainer/policy/std Mean 0.421862 +trainer/policy/std Std 0.019441 +trainer/policy/std Max 0.443676 +trainer/policy/std Min 0.393205 +trainer/Advantage Weights Mean 6.84066 +trainer/Advantage Weights Std 20.343 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.25754e-21 +trainer/Advantage Score Mean -0.322171 +trainer/Advantage Score Std 0.718051 +trainer/Advantage Score Max 1.67611 +trainer/Advantage Score Min -4.81251 +trainer/V1 Predictions Mean -71.9254 +trainer/V1 Predictions Std 19.1304 +trainer/V1 Predictions Max 2.32876 +trainer/V1 Predictions Min -87.1383 +trainer/VF Loss 0.091738 +expl/num steps total 471000 +expl/num paths total 579 +expl/path length Mean 500 +expl/path length Std 465 +expl/path length Max 965 +expl/path length Min 35 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0384186 +expl/Actions Std 0.830087 +expl/Actions Max 2.39145 +expl/Actions Min -2.25927 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 443104 +eval/num paths total 471 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.10986 +eval/Actions Std 0.755396 +eval/Actions Max 0.999641 +eval/Actions Min -0.999291 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.11749e-05 +time/evaluation sampling (s) 5.01249 +time/exploration sampling (s) 5.75756 +time/logging (s) 0.0128912 +time/saving (s) 0.0184858 +time/training (s) 19.1367 +time/epoch (s) 29.9382 +time/total (s) 10759.3 +Epoch -530 +------------------------------ ---------------- +2022-05-15 21:02:17.222200 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -529 finished +------------------------------ ---------------- +epoch -529 +replay_buffer/size 999047 +trainer/num train calls 472000 +trainer/QF1 Loss 0.916843 +trainer/QF2 Loss 0.93231 +trainer/Policy Loss 48.173 +trainer/Q1 Predictions Mean -73.4555 +trainer/Q1 Predictions Std 18.039 +trainer/Q1 Predictions Max -0.255206 +trainer/Q1 Predictions Min -87.1045 +trainer/Q2 Predictions Mean -73.4196 +trainer/Q2 Predictions Std 18.0451 +trainer/Q2 Predictions Max -0.467684 +trainer/Q2 Predictions Min -87.1216 +trainer/Q Targets Mean -73.6219 +trainer/Q Targets Std 18.2781 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5579 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0159839 +trainer/policy/mean Std 0.711963 +trainer/policy/mean Max 0.998459 +trainer/policy/mean Min -0.998962 +trainer/policy/std Mean 0.421287 +trainer/policy/std Std 0.0196462 +trainer/policy/std Max 0.441091 +trainer/policy/std Min 0.389174 +trainer/Advantage Weights Mean 9.44728 +trainer/Advantage Weights Std 22.7671 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.0138e-17 +trainer/Advantage Score Mean -0.127887 +trainer/Advantage Score Std 0.561693 +trainer/Advantage Score Max 1.18876 +trainer/Advantage Score Min -3.91302 +trainer/V1 Predictions Mean -73.4395 +trainer/V1 Predictions Std 18.1788 +trainer/V1 Predictions Max 0.961058 +trainer/V1 Predictions Min -87.4349 +trainer/VF Loss 0.070248 +expl/num steps total 472000 +expl/num paths total 581 +expl/path length Mean 500 +expl/path length Std 430 +expl/path length Max 930 +expl/path length Min 70 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0267608 +expl/Actions Std 0.840078 +expl/Actions Max 2.22947 +expl/Actions Min -2.3894 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 443973 +eval/num paths total 472 +eval/path length Mean 869 +eval/path length Std 0 +eval/path length Max 869 +eval/path length Min 869 +eval/Rewards Mean 0.00115075 +eval/Rewards Std 0.0339032 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0217448 +eval/Actions Std 0.753729 +eval/Actions Max 0.999906 +eval/Actions Min -0.999619 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.06064e-05 +time/evaluation sampling (s) 4.78585 +time/exploration sampling (s) 6.23398 +time/logging (s) 0.0099661 +time/saving (s) 0.0126265 +time/training (s) 19.566 +time/epoch (s) 30.6084 +time/total (s) 10789.9 +Epoch -529 +------------------------------ ---------------- +2022-05-15 21:02:47.194519 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -528 finished +------------------------------ ---------------- +epoch -528 +replay_buffer/size 999047 +trainer/num train calls 473000 +trainer/QF1 Loss 1.1699 +trainer/QF2 Loss 1.34782 +trainer/Policy Loss 6.54017 +trainer/Q1 Predictions Mean -72.6225 +trainer/Q1 Predictions Std 19.6272 +trainer/Q1 Predictions Max -1.48108 +trainer/Q1 Predictions Min -87.6245 +trainer/Q2 Predictions Mean -72.642 +trainer/Q2 Predictions Std 19.6136 +trainer/Q2 Predictions Max -1.71918 +trainer/Q2 Predictions Min -87.1762 +trainer/Q Targets Mean -72.4116 +trainer/Q Targets Std 19.7569 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8518 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.016865 +trainer/policy/mean Std 0.711432 +trainer/policy/mean Max 0.999675 +trainer/policy/mean Min -0.999415 +trainer/policy/std Mean 0.421126 +trainer/policy/std Std 0.0201786 +trainer/policy/std Max 0.442201 +trainer/policy/std Min 0.38717 +trainer/Advantage Weights Mean 2.20056 +trainer/Advantage Weights Std 8.5161 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.6066e-14 +trainer/Advantage Score Mean -0.36552 +trainer/Advantage Score Std 0.569902 +trainer/Advantage Score Max 1.05523 +trainer/Advantage Score Min -3.09534 +trainer/V1 Predictions Mean -72.1726 +trainer/V1 Predictions Std 19.6659 +trainer/V1 Predictions Max -2.12841 +trainer/V1 Predictions Min -86.7158 +trainer/VF Loss 0.055072 +expl/num steps total 473000 +expl/num paths total 582 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0302872 +expl/Actions Std 0.81656 +expl/Actions Max 2.3901 +expl/Actions Min -2.33518 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 444973 +eval/num paths total 473 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.178501 +eval/Actions Std 0.769019 +eval/Actions Max 0.999761 +eval/Actions Min -0.999173 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.22199e-05 +time/evaluation sampling (s) 4.81904 +time/exploration sampling (s) 5.74957 +time/logging (s) 0.0083708 +time/saving (s) 0.0118936 +time/training (s) 19.3645 +time/epoch (s) 29.9534 +time/total (s) 10819.9 +Epoch -528 +------------------------------ ---------------- +2022-05-15 21:03:17.990168 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -527 finished +------------------------------ ---------------- +epoch -527 +replay_buffer/size 999047 +trainer/num train calls 474000 +trainer/QF1 Loss 5.81688 +trainer/QF2 Loss 5.62635 +trainer/Policy Loss 25.1153 +trainer/Q1 Predictions Mean -72.8837 +trainer/Q1 Predictions Std 17.6061 +trainer/Q1 Predictions Max -1.07662 +trainer/Q1 Predictions Min -87.7572 +trainer/Q2 Predictions Mean -72.8377 +trainer/Q2 Predictions Std 17.5734 +trainer/Q2 Predictions Max -0.964019 +trainer/Q2 Predictions Min -87.31 +trainer/Q Targets Mean -73.2626 +trainer/Q Targets Std 16.9841 +trainer/Q Targets Max -2.70594 +trainer/Q Targets Min -89.328 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00916481 +trainer/policy/mean Std 0.717874 +trainer/policy/mean Max 0.998781 +trainer/policy/mean Min -0.997514 +trainer/policy/std Mean 0.41884 +trainer/policy/std Std 0.0211704 +trainer/policy/std Max 0.440966 +trainer/policy/std Min 0.383821 +trainer/Advantage Weights Mean 6.24487 +trainer/Advantage Weights Std 20.3305 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.6485e-17 +trainer/Advantage Score Mean -0.385584 +trainer/Advantage Score Std 0.61361 +trainer/Advantage Score Max 1.54874 +trainer/Advantage Score Min -3.76074 +trainer/V1 Predictions Mean -72.8055 +trainer/V1 Predictions Std 17.4093 +trainer/V1 Predictions Max -0.972583 +trainer/V1 Predictions Min -88.2548 +trainer/VF Loss 0.0867817 +expl/num steps total 474000 +expl/num paths total 584 +expl/path length Mean 500 +expl/path length Std 427 +expl/path length Max 927 +expl/path length Min 73 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0502198 +expl/Actions Std 0.820857 +expl/Actions Max 2.22323 +expl/Actions Min -2.14216 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 445973 +eval/num paths total 474 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.010375 +eval/Actions Std 0.780024 +eval/Actions Max 0.999883 +eval/Actions Min -0.999465 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.28178e-05 +time/evaluation sampling (s) 4.78557 +time/exploration sampling (s) 6.36002 +time/logging (s) 0.00859117 +time/saving (s) 0.0129882 +time/training (s) 19.6185 +time/epoch (s) 30.7857 +time/total (s) 10850.7 +Epoch -527 +------------------------------ ---------------- +2022-05-15 21:03:49.044419 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -526 finished +------------------------------ ---------------- +epoch -526 +replay_buffer/size 999047 +trainer/num train calls 475000 +trainer/QF1 Loss 0.987344 +trainer/QF2 Loss 0.95594 +trainer/Policy Loss 27.3639 +trainer/Q1 Predictions Mean -70.8448 +trainer/Q1 Predictions Std 19.7565 +trainer/Q1 Predictions Max -0.366166 +trainer/Q1 Predictions Min -86.7853 +trainer/Q2 Predictions Mean -70.9023 +trainer/Q2 Predictions Std 19.6516 +trainer/Q2 Predictions Max -0.397243 +trainer/Q2 Predictions Min -86.6622 +trainer/Q Targets Mean -71.0826 +trainer/Q Targets Std 19.3906 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.077 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0105222 +trainer/policy/mean Std 0.714228 +trainer/policy/mean Max 0.999958 +trainer/policy/mean Min -0.999776 +trainer/policy/std Mean 0.420167 +trainer/policy/std Std 0.0218181 +trainer/policy/std Max 0.444589 +trainer/policy/std Min 0.385698 +trainer/Advantage Weights Mean 6.28039 +trainer/Advantage Weights Std 21.2264 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.54628e-19 +trainer/Advantage Score Mean -0.331468 +trainer/Advantage Score Std 0.608192 +trainer/Advantage Score Max 1.74687 +trainer/Advantage Score Min -4.28145 +trainer/V1 Predictions Mean -70.7493 +trainer/V1 Predictions Std 19.7073 +trainer/V1 Predictions Max -0.297053 +trainer/V1 Predictions Min -86.1372 +trainer/VF Loss 0.0842514 +expl/num steps total 475000 +expl/num paths total 586 +expl/path length Mean 500 +expl/path length Std 106 +expl/path length Max 606 +expl/path length Min 394 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0255902 +expl/Actions Std 0.830424 +expl/Actions Max 2.36349 +expl/Actions Min -2.33029 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 446973 +eval/num paths total 475 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.145228 +eval/Actions Std 0.647118 +eval/Actions Max 0.999765 +eval/Actions Min -0.999898 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.26031e-05 +time/evaluation sampling (s) 5.12273 +time/exploration sampling (s) 6.69857 +time/logging (s) 0.0124169 +time/saving (s) 0.0152208 +time/training (s) 19.1985 +time/epoch (s) 31.0475 +time/total (s) 10881.7 +Epoch -526 +------------------------------ ---------------- +2022-05-15 21:04:19.303265 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -525 finished +------------------------------ ---------------- +epoch -525 +replay_buffer/size 999047 +trainer/num train calls 476000 +trainer/QF1 Loss 0.512987 +trainer/QF2 Loss 0.588221 +trainer/Policy Loss 0.967314 +trainer/Q1 Predictions Mean -74.1802 +trainer/Q1 Predictions Std 17.1158 +trainer/Q1 Predictions Max -0.418593 +trainer/Q1 Predictions Min -87.3698 +trainer/Q2 Predictions Mean -74.2707 +trainer/Q2 Predictions Std 17.1281 +trainer/Q2 Predictions Max -0.633045 +trainer/Q2 Predictions Min -87.4937 +trainer/Q Targets Mean -73.8427 +trainer/Q Targets Std 17.1785 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7363 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0191651 +trainer/policy/mean Std 0.725336 +trainer/policy/mean Max 0.999135 +trainer/policy/mean Min -0.999457 +trainer/policy/std Mean 0.421653 +trainer/policy/std Std 0.0200106 +trainer/policy/std Max 0.442546 +trainer/policy/std Min 0.389447 +trainer/Advantage Weights Mean 0.147929 +trainer/Advantage Weights Std 0.842735 +trainer/Advantage Weights Max 10.8821 +trainer/Advantage Weights Min 3.69865e-21 +trainer/Advantage Score Mean -0.696385 +trainer/Advantage Score Std 0.568857 +trainer/Advantage Score Max 0.238712 +trainer/Advantage Score Min -4.70463 +trainer/V1 Predictions Mean -73.6373 +trainer/V1 Predictions Std 17.2205 +trainer/V1 Predictions Max 0.753213 +trainer/V1 Predictions Min -86.5682 +trainer/VF Loss 0.081234 +expl/num steps total 476000 +expl/num paths total 587 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0516541 +expl/Actions Std 0.867801 +expl/Actions Max 2.40885 +expl/Actions Min -2.70029 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 447973 +eval/num paths total 476 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0290452 +eval/Actions Std 0.733872 +eval/Actions Max 0.999391 +eval/Actions Min -0.998969 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.19461e-05 +time/evaluation sampling (s) 4.69553 +time/exploration sampling (s) 6.17186 +time/logging (s) 0.0108183 +time/saving (s) 0.0182415 +time/training (s) 19.3461 +time/epoch (s) 30.2425 +time/total (s) 10912 +Epoch -525 +------------------------------ ---------------- +2022-05-15 21:04:49.808594 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -524 finished +------------------------------ ---------------- +epoch -524 +replay_buffer/size 999047 +trainer/num train calls 477000 +trainer/QF1 Loss 1.21096 +trainer/QF2 Loss 1.05824 +trainer/Policy Loss 67.2916 +trainer/Q1 Predictions Mean -73.4086 +trainer/Q1 Predictions Std 17.7588 +trainer/Q1 Predictions Max -1.14196 +trainer/Q1 Predictions Min -86.8036 +trainer/Q2 Predictions Mean -73.4631 +trainer/Q2 Predictions Std 17.6837 +trainer/Q2 Predictions Max -1.45133 +trainer/Q2 Predictions Min -86.8118 +trainer/Q Targets Mean -73.9372 +trainer/Q Targets Std 17.6205 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2368 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.028736 +trainer/policy/mean Std 0.722231 +trainer/policy/mean Max 0.999381 +trainer/policy/mean Min -0.998005 +trainer/policy/std Mean 0.422357 +trainer/policy/std Std 0.0200529 +trainer/policy/std Max 0.441837 +trainer/policy/std Min 0.388902 +trainer/Advantage Weights Mean 13.4256 +trainer/Advantage Weights Std 29.6498 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.06357e-14 +trainer/Advantage Score Mean -0.0764667 +trainer/Advantage Score Std 0.509106 +trainer/Advantage Score Max 2.18432 +trainer/Advantage Score Min -3.11166 +trainer/V1 Predictions Mean -73.7541 +trainer/V1 Predictions Std 17.661 +trainer/V1 Predictions Max -1.8723 +trainer/V1 Predictions Min -87.1701 +trainer/VF Loss 0.0949919 +expl/num steps total 477000 +expl/num paths total 588 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0375637 +expl/Actions Std 0.838995 +expl/Actions Max 2.32032 +expl/Actions Min -2.33274 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 448573 +eval/num paths total 477 +eval/path length Mean 600 +eval/path length Std 0 +eval/path length Max 600 +eval/path length Min 600 +eval/Rewards Mean 0.00166667 +eval/Rewards Std 0.0407908 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0270201 +eval/Actions Std 0.765688 +eval/Actions Max 0.999218 +eval/Actions Min -0.999398 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.07302e-05 +time/evaluation sampling (s) 4.5947 +time/exploration sampling (s) 6.17618 +time/logging (s) 0.009839 +time/saving (s) 0.01795 +time/training (s) 19.6897 +time/epoch (s) 30.4884 +time/total (s) 10942.5 +Epoch -524 +------------------------------ ---------------- +2022-05-15 21:05:20.561625 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -523 finished +------------------------------ ---------------- +epoch -523 +replay_buffer/size 999047 +trainer/num train calls 478000 +trainer/QF1 Loss 0.880766 +trainer/QF2 Loss 0.905573 +trainer/Policy Loss 29.8158 +trainer/Q1 Predictions Mean -71.9286 +trainer/Q1 Predictions Std 19.1842 +trainer/Q1 Predictions Max -0.457267 +trainer/Q1 Predictions Min -86.5789 +trainer/Q2 Predictions Mean -71.8653 +trainer/Q2 Predictions Std 19.2794 +trainer/Q2 Predictions Max -0.113759 +trainer/Q2 Predictions Min -86.3469 +trainer/Q Targets Mean -72.6549 +trainer/Q Targets Std 19.3596 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2531 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.03582 +trainer/policy/mean Std 0.71299 +trainer/policy/mean Max 0.999189 +trainer/policy/mean Min -0.998009 +trainer/policy/std Mean 0.422136 +trainer/policy/std Std 0.0207465 +trainer/policy/std Max 0.443201 +trainer/policy/std Min 0.385431 +trainer/Advantage Weights Mean 6.97578 +trainer/Advantage Weights Std 17.4062 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.81565e-10 +trainer/Advantage Score Mean -0.136383 +trainer/Advantage Score Std 0.50047 +trainer/Advantage Score Max 0.925537 +trainer/Advantage Score Min -2.24294 +trainer/V1 Predictions Mean -72.4281 +trainer/V1 Predictions Std 19.4265 +trainer/V1 Predictions Max -0.0339149 +trainer/V1 Predictions Min -86.9063 +trainer/VF Loss 0.0511347 +expl/num steps total 478000 +expl/num paths total 590 +expl/path length Mean 500 +expl/path length Std 333 +expl/path length Max 833 +expl/path length Min 167 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0150954 +expl/Actions Std 0.837362 +expl/Actions Max 2.4636 +expl/Actions Min -2.11358 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 449356 +eval/num paths total 478 +eval/path length Mean 783 +eval/path length Std 0 +eval/path length Max 783 +eval/path length Min 783 +eval/Rewards Mean 0.00127714 +eval/Rewards Std 0.0357143 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0193063 +eval/Actions Std 0.752335 +eval/Actions Max 0.999318 +eval/Actions Min -0.999293 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 9.89018e-06 +time/evaluation sampling (s) 4.69006 +time/exploration sampling (s) 6.83642 +time/logging (s) 0.0114568 +time/saving (s) 0.0177028 +time/training (s) 19.184 +time/epoch (s) 30.7397 +time/total (s) 10973.2 +Epoch -523 +------------------------------ ---------------- +2022-05-15 21:05:51.448969 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -522 finished +------------------------------ ---------------- +epoch -522 +replay_buffer/size 999047 +trainer/num train calls 479000 +trainer/QF1 Loss 0.972047 +trainer/QF2 Loss 0.840856 +trainer/Policy Loss 27.7375 +trainer/Q1 Predictions Mean -71.3849 +trainer/Q1 Predictions Std 20.0383 +trainer/Q1 Predictions Max -1.66722 +trainer/Q1 Predictions Min -86.5258 +trainer/Q2 Predictions Mean -71.5014 +trainer/Q2 Predictions Std 19.8581 +trainer/Q2 Predictions Max -2.20965 +trainer/Q2 Predictions Min -86.6068 +trainer/Q Targets Mean -71.9963 +trainer/Q Targets Std 20.0482 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7998 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000278278 +trainer/policy/mean Std 0.720844 +trainer/policy/mean Max 0.999376 +trainer/policy/mean Min -0.99864 +trainer/policy/std Mean 0.421051 +trainer/policy/std Std 0.0209423 +trainer/policy/std Max 0.441788 +trainer/policy/std Min 0.383639 +trainer/Advantage Weights Mean 6.25822 +trainer/Advantage Weights Std 20.242 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.76109e-14 +trainer/Advantage Score Mean -0.221114 +trainer/Advantage Score Std 0.539922 +trainer/Advantage Score Max 1.70201 +trainer/Advantage Score Min -3.09115 +trainer/V1 Predictions Mean -71.7086 +trainer/V1 Predictions Std 20.2346 +trainer/V1 Predictions Max -2.24607 +trainer/V1 Predictions Min -86.6494 +trainer/VF Loss 0.0637 +expl/num steps total 479000 +expl/num paths total 592 +expl/path length Mean 500 +expl/path length Std 119 +expl/path length Max 619 +expl/path length Min 381 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.00093801 +expl/Actions Std 0.821901 +expl/Actions Max 2.24113 +expl/Actions Min -2.33395 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 450356 +eval/num paths total 479 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.125131 +eval/Actions Std 0.626993 +eval/Actions Max 0.99997 +eval/Actions Min -0.998966 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.91599e-06 +time/evaluation sampling (s) 5.50833 +time/exploration sampling (s) 5.96417 +time/logging (s) 0.014411 +time/saving (s) 0.0197791 +time/training (s) 19.3682 +time/epoch (s) 30.8749 +time/total (s) 11004.1 +Epoch -522 +------------------------------ ---------------- +2022-05-15 21:06:22.380792 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -521 finished +------------------------------ ---------------- +epoch -521 +replay_buffer/size 999047 +trainer/num train calls 480000 +trainer/QF1 Loss 0.699222 +trainer/QF2 Loss 0.623163 +trainer/Policy Loss 13.8501 +trainer/Q1 Predictions Mean -73.0704 +trainer/Q1 Predictions Std 17.6203 +trainer/Q1 Predictions Max -0.102471 +trainer/Q1 Predictions Min -87.462 +trainer/Q2 Predictions Mean -73.1089 +trainer/Q2 Predictions Std 17.5985 +trainer/Q2 Predictions Max -0.752877 +trainer/Q2 Predictions Min -87.3452 +trainer/Q Targets Mean -73.2133 +trainer/Q Targets Std 17.5917 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3862 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00286855 +trainer/policy/mean Std 0.717837 +trainer/policy/mean Max 0.999172 +trainer/policy/mean Min -0.998972 +trainer/policy/std Mean 0.420602 +trainer/policy/std Std 0.0210539 +trainer/policy/std Max 0.442903 +trainer/policy/std Min 0.386651 +trainer/Advantage Weights Mean 3.93696 +trainer/Advantage Weights Std 15.5052 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.67874e-16 +trainer/Advantage Score Mean -0.348234 +trainer/Advantage Score Std 0.570454 +trainer/Advantage Score Max 1.48089 +trainer/Advantage Score Min -3.63233 +trainer/V1 Predictions Mean -72.917 +trainer/V1 Predictions Std 17.7527 +trainer/V1 Predictions Max 0.73264 +trainer/V1 Predictions Min -87.2608 +trainer/VF Loss 0.0623336 +expl/num steps total 480000 +expl/num paths total 593 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0107169 +expl/Actions Std 0.799413 +expl/Actions Max 2.24496 +expl/Actions Min -2.42361 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 451356 +eval/num paths total 480 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00527553 +eval/Actions Std 0.739321 +eval/Actions Max 0.999582 +eval/Actions Min -0.997784 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.98913e-06 +time/evaluation sampling (s) 4.8807 +time/exploration sampling (s) 6.52565 +time/logging (s) 0.0124692 +time/saving (s) 0.0187269 +time/training (s) 19.4757 +time/epoch (s) 30.9133 +time/total (s) 11035 +Epoch -521 +------------------------------ ---------------- +2022-05-15 21:06:53.229960 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -520 finished +------------------------------ ---------------- +epoch -520 +replay_buffer/size 999047 +trainer/num train calls 481000 +trainer/QF1 Loss 0.933001 +trainer/QF2 Loss 0.694533 +trainer/Policy Loss 3.06292 +trainer/Q1 Predictions Mean -72.7574 +trainer/Q1 Predictions Std 18.9976 +trainer/Q1 Predictions Max -1.13129 +trainer/Q1 Predictions Min -87.1076 +trainer/Q2 Predictions Mean -72.6672 +trainer/Q2 Predictions Std 18.9887 +trainer/Q2 Predictions Max -0.546441 +trainer/Q2 Predictions Min -87.135 +trainer/Q Targets Mean -72.2246 +trainer/Q Targets Std 18.8837 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.871 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0107195 +trainer/policy/mean Std 0.725393 +trainer/policy/mean Max 0.999351 +trainer/policy/mean Min -0.999266 +trainer/policy/std Mean 0.421428 +trainer/policy/std Std 0.0214836 +trainer/policy/std Max 0.445786 +trainer/policy/std Min 0.386344 +trainer/Advantage Weights Mean 1.20314 +trainer/Advantage Weights Std 8.98036 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.07317e-12 +trainer/Advantage Score Mean -0.508046 +trainer/Advantage Score Std 0.450937 +trainer/Advantage Score Max 1.71847 +trainer/Advantage Score Min -2.75604 +trainer/V1 Predictions Mean -72.0018 +trainer/V1 Predictions Std 19.0456 +trainer/V1 Predictions Max -0.42534 +trainer/V1 Predictions Min -86.6216 +trainer/VF Loss 0.0574669 +expl/num steps total 481000 +expl/num paths total 594 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.133735 +expl/Actions Std 0.832278 +expl/Actions Max 2.42692 +expl/Actions Min -2.21767 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 452356 +eval/num paths total 481 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0423055 +eval/Actions Std 0.752835 +eval/Actions Max 0.999635 +eval/Actions Min -0.999698 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.33007e-05 +time/evaluation sampling (s) 4.93988 +time/exploration sampling (s) 6.64198 +time/logging (s) 0.0124545 +time/saving (s) 0.0173596 +time/training (s) 19.2215 +time/epoch (s) 30.8332 +time/total (s) 11065.9 +Epoch -520 +------------------------------ ---------------- +2022-05-15 21:07:23.459653 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -519 finished +------------------------------ ---------------- +epoch -519 +replay_buffer/size 999047 +trainer/num train calls 482000 +trainer/QF1 Loss 0.838357 +trainer/QF2 Loss 0.957239 +trainer/Policy Loss 14.1286 +trainer/Q1 Predictions Mean -72.1084 +trainer/Q1 Predictions Std 19.4486 +trainer/Q1 Predictions Max -4.34394 +trainer/Q1 Predictions Min -86.1522 +trainer/Q2 Predictions Mean -72.1022 +trainer/Q2 Predictions Std 19.4104 +trainer/Q2 Predictions Max -4.06288 +trainer/Q2 Predictions Min -86.1329 +trainer/Q Targets Mean -72.2048 +trainer/Q Targets Std 19.6173 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4109 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00983586 +trainer/policy/mean Std 0.715544 +trainer/policy/mean Max 0.998861 +trainer/policy/mean Min -0.996623 +trainer/policy/std Mean 0.420574 +trainer/policy/std Std 0.0205573 +trainer/policy/std Max 0.443679 +trainer/policy/std Min 0.385895 +trainer/Advantage Weights Mean 4.0229 +trainer/Advantage Weights Std 14.9336 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.3197e-13 +trainer/Advantage Score Mean -0.299342 +trainer/Advantage Score Std 0.529679 +trainer/Advantage Score Max 1.20717 +trainer/Advantage Score Min -2.7815 +trainer/V1 Predictions Mean -71.97 +trainer/V1 Predictions Std 19.646 +trainer/V1 Predictions Max -3.4386 +trainer/V1 Predictions Min -86.2079 +trainer/VF Loss 0.0513042 +expl/num steps total 482000 +expl/num paths total 595 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0630222 +expl/Actions Std 0.836477 +expl/Actions Max 2.50518 +expl/Actions Min -2.26135 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 453356 +eval/num paths total 482 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.208447 +eval/Actions Std 0.707506 +eval/Actions Max 0.999313 +eval/Actions Min -0.999209 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 6.19888e-06 +time/evaluation sampling (s) 5.22387 +time/exploration sampling (s) 5.72637 +time/logging (s) 0.00986732 +time/saving (s) 0.0142003 +time/training (s) 19.2376 +time/epoch (s) 30.212 +time/total (s) 11096.1 +Epoch -519 +------------------------------ ---------------- +2022-05-15 21:07:54.168702 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -518 finished +------------------------------ ---------------- +epoch -518 +replay_buffer/size 999047 +trainer/num train calls 483000 +trainer/QF1 Loss 1.20485 +trainer/QF2 Loss 1.01982 +trainer/Policy Loss 18.4527 +trainer/Q1 Predictions Mean -72.3746 +trainer/Q1 Predictions Std 18.5585 +trainer/Q1 Predictions Max -0.523213 +trainer/Q1 Predictions Min -86.8795 +trainer/Q2 Predictions Mean -72.4047 +trainer/Q2 Predictions Std 18.5629 +trainer/Q2 Predictions Max 0.466074 +trainer/Q2 Predictions Min -86.8994 +trainer/Q Targets Mean -72.3317 +trainer/Q Targets Std 18.7317 +trainer/Q Targets Max -0.603489 +trainer/Q Targets Min -86.6073 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00136107 +trainer/policy/mean Std 0.729638 +trainer/policy/mean Max 0.999389 +trainer/policy/mean Min -0.998805 +trainer/policy/std Mean 0.419541 +trainer/policy/std Std 0.0212688 +trainer/policy/std Max 0.4432 +trainer/policy/std Min 0.385785 +trainer/Advantage Weights Mean 5.35285 +trainer/Advantage Weights Std 19.3722 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.06622e-19 +trainer/Advantage Score Mean -0.382578 +trainer/Advantage Score Std 0.638296 +trainer/Advantage Score Max 1.0727 +trainer/Advantage Score Min -4.21265 +trainer/V1 Predictions Mean -72.1299 +trainer/V1 Predictions Std 18.8571 +trainer/V1 Predictions Max 0.393045 +trainer/V1 Predictions Min -86.3873 +trainer/VF Loss 0.0754602 +expl/num steps total 483000 +expl/num paths total 596 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0443539 +expl/Actions Std 0.832085 +expl/Actions Max 2.43948 +expl/Actions Min -2.51558 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 454356 +eval/num paths total 483 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0859189 +eval/Actions Std 0.659304 +eval/Actions Max 0.999369 +eval/Actions Min -0.999143 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.88989e-06 +time/evaluation sampling (s) 4.93752 +time/exploration sampling (s) 6.43915 +time/logging (s) 0.011538 +time/saving (s) 0.016054 +time/training (s) 19.2931 +time/epoch (s) 30.6973 +time/total (s) 11126.8 +Epoch -518 +------------------------------ ---------------- +2022-05-15 21:08:25.010902 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -517 finished +------------------------------ ---------------- +epoch -517 +replay_buffer/size 999047 +trainer/num train calls 484000 +trainer/QF1 Loss 1.04034 +trainer/QF2 Loss 0.84725 +trainer/Policy Loss 37.6091 +trainer/Q1 Predictions Mean -72.7149 +trainer/Q1 Predictions Std 18.369 +trainer/Q1 Predictions Max -0.813584 +trainer/Q1 Predictions Min -85.8697 +trainer/Q2 Predictions Mean -72.781 +trainer/Q2 Predictions Std 18.2972 +trainer/Q2 Predictions Max -1.02441 +trainer/Q2 Predictions Min -86.2057 +trainer/Q Targets Mean -73.2113 +trainer/Q Targets Std 17.9593 +trainer/Q Targets Max -1.74711 +trainer/Q Targets Min -86.3927 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00484737 +trainer/policy/mean Std 0.743178 +trainer/policy/mean Max 0.999741 +trainer/policy/mean Min -0.998245 +trainer/policy/std Mean 0.419731 +trainer/policy/std Std 0.0213228 +trainer/policy/std Max 0.442389 +trainer/policy/std Min 0.385078 +trainer/Advantage Weights Mean 9.63004 +trainer/Advantage Weights Std 25.0672 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.66552e-17 +trainer/Advantage Score Mean -0.195484 +trainer/Advantage Score Std 0.623492 +trainer/Advantage Score Max 1.52005 +trainer/Advantage Score Min -3.81635 +trainer/V1 Predictions Mean -72.865 +trainer/V1 Predictions Std 18.3471 +trainer/V1 Predictions Max -1.2961 +trainer/V1 Predictions Min -86.2395 +trainer/VF Loss 0.0933918 +expl/num steps total 484000 +expl/num paths total 598 +expl/path length Mean 500 +expl/path length Std 281 +expl/path length Max 781 +expl/path length Min 219 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0242191 +expl/Actions Std 0.832925 +expl/Actions Max 2.12194 +expl/Actions Min -2.3098 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 454888 +eval/num paths total 484 +eval/path length Mean 532 +eval/path length Std 0 +eval/path length Max 532 +eval/path length Min 532 +eval/Rewards Mean 0.0018797 +eval/Rewards Std 0.0433147 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.016818 +eval/Actions Std 0.731993 +eval/Actions Max 0.9999 +eval/Actions Min -0.999952 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.33971e-05 +time/evaluation sampling (s) 4.62545 +time/exploration sampling (s) 6.85461 +time/logging (s) 0.00986441 +time/saving (s) 0.0155092 +time/training (s) 19.3118 +time/epoch (s) 30.8172 +time/total (s) 11157.6 +Epoch -517 +------------------------------ ---------------- +2022-05-15 21:08:55.179714 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -516 finished +------------------------------ ---------------- +epoch -516 +replay_buffer/size 999047 +trainer/num train calls 485000 +trainer/QF1 Loss 1.05554 +trainer/QF2 Loss 0.958278 +trainer/Policy Loss 28.2125 +trainer/Q1 Predictions Mean -70.5378 +trainer/Q1 Predictions Std 21.1528 +trainer/Q1 Predictions Max -0.24636 +trainer/Q1 Predictions Min -86.9588 +trainer/Q2 Predictions Mean -70.5774 +trainer/Q2 Predictions Std 21.1136 +trainer/Q2 Predictions Max 0.155371 +trainer/Q2 Predictions Min -87.1503 +trainer/Q Targets Mean -70.8027 +trainer/Q Targets Std 21.096 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0775 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0050643 +trainer/policy/mean Std 0.721718 +trainer/policy/mean Max 0.999438 +trainer/policy/mean Min -0.997386 +trainer/policy/std Mean 0.420071 +trainer/policy/std Std 0.0213144 +trainer/policy/std Max 0.441722 +trainer/policy/std Min 0.384302 +trainer/Advantage Weights Mean 8.50478 +trainer/Advantage Weights Std 24.002 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.03519e-12 +trainer/Advantage Score Mean -0.249341 +trainer/Advantage Score Std 0.602267 +trainer/Advantage Score Max 1.38524 +trainer/Advantage Score Min -2.69204 +trainer/V1 Predictions Mean -70.5894 +trainer/V1 Predictions Std 21.0101 +trainer/V1 Predictions Max 0.497515 +trainer/V1 Predictions Min -86.9771 +trainer/VF Loss 0.0846248 +expl/num steps total 485000 +expl/num paths total 600 +expl/path length Mean 500 +expl/path length Std 331 +expl/path length Max 831 +expl/path length Min 169 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0475391 +expl/Actions Std 0.827009 +expl/Actions Max 2.43059 +expl/Actions Min -2.15312 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 455888 +eval/num paths total 485 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0212852 +eval/Actions Std 0.733348 +eval/Actions Max 0.999754 +eval/Actions Min -0.99975 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.28248e-05 +time/evaluation sampling (s) 5.17473 +time/exploration sampling (s) 6.40253 +time/logging (s) 0.00779754 +time/saving (s) 0.0166904 +time/training (s) 18.5503 +time/epoch (s) 30.1521 +time/total (s) 11187.8 +Epoch -516 +------------------------------ ---------------- +2022-05-15 21:09:25.818786 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -515 finished +------------------------------ ---------------- +epoch -515 +replay_buffer/size 999047 +trainer/num train calls 486000 +trainer/QF1 Loss 0.608254 +trainer/QF2 Loss 0.609766 +trainer/Policy Loss 24.1731 +trainer/Q1 Predictions Mean -73.7966 +trainer/Q1 Predictions Std 15.422 +trainer/Q1 Predictions Max -4.47005 +trainer/Q1 Predictions Min -86.7487 +trainer/Q2 Predictions Mean -73.8046 +trainer/Q2 Predictions Std 15.378 +trainer/Q2 Predictions Max -5.28215 +trainer/Q2 Predictions Min -87.2282 +trainer/Q Targets Mean -74.152 +trainer/Q Targets Std 15.3648 +trainer/Q Targets Max -4.57751 +trainer/Q Targets Min -86.5818 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00727292 +trainer/policy/mean Std 0.710285 +trainer/policy/mean Max 0.998491 +trainer/policy/mean Min -0.998908 +trainer/policy/std Mean 0.420414 +trainer/policy/std Std 0.0215103 +trainer/policy/std Max 0.442541 +trainer/policy/std Min 0.385556 +trainer/Advantage Weights Mean 4.58553 +trainer/Advantage Weights Std 16.0795 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.19362e-11 +trainer/Advantage Score Mean -0.319087 +trainer/Advantage Score Std 0.509048 +trainer/Advantage Score Max 0.976125 +trainer/Advantage Score Min -2.51514 +trainer/V1 Predictions Mean -73.8689 +trainer/V1 Predictions Std 15.4137 +trainer/V1 Predictions Max -6.16086 +trainer/V1 Predictions Min -86.022 +trainer/VF Loss 0.0530167 +expl/num steps total 486000 +expl/num paths total 601 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0292492 +expl/Actions Std 0.842596 +expl/Actions Max 2.38531 +expl/Actions Min -2.63586 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 456888 +eval/num paths total 486 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0951 +eval/Actions Std 0.692621 +eval/Actions Max 0.999784 +eval/Actions Min -0.99874 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.94372e-06 +time/evaluation sampling (s) 4.98619 +time/exploration sampling (s) 7.22618 +time/logging (s) 0.00996989 +time/saving (s) 0.012583 +time/training (s) 18.3925 +time/epoch (s) 30.6274 +time/total (s) 11218.4 +Epoch -515 +------------------------------ ---------------- +2022-05-15 21:09:56.039024 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -514 finished +------------------------------ ---------------- +epoch -514 +replay_buffer/size 999047 +trainer/num train calls 487000 +trainer/QF1 Loss 0.984123 +trainer/QF2 Loss 0.806206 +trainer/Policy Loss 11.7776 +trainer/Q1 Predictions Mean -70.471 +trainer/Q1 Predictions Std 21.051 +trainer/Q1 Predictions Max -0.410052 +trainer/Q1 Predictions Min -86.5283 +trainer/Q2 Predictions Mean -70.4776 +trainer/Q2 Predictions Std 21.0897 +trainer/Q2 Predictions Max 0.180272 +trainer/Q2 Predictions Min -86.4694 +trainer/Q Targets Mean -70.7138 +trainer/Q Targets Std 21.182 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7569 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00290034 +trainer/policy/mean Std 0.730225 +trainer/policy/mean Max 0.997897 +trainer/policy/mean Min -0.999459 +trainer/policy/std Mean 0.419052 +trainer/policy/std Std 0.0217437 +trainer/policy/std Max 0.443942 +trainer/policy/std Min 0.381883 +trainer/Advantage Weights Mean 4.08313 +trainer/Advantage Weights Std 15.5012 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.87483e-20 +trainer/Advantage Score Mean -0.303438 +trainer/Advantage Score Std 0.605725 +trainer/Advantage Score Max 0.956366 +trainer/Advantage Score Min -4.49957 +trainer/V1 Predictions Mean -70.3983 +trainer/V1 Predictions Std 21.3916 +trainer/V1 Predictions Max 0.891768 +trainer/V1 Predictions Min -86.5489 +trainer/VF Loss 0.0597139 +expl/num steps total 487000 +expl/num paths total 602 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.1128 +expl/Actions Std 0.806684 +expl/Actions Max 2.31866 +expl/Actions Min -2.32348 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 457484 +eval/num paths total 487 +eval/path length Mean 596 +eval/path length Std 0 +eval/path length Max 596 +eval/path length Min 596 +eval/Rewards Mean 0.00167785 +eval/Rewards Std 0.0409272 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0258088 +eval/Actions Std 0.725264 +eval/Actions Max 0.999501 +eval/Actions Min -0.999602 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.23288e-05 +time/evaluation sampling (s) 5.07043 +time/exploration sampling (s) 6.46784 +time/logging (s) 0.0109087 +time/saving (s) 0.0164719 +time/training (s) 18.638 +time/epoch (s) 30.2037 +time/total (s) 11248.6 +Epoch -514 +------------------------------ ---------------- +2022-05-15 21:10:26.329489 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -513 finished +------------------------------ ---------------- +epoch -513 +replay_buffer/size 999047 +trainer/num train calls 488000 +trainer/QF1 Loss 0.628572 +trainer/QF2 Loss 0.651707 +trainer/Policy Loss 9.26479 +trainer/Q1 Predictions Mean -72.7216 +trainer/Q1 Predictions Std 18.0303 +trainer/Q1 Predictions Max -0.337027 +trainer/Q1 Predictions Min -86.371 +trainer/Q2 Predictions Mean -72.718 +trainer/Q2 Predictions Std 18.0025 +trainer/Q2 Predictions Max -0.395628 +trainer/Q2 Predictions Min -86.1954 +trainer/Q Targets Mean -72.8489 +trainer/Q Targets Std 18.1543 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4851 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0269033 +trainer/policy/mean Std 0.733288 +trainer/policy/mean Max 0.999311 +trainer/policy/mean Min -0.999502 +trainer/policy/std Mean 0.419885 +trainer/policy/std Std 0.0199416 +trainer/policy/std Max 0.440523 +trainer/policy/std Min 0.385425 +trainer/Advantage Weights Mean 1.13968 +trainer/Advantage Weights Std 6.62111 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.56756e-17 +trainer/Advantage Score Mean -0.375371 +trainer/Advantage Score Std 0.543119 +trainer/Advantage Score Max 2.64309 +trainer/Advantage Score Min -3.8201 +trainer/V1 Predictions Mean -72.5961 +trainer/V1 Predictions Std 18.3112 +trainer/V1 Predictions Max 0.719268 +trainer/V1 Predictions Min -86.3783 +trainer/VF Loss 0.0675805 +expl/num steps total 488000 +expl/num paths total 603 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0254778 +expl/Actions Std 0.86945 +expl/Actions Max 2.54899 +expl/Actions Min -2.31455 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 458484 +eval/num paths total 488 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0529928 +eval/Actions Std 0.69288 +eval/Actions Max 0.999758 +eval/Actions Min -0.9988 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.25361e-05 +time/evaluation sampling (s) 4.9143 +time/exploration sampling (s) 6.60809 +time/logging (s) 0.0112144 +time/saving (s) 0.014962 +time/training (s) 18.73 +time/epoch (s) 30.2785 +time/total (s) 11278.9 +Epoch -513 +------------------------------ ---------------- +2022-05-15 21:10:57.339072 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -512 finished +------------------------------ ---------------- +epoch -512 +replay_buffer/size 999047 +trainer/num train calls 489000 +trainer/QF1 Loss 0.894331 +trainer/QF2 Loss 0.857427 +trainer/Policy Loss 10.332 +trainer/Q1 Predictions Mean -75.4676 +trainer/Q1 Predictions Std 16.1014 +trainer/Q1 Predictions Max -1.7765 +trainer/Q1 Predictions Min -86.7968 +trainer/Q2 Predictions Mean -75.379 +trainer/Q2 Predictions Std 16.1049 +trainer/Q2 Predictions Max -0.691794 +trainer/Q2 Predictions Min -86.7248 +trainer/Q Targets Mean -75.4188 +trainer/Q Targets Std 15.7004 +trainer/Q Targets Max -2.71186 +trainer/Q Targets Min -86.5976 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00295136 +trainer/policy/mean Std 0.712581 +trainer/policy/mean Max 0.999394 +trainer/policy/mean Min -0.99868 +trainer/policy/std Mean 0.419789 +trainer/policy/std Std 0.0209635 +trainer/policy/std Max 0.441266 +trainer/policy/std Min 0.384962 +trainer/Advantage Weights Mean 2.72349 +trainer/Advantage Weights Std 14.894 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.60903e-14 +trainer/Advantage Score Mean -0.451404 +trainer/Advantage Score Std 0.476646 +trainer/Advantage Score Max 1.40757 +trainer/Advantage Score Min -3.17606 +trainer/V1 Predictions Mean -75.0915 +trainer/V1 Predictions Std 15.9607 +trainer/V1 Predictions Max -2.26513 +trainer/V1 Predictions Min -86.4235 +trainer/VF Loss 0.0589294 +expl/num steps total 489000 +expl/num paths total 605 +expl/path length Mean 500 +expl/path length Std 215 +expl/path length Max 715 +expl/path length Min 285 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0099897 +expl/Actions Std 0.819427 +expl/Actions Max 2.20765 +expl/Actions Min -2.80119 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 459484 +eval/num paths total 489 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.15582 +eval/Actions Std 0.727574 +eval/Actions Max 0.998893 +eval/Actions Min -0.999654 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.11924e-06 +time/evaluation sampling (s) 5.1994 +time/exploration sampling (s) 6.86604 +time/logging (s) 0.0126949 +time/saving (s) 0.0222581 +time/training (s) 18.898 +time/epoch (s) 30.9984 +time/total (s) 11309.9 +Epoch -512 +------------------------------ ---------------- +2022-05-15 21:11:27.517918 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -511 finished +------------------------------ ---------------- +epoch -511 +replay_buffer/size 999047 +trainer/num train calls 490000 +trainer/QF1 Loss 1.15247 +trainer/QF2 Loss 1.17168 +trainer/Policy Loss 38.587 +trainer/Q1 Predictions Mean -74.1429 +trainer/Q1 Predictions Std 17.3824 +trainer/Q1 Predictions Max -4.41164 +trainer/Q1 Predictions Min -87.827 +trainer/Q2 Predictions Mean -74.0416 +trainer/Q2 Predictions Std 17.3373 +trainer/Q2 Predictions Max -3.92437 +trainer/Q2 Predictions Min -87.4617 +trainer/Q Targets Mean -74.0496 +trainer/Q Targets Std 17.4378 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5967 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00363221 +trainer/policy/mean Std 0.71915 +trainer/policy/mean Max 0.999792 +trainer/policy/mean Min -0.999748 +trainer/policy/std Mean 0.420318 +trainer/policy/std Std 0.020636 +trainer/policy/std Max 0.440488 +trainer/policy/std Min 0.384939 +trainer/Advantage Weights Mean 8.44876 +trainer/Advantage Weights Std 23.0144 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.29102e-13 +trainer/Advantage Score Mean -0.204758 +trainer/Advantage Score Std 0.531902 +trainer/Advantage Score Max 1.82257 +trainer/Advantage Score Min -2.80945 +trainer/V1 Predictions Mean -73.7999 +trainer/V1 Predictions Std 17.5377 +trainer/V1 Predictions Max -3.15831 +trainer/V1 Predictions Min -87.7601 +trainer/VF Loss 0.0664118 +expl/num steps total 490000 +expl/num paths total 607 +expl/path length Mean 500 +expl/path length Std 423 +expl/path length Max 923 +expl/path length Min 77 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0463118 +expl/Actions Std 0.818941 +expl/Actions Max 2.35759 +expl/Actions Min -2.22979 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 460484 +eval/num paths total 490 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.176765 +eval/Actions Std 0.719292 +eval/Actions Max 0.999509 +eval/Actions Min -0.999365 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.38762e-05 +time/evaluation sampling (s) 5.16403 +time/exploration sampling (s) 5.78196 +time/logging (s) 0.0110103 +time/saving (s) 0.016734 +time/training (s) 19.1877 +time/epoch (s) 30.1614 +time/total (s) 11340.1 +Epoch -511 +------------------------------ ---------------- +2022-05-15 21:11:58.350550 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -510 finished +------------------------------ ---------------- +epoch -510 +replay_buffer/size 999047 +trainer/num train calls 491000 +trainer/QF1 Loss 0.636962 +trainer/QF2 Loss 0.582861 +trainer/Policy Loss 18.817 +trainer/Q1 Predictions Mean -73.8099 +trainer/Q1 Predictions Std 17.3794 +trainer/Q1 Predictions Max -0.332354 +trainer/Q1 Predictions Min -86.7153 +trainer/Q2 Predictions Mean -73.7595 +trainer/Q2 Predictions Std 17.4416 +trainer/Q2 Predictions Max -0.499155 +trainer/Q2 Predictions Min -86.425 +trainer/Q Targets Mean -73.6309 +trainer/Q Targets Std 17.5571 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5887 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000921987 +trainer/policy/mean Std 0.726386 +trainer/policy/mean Max 0.999712 +trainer/policy/mean Min -0.998586 +trainer/policy/std Mean 0.421017 +trainer/policy/std Std 0.020679 +trainer/policy/std Max 0.442735 +trainer/policy/std Min 0.386505 +trainer/Advantage Weights Mean 3.83385 +trainer/Advantage Weights Std 15.1303 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.04338e-14 +trainer/Advantage Score Mean -0.307825 +trainer/Advantage Score Std 0.527795 +trainer/Advantage Score Max 0.725609 +trainer/Advantage Score Min -3.21937 +trainer/V1 Predictions Mean -73.3417 +trainer/V1 Predictions Std 17.7695 +trainer/V1 Predictions Max -0.0859206 +trainer/V1 Predictions Min -86.373 +trainer/VF Loss 0.0469574 +expl/num steps total 491000 +expl/num paths total 609 +expl/path length Mean 500 +expl/path length Std 387 +expl/path length Max 887 +expl/path length Min 113 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0474731 +expl/Actions Std 0.845432 +expl/Actions Max 2.21303 +expl/Actions Min -2.26145 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 461484 +eval/num paths total 491 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.137232 +eval/Actions Std 0.718483 +eval/Actions Max 0.999468 +eval/Actions Min -0.999542 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.06888e-05 +time/evaluation sampling (s) 4.57768 +time/exploration sampling (s) 6.91392 +time/logging (s) 0.0115524 +time/saving (s) 0.0155196 +time/training (s) 19.3011 +time/epoch (s) 30.8198 +time/total (s) 11370.9 +Epoch -510 +------------------------------ ---------------- +2022-05-15 21:12:28.649181 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -509 finished +------------------------------ ---------------- +epoch -509 +replay_buffer/size 999047 +trainer/num train calls 492000 +trainer/QF1 Loss 0.925526 +trainer/QF2 Loss 0.832839 +trainer/Policy Loss 7.93698 +trainer/Q1 Predictions Mean -74.7294 +trainer/Q1 Predictions Std 15.8727 +trainer/Q1 Predictions Max -2.68495 +trainer/Q1 Predictions Min -88.1202 +trainer/Q2 Predictions Mean -74.772 +trainer/Q2 Predictions Std 15.8546 +trainer/Q2 Predictions Max -3.97186 +trainer/Q2 Predictions Min -88.0341 +trainer/Q Targets Mean -74.3597 +trainer/Q Targets Std 15.9018 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4775 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.036551 +trainer/policy/mean Std 0.72684 +trainer/policy/mean Max 0.999317 +trainer/policy/mean Min -0.998938 +trainer/policy/std Mean 0.420449 +trainer/policy/std Std 0.0198873 +trainer/policy/std Max 0.440542 +trainer/policy/std Min 0.387187 +trainer/Advantage Weights Mean 2.7823 +trainer/Advantage Weights Std 11.9965 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.48694e-12 +trainer/Advantage Score Mean -0.369382 +trainer/Advantage Score Std 0.538454 +trainer/Advantage Score Max 2.13752 +trainer/Advantage Score Min -2.59287 +trainer/V1 Predictions Mean -74.1268 +trainer/V1 Predictions Std 16.032 +trainer/V1 Predictions Max -3.73736 +trainer/V1 Predictions Min -87.3525 +trainer/VF Loss 0.0638706 +expl/num steps total 492000 +expl/num paths total 610 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0100645 +expl/Actions Std 0.866626 +expl/Actions Max 2.3176 +expl/Actions Min -2.38265 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 462484 +eval/num paths total 492 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.394488 +eval/Actions Std 0.605578 +eval/Actions Max 0.999058 +eval/Actions Min -0.998809 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.04682e-06 +time/evaluation sampling (s) 4.70871 +time/exploration sampling (s) 6.70029 +time/logging (s) 0.01461 +time/saving (s) 0.0198592 +time/training (s) 18.8444 +time/epoch (s) 30.2878 +time/total (s) 11401.2 +Epoch -509 +------------------------------ ---------------- +2022-05-15 21:13:00.087536 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -508 finished +------------------------------ ---------------- +epoch -508 +replay_buffer/size 999047 +trainer/num train calls 493000 +trainer/QF1 Loss 0.746321 +trainer/QF2 Loss 0.759822 +trainer/Policy Loss 26.6311 +trainer/Q1 Predictions Mean -73.011 +trainer/Q1 Predictions Std 18.2583 +trainer/Q1 Predictions Max -0.731496 +trainer/Q1 Predictions Min -86.3065 +trainer/Q2 Predictions Mean -73.0798 +trainer/Q2 Predictions Std 18.1505 +trainer/Q2 Predictions Max -0.989461 +trainer/Q2 Predictions Min -86.2277 +trainer/Q Targets Mean -73.3566 +trainer/Q Targets Std 18.2843 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6852 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0260762 +trainer/policy/mean Std 0.72235 +trainer/policy/mean Max 0.999226 +trainer/policy/mean Min -0.998214 +trainer/policy/std Mean 0.419747 +trainer/policy/std Std 0.0205609 +trainer/policy/std Max 0.439762 +trainer/policy/std Min 0.382869 +trainer/Advantage Weights Mean 7.20209 +trainer/Advantage Weights Std 21.5338 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.17296e-11 +trainer/Advantage Score Mean -0.17649 +trainer/Advantage Score Std 0.483709 +trainer/Advantage Score Max 2.18934 +trainer/Advantage Score Min -2.41738 +trainer/V1 Predictions Mean -73.1648 +trainer/V1 Predictions Std 18.2544 +trainer/V1 Predictions Max 0.216124 +trainer/V1 Predictions Min -86.517 +trainer/VF Loss 0.0632788 +expl/num steps total 493000 +expl/num paths total 612 +expl/path length Mean 500 +expl/path length Std 141 +expl/path length Max 641 +expl/path length Min 359 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0345439 +expl/Actions Std 0.829766 +expl/Actions Max 2.32157 +expl/Actions Min -2.76361 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 463172 +eval/num paths total 493 +eval/path length Mean 688 +eval/path length Std 0 +eval/path length Max 688 +eval/path length Min 688 +eval/Rewards Mean 0.00145349 +eval/Rewards Std 0.0380969 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0110585 +eval/Actions Std 0.727814 +eval/Actions Max 0.999918 +eval/Actions Min -0.999892 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 6.21798e-06 +time/evaluation sampling (s) 5.08708 +time/exploration sampling (s) 7.45349 +time/logging (s) 0.0108467 +time/saving (s) 0.016893 +time/training (s) 18.8484 +time/epoch (s) 31.4168 +time/total (s) 11432.6 +Epoch -508 +------------------------------ ---------------- +2022-05-15 21:13:30.348415 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -507 finished +------------------------------ ---------------- +epoch -507 +replay_buffer/size 999047 +trainer/num train calls 494000 +trainer/QF1 Loss 1.30974 +trainer/QF2 Loss 1.30466 +trainer/Policy Loss 30.1333 +trainer/Q1 Predictions Mean -71.9245 +trainer/Q1 Predictions Std 18.8272 +trainer/Q1 Predictions Max -1.43713 +trainer/Q1 Predictions Min -86.8655 +trainer/Q2 Predictions Mean -71.9345 +trainer/Q2 Predictions Std 18.7571 +trainer/Q2 Predictions Max -0.951228 +trainer/Q2 Predictions Min -86.79 +trainer/Q Targets Mean -72.3158 +trainer/Q Targets Std 18.5717 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5712 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0128898 +trainer/policy/mean Std 0.729236 +trainer/policy/mean Max 0.999778 +trainer/policy/mean Min -0.998167 +trainer/policy/std Mean 0.418897 +trainer/policy/std Std 0.020302 +trainer/policy/std Max 0.439881 +trainer/policy/std Min 0.383072 +trainer/Advantage Weights Mean 7.46163 +trainer/Advantage Weights Std 21.3953 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.44937e-10 +trainer/Advantage Score Mean -0.1593 +trainer/Advantage Score Std 0.484226 +trainer/Advantage Score Max 1.81213 +trainer/Advantage Score Min -2.26547 +trainer/V1 Predictions Mean -72.0693 +trainer/V1 Predictions Std 18.6507 +trainer/V1 Predictions Max -0.592013 +trainer/V1 Predictions Min -86.4011 +trainer/VF Loss 0.0728847 +expl/num steps total 494000 +expl/num paths total 614 +expl/path length Mean 500 +expl/path length Std 208 +expl/path length Max 708 +expl/path length Min 292 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0426067 +expl/Actions Std 0.854371 +expl/Actions Max 2.21764 +expl/Actions Min -2.43004 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 464049 +eval/num paths total 494 +eval/path length Mean 877 +eval/path length Std 0 +eval/path length Max 877 +eval/path length Min 877 +eval/Rewards Mean 0.00114025 +eval/Rewards Std 0.0337483 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0679447 +eval/Actions Std 0.75657 +eval/Actions Max 0.999944 +eval/Actions Min -0.999646 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.27261e-05 +time/evaluation sampling (s) 4.8281 +time/exploration sampling (s) 6.9108 +time/logging (s) 0.0138664 +time/saving (s) 0.0203189 +time/training (s) 18.4761 +time/epoch (s) 30.2492 +time/total (s) 11462.9 +Epoch -507 +------------------------------ ---------------- +2022-05-15 21:14:01.242887 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -506 finished +------------------------------ ---------------- +epoch -506 +replay_buffer/size 999047 +trainer/num train calls 495000 +trainer/QF1 Loss 0.438705 +trainer/QF2 Loss 0.444705 +trainer/Policy Loss 10.7275 +trainer/Q1 Predictions Mean -73.4395 +trainer/Q1 Predictions Std 18.9018 +trainer/Q1 Predictions Max -0.598781 +trainer/Q1 Predictions Min -86.2538 +trainer/Q2 Predictions Mean -73.4127 +trainer/Q2 Predictions Std 18.8373 +trainer/Q2 Predictions Max -0.596938 +trainer/Q2 Predictions Min -87.0975 +trainer/Q Targets Mean -73.4039 +trainer/Q Targets Std 18.9054 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.2525 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0136258 +trainer/policy/mean Std 0.718571 +trainer/policy/mean Max 0.998719 +trainer/policy/mean Min -0.997849 +trainer/policy/std Mean 0.420154 +trainer/policy/std Std 0.0184937 +trainer/policy/std Max 0.439188 +trainer/policy/std Min 0.387821 +trainer/Advantage Weights Mean 0.862932 +trainer/Advantage Weights Std 7.03439 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.34502e-16 +trainer/Advantage Score Mean -0.551334 +trainer/Advantage Score Std 0.530008 +trainer/Advantage Score Max 0.806412 +trainer/Advantage Score Min -3.56339 +trainer/V1 Predictions Mean -73.075 +trainer/V1 Predictions Std 19.0621 +trainer/V1 Predictions Max -0.0500981 +trainer/V1 Predictions Min -86.4434 +trainer/VF Loss 0.0616763 +expl/num steps total 495000 +expl/num paths total 615 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0509384 +expl/Actions Std 0.806349 +expl/Actions Max 2.65695 +expl/Actions Min -2.33636 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 464633 +eval/num paths total 495 +eval/path length Mean 584 +eval/path length Std 0 +eval/path length Max 584 +eval/path length Min 584 +eval/Rewards Mean 0.00171233 +eval/Rewards Std 0.0413449 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0321955 +eval/Actions Std 0.724607 +eval/Actions Max 0.9996 +eval/Actions Min -0.99739 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.94531e-06 +time/evaluation sampling (s) 5.70918 +time/exploration sampling (s) 6.5893 +time/logging (s) 0.0106881 +time/saving (s) 0.0156658 +time/training (s) 18.5504 +time/epoch (s) 30.8753 +time/total (s) 11493.8 +Epoch -506 +------------------------------ ---------------- +2022-05-15 21:14:32.542734 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -505 finished +------------------------------ ---------------- +epoch -505 +replay_buffer/size 999047 +trainer/num train calls 496000 +trainer/QF1 Loss 1.10598 +trainer/QF2 Loss 1.08721 +trainer/Policy Loss 44.862 +trainer/Q1 Predictions Mean -71.9175 +trainer/Q1 Predictions Std 19.1481 +trainer/Q1 Predictions Max -0.244616 +trainer/Q1 Predictions Min -86.6706 +trainer/Q2 Predictions Mean -71.8646 +trainer/Q2 Predictions Std 19.1809 +trainer/Q2 Predictions Max -0.466355 +trainer/Q2 Predictions Min -86.3781 +trainer/Q Targets Mean -72.4554 +trainer/Q Targets Std 19.4966 +trainer/Q Targets Max 1.82071 +trainer/Q Targets Min -86.7473 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00801303 +trainer/policy/mean Std 0.729858 +trainer/policy/mean Max 0.998605 +trainer/policy/mean Min -0.998203 +trainer/policy/std Mean 0.419817 +trainer/policy/std Std 0.019769 +trainer/policy/std Max 0.441582 +trainer/policy/std Min 0.386105 +trainer/Advantage Weights Mean 10.7417 +trainer/Advantage Weights Std 24.8959 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.91174e-18 +trainer/Advantage Score Mean -0.168761 +trainer/Advantage Score Std 0.628314 +trainer/Advantage Score Max 1.90894 +trainer/Advantage Score Min -4.00826 +trainer/V1 Predictions Mean -72.2677 +trainer/V1 Predictions Std 19.5121 +trainer/V1 Predictions Max 1.16906 +trainer/V1 Predictions Min -86.6148 +trainer/VF Loss 0.0880704 +expl/num steps total 496000 +expl/num paths total 616 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.102635 +expl/Actions Std 0.820364 +expl/Actions Max 2.49498 +expl/Actions Min -2.19637 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 465348 +eval/num paths total 496 +eval/path length Mean 715 +eval/path length Std 0 +eval/path length Max 715 +eval/path length Min 715 +eval/Rewards Mean 0.0013986 +eval/Rewards Std 0.0373717 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00386122 +eval/Actions Std 0.741845 +eval/Actions Max 0.999144 +eval/Actions Min -0.999824 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.20701e-06 +time/evaluation sampling (s) 4.75652 +time/exploration sampling (s) 7.60964 +time/logging (s) 0.011009 +time/saving (s) 0.0223124 +time/training (s) 18.8894 +time/epoch (s) 31.2889 +time/total (s) 11525.1 +Epoch -505 +------------------------------ ---------------- +2022-05-15 21:15:03.245017 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -504 finished +------------------------------ ---------------- +epoch -504 +replay_buffer/size 999047 +trainer/num train calls 497000 +trainer/QF1 Loss 0.671757 +trainer/QF2 Loss 0.575399 +trainer/Policy Loss 15.8453 +trainer/Q1 Predictions Mean -72.6659 +trainer/Q1 Predictions Std 19.2938 +trainer/Q1 Predictions Max -0.26298 +trainer/Q1 Predictions Min -86.6608 +trainer/Q2 Predictions Mean -72.6196 +trainer/Q2 Predictions Std 19.3995 +trainer/Q2 Predictions Max -0.397343 +trainer/Q2 Predictions Min -86.6298 +trainer/Q Targets Mean -72.6858 +trainer/Q Targets Std 19.358 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6273 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0131564 +trainer/policy/mean Std 0.719538 +trainer/policy/mean Max 0.99979 +trainer/policy/mean Min -0.998942 +trainer/policy/std Mean 0.419318 +trainer/policy/std Std 0.0202569 +trainer/policy/std Max 0.439652 +trainer/policy/std Min 0.384038 +trainer/Advantage Weights Mean 3.7145 +trainer/Advantage Weights Std 17.5068 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.66424e-20 +trainer/Advantage Score Mean -0.415707 +trainer/Advantage Score Std 0.577418 +trainer/Advantage Score Max 1.50876 +trainer/Advantage Score Min -4.55423 +trainer/V1 Predictions Mean -72.4342 +trainer/V1 Predictions Std 19.3725 +trainer/V1 Predictions Max 0.745861 +trainer/V1 Predictions Min -86.4571 +trainer/VF Loss 0.0670526 +expl/num steps total 497000 +expl/num paths total 618 +expl/path length Mean 500 +expl/path length Std 276 +expl/path length Max 776 +expl/path length Min 224 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0378081 +expl/Actions Std 0.838235 +expl/Actions Max 2.24729 +expl/Actions Min -2.81369 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 466348 +eval/num paths total 497 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0836731 +eval/Actions Std 0.708623 +eval/Actions Max 0.997864 +eval/Actions Min -0.999477 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.39298e-05 +time/evaluation sampling (s) 5.03721 +time/exploration sampling (s) 6.43821 +time/logging (s) 0.0126685 +time/saving (s) 0.0187713 +time/training (s) 19.1809 +time/epoch (s) 30.6878 +time/total (s) 11555.8 +Epoch -504 +------------------------------ ---------------- +2022-05-15 21:15:33.476790 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -503 finished +------------------------------ ---------------- +epoch -503 +replay_buffer/size 999047 +trainer/num train calls 498000 +trainer/QF1 Loss 0.577226 +trainer/QF2 Loss 0.624909 +trainer/Policy Loss 30.3226 +trainer/Q1 Predictions Mean -73.2179 +trainer/Q1 Predictions Std 17.1601 +trainer/Q1 Predictions Max -0.547969 +trainer/Q1 Predictions Min -87.8306 +trainer/Q2 Predictions Mean -73.192 +trainer/Q2 Predictions Std 17.1364 +trainer/Q2 Predictions Max -1.21524 +trainer/Q2 Predictions Min -88.4105 +trainer/Q Targets Mean -73.2605 +trainer/Q Targets Std 16.8781 +trainer/Q Targets Max -2.06596 +trainer/Q Targets Min -87.425 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0122918 +trainer/policy/mean Std 0.708848 +trainer/policy/mean Max 0.999392 +trainer/policy/mean Min -0.99781 +trainer/policy/std Mean 0.421512 +trainer/policy/std Std 0.0202959 +trainer/policy/std Max 0.444057 +trainer/policy/std Min 0.387771 +trainer/Advantage Weights Mean 6.49809 +trainer/Advantage Weights Std 21.2831 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.89414e-18 +trainer/Advantage Score Mean -0.222484 +trainer/Advantage Score Std 0.57093 +trainer/Advantage Score Max 2.84269 +trainer/Advantage Score Min -3.98585 +trainer/V1 Predictions Mean -72.9424 +trainer/V1 Predictions Std 17.1848 +trainer/V1 Predictions Max -0.587264 +trainer/V1 Predictions Min -87.1852 +trainer/VF Loss 0.0948581 +expl/num steps total 498000 +expl/num paths total 619 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.36409 +expl/Actions Std 0.823161 +expl/Actions Max 2.97481 +expl/Actions Min -2.48927 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 467140 +eval/num paths total 498 +eval/path length Mean 792 +eval/path length Std 0 +eval/path length Max 792 +eval/path length Min 792 +eval/Rewards Mean 0.00126263 +eval/Rewards Std 0.035511 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0116718 +eval/Actions Std 0.714382 +eval/Actions Max 0.999267 +eval/Actions Min -0.999674 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.11251e-05 +time/evaluation sampling (s) 4.74872 +time/exploration sampling (s) 6.69649 +time/logging (s) 0.0112134 +time/saving (s) 0.0175735 +time/training (s) 18.7402 +time/epoch (s) 30.2142 +time/total (s) 11586 +Epoch -503 +------------------------------ ---------------- +2022-05-15 21:16:04.151535 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -502 finished +------------------------------ ---------------- +epoch -502 +replay_buffer/size 999047 +trainer/num train calls 499000 +trainer/QF1 Loss 1.01999 +trainer/QF2 Loss 0.967046 +trainer/Policy Loss 34.4326 +trainer/Q1 Predictions Mean -73.5408 +trainer/Q1 Predictions Std 17.0815 +trainer/Q1 Predictions Max -0.827392 +trainer/Q1 Predictions Min -86.6153 +trainer/Q2 Predictions Mean -73.5296 +trainer/Q2 Predictions Std 17.101 +trainer/Q2 Predictions Max -0.571636 +trainer/Q2 Predictions Min -86.4299 +trainer/Q Targets Mean -73.8453 +trainer/Q Targets Std 17.2647 +trainer/Q Targets Max -2.28549 +trainer/Q Targets Min -86.5534 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0158744 +trainer/policy/mean Std 0.717726 +trainer/policy/mean Max 0.999525 +trainer/policy/mean Min -0.99918 +trainer/policy/std Mean 0.41895 +trainer/policy/std Std 0.0209099 +trainer/policy/std Max 0.44438 +trainer/policy/std Min 0.384977 +trainer/Advantage Weights Mean 7.72272 +trainer/Advantage Weights Std 21.0902 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.99899e-12 +trainer/Advantage Score Mean -0.273505 +trainer/Advantage Score Std 0.606757 +trainer/Advantage Score Max 1.34882 +trainer/Advantage Score Min -2.65327 +trainer/V1 Predictions Mean -73.5397 +trainer/V1 Predictions Std 17.4361 +trainer/V1 Predictions Max -1.52699 +trainer/V1 Predictions Min -86.3494 +trainer/VF Loss 0.0690479 +expl/num steps total 499000 +expl/num paths total 620 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0570149 +expl/Actions Std 0.884743 +expl/Actions Max 2.74148 +expl/Actions Min -2.42642 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 467753 +eval/num paths total 499 +eval/path length Mean 613 +eval/path length Std 0 +eval/path length Max 613 +eval/path length Min 613 +eval/Rewards Mean 0.00163132 +eval/Rewards Std 0.0403567 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0280935 +eval/Actions Std 0.734849 +eval/Actions Max 0.999945 +eval/Actions Min -0.999726 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.02278e-05 +time/evaluation sampling (s) 4.67059 +time/exploration sampling (s) 6.83826 +time/logging (s) 0.0106908 +time/saving (s) 0.0166346 +time/training (s) 19.1235 +time/epoch (s) 30.6597 +time/total (s) 11616.6 +Epoch -502 +------------------------------ ---------------- +2022-05-15 21:16:35.039970 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -501 finished +------------------------------ ---------------- +epoch -501 +replay_buffer/size 999047 +trainer/num train calls 500000 +trainer/QF1 Loss 0.938554 +trainer/QF2 Loss 1.00409 +trainer/Policy Loss 50.9796 +trainer/Q1 Predictions Mean -71.2464 +trainer/Q1 Predictions Std 20.7342 +trainer/Q1 Predictions Max -0.531943 +trainer/Q1 Predictions Min -85.8356 +trainer/Q2 Predictions Mean -71.2876 +trainer/Q2 Predictions Std 20.7532 +trainer/Q2 Predictions Max -0.510479 +trainer/Q2 Predictions Min -85.8905 +trainer/Q Targets Mean -71.7099 +trainer/Q Targets Std 20.4759 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3874 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00145174 +trainer/policy/mean Std 0.725146 +trainer/policy/mean Max 0.999878 +trainer/policy/mean Min -0.999876 +trainer/policy/std Mean 0.418968 +trainer/policy/std Std 0.0201264 +trainer/policy/std Max 0.442393 +trainer/policy/std Min 0.388288 +trainer/Advantage Weights Mean 9.832 +trainer/Advantage Weights Std 26.2995 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.76544e-12 +trainer/Advantage Score Mean -0.170859 +trainer/Advantage Score Std 0.551433 +trainer/Advantage Score Max 2.2022 +trainer/Advantage Score Min -2.57192 +trainer/V1 Predictions Mean -71.4524 +trainer/V1 Predictions Std 20.5012 +trainer/V1 Predictions Max -0.0939435 +trainer/V1 Predictions Min -86.3294 +trainer/VF Loss 0.0867755 +expl/num steps total 500000 +expl/num paths total 621 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00225226 +expl/Actions Std 0.876541 +expl/Actions Max 2.20312 +expl/Actions Min -2.2105 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 468753 +eval/num paths total 500 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0494102 +eval/Actions Std 0.744564 +eval/Actions Max 0.999841 +eval/Actions Min -0.99883 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.2354e-05 +time/evaluation sampling (s) 4.81178 +time/exploration sampling (s) 7.0748 +time/logging (s) 0.0096737 +time/saving (s) 0.0199542 +time/training (s) 18.957 +time/epoch (s) 30.8732 +time/total (s) 11647.5 +Epoch -501 +------------------------------ ---------------- +2022-05-15 21:17:05.881192 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -500 finished +------------------------------ ---------------- +epoch -500 +replay_buffer/size 999047 +trainer/num train calls 501000 +trainer/QF1 Loss 0.517913 +trainer/QF2 Loss 0.588175 +trainer/Policy Loss 20.9497 +trainer/Q1 Predictions Mean -73.4416 +trainer/Q1 Predictions Std 18.4734 +trainer/Q1 Predictions Max -0.575414 +trainer/Q1 Predictions Min -86.7122 +trainer/Q2 Predictions Mean -73.4402 +trainer/Q2 Predictions Std 18.5333 +trainer/Q2 Predictions Max -0.505312 +trainer/Q2 Predictions Min -86.8202 +trainer/Q Targets Mean -73.1989 +trainer/Q Targets Std 18.5225 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7499 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0106175 +trainer/policy/mean Std 0.720326 +trainer/policy/mean Max 0.997642 +trainer/policy/mean Min -0.998201 +trainer/policy/std Mean 0.419774 +trainer/policy/std Std 0.0212147 +trainer/policy/std Max 0.4459 +trainer/policy/std Min 0.386925 +trainer/Advantage Weights Mean 4.05106 +trainer/Advantage Weights Std 16.7615 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.13779e-18 +trainer/Advantage Score Mean -0.354414 +trainer/Advantage Score Std 0.614817 +trainer/Advantage Score Max 1.26984 +trainer/Advantage Score Min -4.13174 +trainer/V1 Predictions Mean -72.9595 +trainer/V1 Predictions Std 18.7401 +trainer/V1 Predictions Max 1.30827 +trainer/V1 Predictions Min -86.4079 +trainer/VF Loss 0.0699266 +expl/num steps total 501000 +expl/num paths total 622 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0576391 +expl/Actions Std 0.811426 +expl/Actions Max 2.30349 +expl/Actions Min -2.2797 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 469753 +eval/num paths total 501 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.182739 +eval/Actions Std 0.780372 +eval/Actions Max 0.999728 +eval/Actions Min -0.998573 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.1004e-05 +time/evaluation sampling (s) 5.12016 +time/exploration sampling (s) 6.86504 +time/logging (s) 0.0123525 +time/saving (s) 0.0271794 +time/training (s) 18.8058 +time/epoch (s) 30.8305 +time/total (s) 11678.4 +Epoch -500 +------------------------------ ---------------- +2022-05-15 21:17:36.268839 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -499 finished +------------------------------ ---------------- +epoch -499 +replay_buffer/size 999047 +trainer/num train calls 502000 +trainer/QF1 Loss 0.929557 +trainer/QF2 Loss 0.866106 +trainer/Policy Loss 12.3432 +trainer/Q1 Predictions Mean -73.5897 +trainer/Q1 Predictions Std 17.703 +trainer/Q1 Predictions Max -0.950456 +trainer/Q1 Predictions Min -86.8946 +trainer/Q2 Predictions Mean -73.5459 +trainer/Q2 Predictions Std 17.735 +trainer/Q2 Predictions Max -0.0312251 +trainer/Q2 Predictions Min -87.2171 +trainer/Q Targets Mean -73.0687 +trainer/Q Targets Std 17.8006 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6512 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0148118 +trainer/policy/mean Std 0.717351 +trainer/policy/mean Max 0.99897 +trainer/policy/mean Min -0.998411 +trainer/policy/std Mean 0.419649 +trainer/policy/std Std 0.0211437 +trainer/policy/std Max 0.445413 +trainer/policy/std Min 0.38522 +trainer/Advantage Weights Mean 1.26471 +trainer/Advantage Weights Std 9.96037 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.25592e-15 +trainer/Advantage Score Mean -0.752135 +trainer/Advantage Score Std 0.644344 +trainer/Advantage Score Max 0.819325 +trainer/Advantage Score Min -3.43109 +trainer/V1 Predictions Mean -72.8087 +trainer/V1 Predictions Std 17.9677 +trainer/V1 Predictions Max -0.351695 +trainer/V1 Predictions Min -86.4222 +trainer/VF Loss 0.101951 +expl/num steps total 502000 +expl/num paths total 623 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0663864 +expl/Actions Std 0.848152 +expl/Actions Max 2.37029 +expl/Actions Min -2.21777 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 470500 +eval/num paths total 502 +eval/path length Mean 747 +eval/path length Std 0 +eval/path length Max 747 +eval/path length Min 747 +eval/Rewards Mean 0.00133869 +eval/Rewards Std 0.0365636 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0500193 +eval/Actions Std 0.741118 +eval/Actions Max 0.99985 +eval/Actions Min -0.99956 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 5.95814e-06 +time/evaluation sampling (s) 5.41168 +time/exploration sampling (s) 6.29787 +time/logging (s) 0.0116959 +time/saving (s) 0.0187893 +time/training (s) 18.632 +time/epoch (s) 30.3721 +time/total (s) 11708.7 +Epoch -499 +------------------------------ ---------------- +2022-05-15 21:18:06.641376 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -498 finished +------------------------------ ---------------- +epoch -498 +replay_buffer/size 999047 +trainer/num train calls 503000 +trainer/QF1 Loss 0.747063 +trainer/QF2 Loss 0.830752 +trainer/Policy Loss 18.0544 +trainer/Q1 Predictions Mean -74.1816 +trainer/Q1 Predictions Std 17.5648 +trainer/Q1 Predictions Max 0.466651 +trainer/Q1 Predictions Min -86.9464 +trainer/Q2 Predictions Mean -74.265 +trainer/Q2 Predictions Std 17.6245 +trainer/Q2 Predictions Max 0.522668 +trainer/Q2 Predictions Min -87.2603 +trainer/Q Targets Mean -74.0854 +trainer/Q Targets Std 17.8779 +trainer/Q Targets Max 0.976239 +trainer/Q Targets Min -86.7909 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.010519 +trainer/policy/mean Std 0.729727 +trainer/policy/mean Max 0.999762 +trainer/policy/mean Min -0.998875 +trainer/policy/std Mean 0.419054 +trainer/policy/std Std 0.0195678 +trainer/policy/std Max 0.441704 +trainer/policy/std Min 0.389767 +trainer/Advantage Weights Mean 3.00419 +trainer/Advantage Weights Std 14.0137 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.68709e-13 +trainer/Advantage Score Mean -0.428353 +trainer/Advantage Score Std 0.514814 +trainer/Advantage Score Max 1.32463 +trainer/Advantage Score Min -2.86288 +trainer/V1 Predictions Mean -73.8898 +trainer/V1 Predictions Std 17.8782 +trainer/V1 Predictions Max 1.03978 +trainer/V1 Predictions Min -86.6667 +trainer/VF Loss 0.0578572 +expl/num steps total 503000 +expl/num paths total 624 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.061204 +expl/Actions Std 0.846853 +expl/Actions Max 2.24802 +expl/Actions Min -2.3055 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 471500 +eval/num paths total 503 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0447759 +eval/Actions Std 0.741846 +eval/Actions Max 0.999851 +eval/Actions Min -0.999561 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.19638e-05 +time/evaluation sampling (s) 4.86444 +time/exploration sampling (s) 6.38299 +time/logging (s) 0.0144986 +time/saving (s) 0.0204979 +time/training (s) 19.0758 +time/epoch (s) 30.3582 +time/total (s) 11739.1 +Epoch -498 +------------------------------ ---------------- +2022-05-15 21:18:37.829178 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -497 finished +------------------------------ ---------------- +epoch -497 +replay_buffer/size 999047 +trainer/num train calls 504000 +trainer/QF1 Loss 0.879324 +trainer/QF2 Loss 0.906016 +trainer/Policy Loss 8.6803 +trainer/Q1 Predictions Mean -75.2386 +trainer/Q1 Predictions Std 15.4336 +trainer/Q1 Predictions Max -4.34047 +trainer/Q1 Predictions Min -87.2069 +trainer/Q2 Predictions Mean -75.1659 +trainer/Q2 Predictions Std 15.3544 +trainer/Q2 Predictions Max -3.23534 +trainer/Q2 Predictions Min -87.3431 +trainer/Q Targets Mean -75.001 +trainer/Q Targets Std 15.2373 +trainer/Q Targets Max -4.16022 +trainer/Q Targets Min -87.1089 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00941387 +trainer/policy/mean Std 0.729047 +trainer/policy/mean Max 0.998606 +trainer/policy/mean Min -0.998795 +trainer/policy/std Mean 0.419712 +trainer/policy/std Std 0.0207374 +trainer/policy/std Max 0.445992 +trainer/policy/std Min 0.38841 +trainer/Advantage Weights Mean 2.53908 +trainer/Advantage Weights Std 12.7438 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.84786e-12 +trainer/Advantage Score Mean -0.403875 +trainer/Advantage Score Std 0.497633 +trainer/Advantage Score Max 0.949005 +trainer/Advantage Score Min -2.58649 +trainer/V1 Predictions Mean -74.7306 +trainer/V1 Predictions Std 15.403 +trainer/V1 Predictions Max -3.99111 +trainer/V1 Predictions Min -86.7947 +trainer/VF Loss 0.0507137 +expl/num steps total 504000 +expl/num paths total 626 +expl/path length Mean 500 +expl/path length Std 317 +expl/path length Max 817 +expl/path length Min 183 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0387309 +expl/Actions Std 0.822569 +expl/Actions Max 2.31264 +expl/Actions Min -2.26487 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 472094 +eval/num paths total 504 +eval/path length Mean 594 +eval/path length Std 0 +eval/path length Max 594 +eval/path length Min 594 +eval/Rewards Mean 0.0016835 +eval/Rewards Std 0.0409959 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0507039 +eval/Actions Std 0.738001 +eval/Actions Max 0.999913 +eval/Actions Min -0.999467 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.59759e-05 +time/evaluation sampling (s) 5.19184 +time/exploration sampling (s) 6.85792 +time/logging (s) 0.0106237 +time/saving (s) 0.0187204 +time/training (s) 19.0868 +time/epoch (s) 31.1659 +time/total (s) 11770.3 +Epoch -497 +------------------------------ ---------------- +2022-05-15 21:19:09.029853 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -496 finished +------------------------------ ---------------- +epoch -496 +replay_buffer/size 999047 +trainer/num train calls 505000 +trainer/QF1 Loss 0.501151 +trainer/QF2 Loss 0.543325 +trainer/Policy Loss 15.784 +trainer/Q1 Predictions Mean -72.5563 +trainer/Q1 Predictions Std 18.1247 +trainer/Q1 Predictions Max -1.74053 +trainer/Q1 Predictions Min -86.7502 +trainer/Q2 Predictions Mean -72.6333 +trainer/Q2 Predictions Std 18.119 +trainer/Q2 Predictions Max -1.9326 +trainer/Q2 Predictions Min -86.3692 +trainer/Q Targets Mean -72.6699 +trainer/Q Targets Std 18.426 +trainer/Q Targets Max -0.650081 +trainer/Q Targets Min -86.7639 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0081415 +trainer/policy/mean Std 0.721739 +trainer/policy/mean Max 0.999182 +trainer/policy/mean Min -0.997672 +trainer/policy/std Mean 0.419013 +trainer/policy/std Std 0.0199694 +trainer/policy/std Max 0.443209 +trainer/policy/std Min 0.388832 +trainer/Advantage Weights Mean 3.485 +trainer/Advantage Weights Std 14.6405 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.64345e-12 +trainer/Advantage Score Mean -0.351766 +trainer/Advantage Score Std 0.559316 +trainer/Advantage Score Max 0.798538 +trainer/Advantage Score Min -2.66589 +trainer/V1 Predictions Mean -72.4544 +trainer/V1 Predictions Std 18.4793 +trainer/V1 Predictions Max -0.106377 +trainer/V1 Predictions Min -86.6112 +trainer/VF Loss 0.0534321 +expl/num steps total 505000 +expl/num paths total 628 +expl/path length Mean 500 +expl/path length Std 340 +expl/path length Max 840 +expl/path length Min 160 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0105421 +expl/Actions Std 0.832405 +expl/Actions Max 2.47635 +expl/Actions Min -2.61571 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 472585 +eval/num paths total 505 +eval/path length Mean 491 +eval/path length Std 0 +eval/path length Max 491 +eval/path length Min 491 +eval/Rewards Mean 0.00203666 +eval/Rewards Std 0.0450834 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00564878 +eval/Actions Std 0.74548 +eval/Actions Max 0.99978 +eval/Actions Min -0.999845 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.29449e-05 +time/evaluation sampling (s) 5.0875 +time/exploration sampling (s) 6.94217 +time/logging (s) 0.00994502 +time/saving (s) 0.0153065 +time/training (s) 19.1285 +time/epoch (s) 31.1834 +time/total (s) 11801.5 +Epoch -496 +------------------------------ ---------------- +2022-05-15 21:19:40.051706 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -495 finished +------------------------------ ---------------- +epoch -495 +replay_buffer/size 999047 +trainer/num train calls 506000 +trainer/QF1 Loss 0.54001 +trainer/QF2 Loss 0.579638 +trainer/Policy Loss 27.295 +trainer/Q1 Predictions Mean -73.4793 +trainer/Q1 Predictions Std 17.274 +trainer/Q1 Predictions Max -1.32283 +trainer/Q1 Predictions Min -87.1693 +trainer/Q2 Predictions Mean -73.4615 +trainer/Q2 Predictions Std 17.2587 +trainer/Q2 Predictions Max -1.39243 +trainer/Q2 Predictions Min -87.2793 +trainer/Q Targets Mean -73.7289 +trainer/Q Targets Std 17.2073 +trainer/Q Targets Max -0.422246 +trainer/Q Targets Min -86.6197 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00201466 +trainer/policy/mean Std 0.715327 +trainer/policy/mean Max 0.999499 +trainer/policy/mean Min -0.999553 +trainer/policy/std Mean 0.419417 +trainer/policy/std Std 0.020781 +trainer/policy/std Max 0.442976 +trainer/policy/std Min 0.387749 +trainer/Advantage Weights Mean 4.84959 +trainer/Advantage Weights Std 18.971 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.007e-18 +trainer/Advantage Score Mean -0.351224 +trainer/Advantage Score Std 0.568774 +trainer/Advantage Score Max 1.53618 +trainer/Advantage Score Min -3.96536 +trainer/V1 Predictions Mean -73.4839 +trainer/V1 Predictions Std 17.3521 +trainer/V1 Predictions Max 0.657399 +trainer/V1 Predictions Min -86.7605 +trainer/VF Loss 0.0749085 +expl/num steps total 506000 +expl/num paths total 629 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.030035 +expl/Actions Std 0.822757 +expl/Actions Max 2.41101 +expl/Actions Min -2.12217 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 473585 +eval/num paths total 506 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0950177 +eval/Actions Std 0.713026 +eval/Actions Max 0.99967 +eval/Actions Min -0.999516 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.81097e-06 +time/evaluation sampling (s) 5.05688 +time/exploration sampling (s) 6.97534 +time/logging (s) 0.00829359 +time/saving (s) 0.0120024 +time/training (s) 18.9545 +time/epoch (s) 31.007 +time/total (s) 11832.5 +Epoch -495 +------------------------------ ---------------- +2022-05-15 21:20:11.057265 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -494 finished +------------------------------ ---------------- +epoch -494 +replay_buffer/size 999047 +trainer/num train calls 507000 +trainer/QF1 Loss 0.872433 +trainer/QF2 Loss 0.840048 +trainer/Policy Loss 15.6156 +trainer/Q1 Predictions Mean -72.8632 +trainer/Q1 Predictions Std 18.1342 +trainer/Q1 Predictions Max -0.14545 +trainer/Q1 Predictions Min -86.3696 +trainer/Q2 Predictions Mean -72.8999 +trainer/Q2 Predictions Std 18.0412 +trainer/Q2 Predictions Max -0.121603 +trainer/Q2 Predictions Min -86.4603 +trainer/Q Targets Mean -73.2217 +trainer/Q Targets Std 17.9849 +trainer/Q Targets Max 0.227201 +trainer/Q Targets Min -86.7931 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00532325 +trainer/policy/mean Std 0.705682 +trainer/policy/mean Max 0.999017 +trainer/policy/mean Min -0.998699 +trainer/policy/std Mean 0.419761 +trainer/policy/std Std 0.0190621 +trainer/policy/std Max 0.441652 +trainer/policy/std Min 0.391487 +trainer/Advantage Weights Mean 4.04659 +trainer/Advantage Weights Std 16.5902 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.04673e-15 +trainer/Advantage Score Mean -0.375643 +trainer/Advantage Score Std 0.564758 +trainer/Advantage Score Max 2.17563 +trainer/Advantage Score Min -3.25862 +trainer/V1 Predictions Mean -72.8643 +trainer/V1 Predictions Std 18.2644 +trainer/V1 Predictions Max 0.317835 +trainer/V1 Predictions Min -86.5681 +trainer/VF Loss 0.0769235 +expl/num steps total 507000 +expl/num paths total 631 +expl/path length Mean 500 +expl/path length Std 148 +expl/path length Max 648 +expl/path length Min 352 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0413758 +expl/Actions Std 0.846765 +expl/Actions Max 2.60268 +expl/Actions Min -2.5638 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 474585 +eval/num paths total 507 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0958087 +eval/Actions Std 0.736041 +eval/Actions Max 0.999605 +eval/Actions Min -0.999512 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.12459e-06 +time/evaluation sampling (s) 4.74508 +time/exploration sampling (s) 7.19743 +time/logging (s) 0.00810565 +time/saving (s) 0.0122236 +time/training (s) 19.0326 +time/epoch (s) 30.9954 +time/total (s) 11863.5 +Epoch -494 +------------------------------ ---------------- +2022-05-15 21:20:40.907131 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -493 finished +------------------------------ --------------- +epoch -493 +replay_buffer/size 999047 +trainer/num train calls 508000 +trainer/QF1 Loss 0.617986 +trainer/QF2 Loss 0.641087 +trainer/Policy Loss 9.36143 +trainer/Q1 Predictions Mean -74.1959 +trainer/Q1 Predictions Std 16.8605 +trainer/Q1 Predictions Max -2.86632 +trainer/Q1 Predictions Min -87.1213 +trainer/Q2 Predictions Mean -74.1419 +trainer/Q2 Predictions Std 16.9106 +trainer/Q2 Predictions Max -1.22862 +trainer/Q2 Predictions Min -87.1562 +trainer/Q Targets Mean -73.9281 +trainer/Q Targets Std 16.8423 +trainer/Q Targets Max -3.59131 +trainer/Q Targets Min -87.0981 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.027407 +trainer/policy/mean Std 0.730473 +trainer/policy/mean Max 0.999561 +trainer/policy/mean Min -0.99897 +trainer/policy/std Mean 0.418598 +trainer/policy/std Std 0.0214079 +trainer/policy/std Max 0.442015 +trainer/policy/std Min 0.385019 +trainer/Advantage Weights Mean 2.94113 +trainer/Advantage Weights Std 15.0361 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.4479e-14 +trainer/Advantage Score Mean -0.433073 +trainer/Advantage Score Std 0.512313 +trainer/Advantage Score Max 1.44941 +trainer/Advantage Score Min -3.18661 +trainer/V1 Predictions Mean -73.6931 +trainer/V1 Predictions Std 16.9264 +trainer/V1 Predictions Max -2.68197 +trainer/V1 Predictions Min -87.3548 +trainer/VF Loss 0.0628966 +expl/num steps total 508000 +expl/num paths total 633 +expl/path length Mean 500 +expl/path length Std 136 +expl/path length Max 636 +expl/path length Min 364 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0259734 +expl/Actions Std 0.822925 +expl/Actions Max 2.35365 +expl/Actions Min -2.88089 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 475347 +eval/num paths total 508 +eval/path length Mean 762 +eval/path length Std 0 +eval/path length Max 762 +eval/path length Min 762 +eval/Rewards Mean 0.00131234 +eval/Rewards Std 0.0362024 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0549294 +eval/Actions Std 0.737936 +eval/Actions Max 0.999917 +eval/Actions Min -0.999293 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.2733e-05 +time/evaluation sampling (s) 4.69584 +time/exploration sampling (s) 6.8513 +time/logging (s) 0.0113566 +time/saving (s) 0.0179224 +time/training (s) 18.2657 +time/epoch (s) 29.8421 +time/total (s) 11893.3 +Epoch -493 +------------------------------ --------------- +2022-05-15 21:21:11.072838 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -492 finished +------------------------------ ---------------- +epoch -492 +replay_buffer/size 999047 +trainer/num train calls 509000 +trainer/QF1 Loss 0.606643 +trainer/QF2 Loss 0.688268 +trainer/Policy Loss 12.706 +trainer/Q1 Predictions Mean -73.3644 +trainer/Q1 Predictions Std 17.9611 +trainer/Q1 Predictions Max -0.576586 +trainer/Q1 Predictions Min -86.826 +trainer/Q2 Predictions Mean -73.375 +trainer/Q2 Predictions Std 17.9483 +trainer/Q2 Predictions Max -0.444034 +trainer/Q2 Predictions Min -86.6156 +trainer/Q Targets Mean -73.2692 +trainer/Q Targets Std 18.0291 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4905 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0184127 +trainer/policy/mean Std 0.712934 +trainer/policy/mean Max 0.99895 +trainer/policy/mean Min -0.996157 +trainer/policy/std Mean 0.41925 +trainer/policy/std Std 0.0207293 +trainer/policy/std Max 0.441708 +trainer/policy/std Min 0.384837 +trainer/Advantage Weights Mean 2.45963 +trainer/Advantage Weights Std 11.5802 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.16093e-20 +trainer/Advantage Score Mean -0.493578 +trainer/Advantage Score Std 0.830753 +trainer/Advantage Score Max 1.1032 +trainer/Advantage Score Min -4.49008 +trainer/V1 Predictions Mean -72.9565 +trainer/V1 Predictions Std 18.3063 +trainer/V1 Predictions Max 1.54909 +trainer/V1 Predictions Min -86.3796 +trainer/VF Loss 0.103755 +expl/num steps total 509000 +expl/num paths total 635 +expl/path length Mean 500 +expl/path length Std 365 +expl/path length Max 865 +expl/path length Min 135 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0539263 +expl/Actions Std 0.830416 +expl/Actions Max 2.3265 +expl/Actions Min -2.49216 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 476347 +eval/num paths total 509 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0827526 +eval/Actions Std 0.711386 +eval/Actions Max 0.999672 +eval/Actions Min -0.999015 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.2564e-05 +time/evaluation sampling (s) 4.73978 +time/exploration sampling (s) 6.78024 +time/logging (s) 0.0108285 +time/saving (s) 0.0167473 +time/training (s) 18.6017 +time/epoch (s) 30.1493 +time/total (s) 11923.5 +Epoch -492 +------------------------------ ---------------- +2022-05-15 21:21:41.257105 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -491 finished +------------------------------ ---------------- +epoch -491 +replay_buffer/size 999047 +trainer/num train calls 510000 +trainer/QF1 Loss 0.681242 +trainer/QF2 Loss 0.683264 +trainer/Policy Loss 27.4804 +trainer/Q1 Predictions Mean -74.178 +trainer/Q1 Predictions Std 17.9431 +trainer/Q1 Predictions Max -0.779016 +trainer/Q1 Predictions Min -87.2696 +trainer/Q2 Predictions Mean -74.2545 +trainer/Q2 Predictions Std 17.9457 +trainer/Q2 Predictions Max -0.720354 +trainer/Q2 Predictions Min -87.2063 +trainer/Q Targets Mean -74.2752 +trainer/Q Targets Std 18.1081 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1626 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0102893 +trainer/policy/mean Std 0.735095 +trainer/policy/mean Max 0.999054 +trainer/policy/mean Min -0.999699 +trainer/policy/std Mean 0.418106 +trainer/policy/std Std 0.0198784 +trainer/policy/std Max 0.43856 +trainer/policy/std Min 0.386093 +trainer/Advantage Weights Mean 5.75485 +trainer/Advantage Weights Std 20.3363 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.17142e-19 +trainer/Advantage Score Mean -0.318998 +trainer/Advantage Score Std 0.657258 +trainer/Advantage Score Max 1.40875 +trainer/Advantage Score Min -4.19292 +trainer/V1 Predictions Mean -74.0548 +trainer/V1 Predictions Std 18.2888 +trainer/V1 Predictions Max 0.924886 +trainer/V1 Predictions Min -86.9333 +trainer/VF Loss 0.0803495 +expl/num steps total 510000 +expl/num paths total 636 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.27728 +expl/Actions Std 0.831605 +expl/Actions Max 2.26604 +expl/Actions Min -2.3198 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 477347 +eval/num paths total 510 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00881525 +eval/Actions Std 0.77058 +eval/Actions Max 0.99989 +eval/Actions Min -0.999119 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.4069e-05 +time/evaluation sampling (s) 4.99737 +time/exploration sampling (s) 6.38019 +time/logging (s) 0.0127652 +time/saving (s) 0.0184529 +time/training (s) 18.7654 +time/epoch (s) 30.1742 +time/total (s) 11953.7 +Epoch -491 +------------------------------ ---------------- +2022-05-15 21:22:11.629152 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -490 finished +------------------------------ ---------------- +epoch -490 +replay_buffer/size 999047 +trainer/num train calls 511000 +trainer/QF1 Loss 0.414122 +trainer/QF2 Loss 0.482523 +trainer/Policy Loss 12.076 +trainer/Q1 Predictions Mean -75.4629 +trainer/Q1 Predictions Std 14.6498 +trainer/Q1 Predictions Max -1.58232 +trainer/Q1 Predictions Min -86.7872 +trainer/Q2 Predictions Mean -75.4238 +trainer/Q2 Predictions Std 14.5858 +trainer/Q2 Predictions Max -1.84263 +trainer/Q2 Predictions Min -86.7685 +trainer/Q Targets Mean -75.379 +trainer/Q Targets Std 14.6884 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6344 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0161136 +trainer/policy/mean Std 0.7217 +trainer/policy/mean Max 0.998892 +trainer/policy/mean Min -0.99756 +trainer/policy/std Mean 0.417982 +trainer/policy/std Std 0.0208071 +trainer/policy/std Max 0.443804 +trainer/policy/std Min 0.382184 +trainer/Advantage Weights Mean 2.17456 +trainer/Advantage Weights Std 12.6865 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.72661e-16 +trainer/Advantage Score Mean -0.435983 +trainer/Advantage Score Std 0.54687 +trainer/Advantage Score Max 1.04977 +trainer/Advantage Score Min -3.62952 +trainer/V1 Predictions Mean -75.1393 +trainer/V1 Predictions Std 14.751 +trainer/V1 Predictions Max -1.0588 +trainer/V1 Predictions Min -86.4655 +trainer/VF Loss 0.058123 +expl/num steps total 511000 +expl/num paths total 638 +expl/path length Mean 500 +expl/path length Std 356 +expl/path length Max 856 +expl/path length Min 144 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0262833 +expl/Actions Std 0.836142 +expl/Actions Max 2.39179 +expl/Actions Min -2.46203 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 478347 +eval/num paths total 511 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.114108 +eval/Actions Std 0.71831 +eval/Actions Max 0.999873 +eval/Actions Min -0.999528 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.35424e-05 +time/evaluation sampling (s) 5.01193 +time/exploration sampling (s) 6.65994 +time/logging (s) 0.0117746 +time/saving (s) 0.0181492 +time/training (s) 18.6467 +time/epoch (s) 30.3485 +time/total (s) 11984 +Epoch -490 +------------------------------ ---------------- +2022-05-15 21:22:42.223184 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -489 finished +------------------------------ ---------------- +epoch -489 +replay_buffer/size 999047 +trainer/num train calls 512000 +trainer/QF1 Loss 0.725806 +trainer/QF2 Loss 0.745743 +trainer/Policy Loss 16.222 +trainer/Q1 Predictions Mean -73.2414 +trainer/Q1 Predictions Std 18.5555 +trainer/Q1 Predictions Max -0.480649 +trainer/Q1 Predictions Min -86.93 +trainer/Q2 Predictions Mean -73.1436 +trainer/Q2 Predictions Std 18.4498 +trainer/Q2 Predictions Max -0.475646 +trainer/Q2 Predictions Min -86.6584 +trainer/Q Targets Mean -73.1821 +trainer/Q Targets Std 18.7226 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4145 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0220826 +trainer/policy/mean Std 0.728971 +trainer/policy/mean Max 0.998942 +trainer/policy/mean Min -0.999053 +trainer/policy/std Mean 0.418353 +trainer/policy/std Std 0.0206168 +trainer/policy/std Max 0.444217 +trainer/policy/std Min 0.381921 +trainer/Advantage Weights Mean 3.2775 +trainer/Advantage Weights Std 16.0321 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.88408e-15 +trainer/Advantage Score Mean -0.409251 +trainer/Advantage Score Std 0.52269 +trainer/Advantage Score Max 1.26144 +trainer/Advantage Score Min -3.27665 +trainer/V1 Predictions Mean -72.9613 +trainer/V1 Predictions Std 18.7979 +trainer/V1 Predictions Max 1.16313 +trainer/V1 Predictions Min -86.9582 +trainer/VF Loss 0.0641883 +expl/num steps total 512000 +expl/num paths total 639 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.038545 +expl/Actions Std 0.845966 +expl/Actions Max 2.41055 +expl/Actions Min -2.29828 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 479347 +eval/num paths total 512 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.055999 +eval/Actions Std 0.68038 +eval/Actions Max 0.999646 +eval/Actions Min -0.999888 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.27782e-05 +time/evaluation sampling (s) 4.84368 +time/exploration sampling (s) 6.64412 +time/logging (s) 0.011218 +time/saving (s) 0.0146284 +time/training (s) 19.0693 +time/epoch (s) 30.583 +time/total (s) 12014.6 +Epoch -489 +------------------------------ ---------------- +2022-05-15 21:23:12.612927 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -488 finished +------------------------------ ---------------- +epoch -488 +replay_buffer/size 999047 +trainer/num train calls 513000 +trainer/QF1 Loss 0.648723 +trainer/QF2 Loss 0.593629 +trainer/Policy Loss 9.27375 +trainer/Q1 Predictions Mean -75.4503 +trainer/Q1 Predictions Std 16.5672 +trainer/Q1 Predictions Max -0.745062 +trainer/Q1 Predictions Min -87.112 +trainer/Q2 Predictions Mean -75.3936 +trainer/Q2 Predictions Std 16.6154 +trainer/Q2 Predictions Max -0.917289 +trainer/Q2 Predictions Min -86.9867 +trainer/Q Targets Mean -75.1107 +trainer/Q Targets Std 16.5194 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0247 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000302415 +trainer/policy/mean Std 0.724472 +trainer/policy/mean Max 0.998976 +trainer/policy/mean Min -0.998793 +trainer/policy/std Mean 0.41882 +trainer/policy/std Std 0.0207489 +trainer/policy/std Max 0.44176 +trainer/policy/std Min 0.384231 +trainer/Advantage Weights Mean 3.18809 +trainer/Advantage Weights Std 15.2768 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.11671e-12 +trainer/Advantage Score Mean -0.390195 +trainer/Advantage Score Std 0.5115 +trainer/Advantage Score Max 0.936928 +trainer/Advantage Score Min -2.75206 +trainer/V1 Predictions Mean -74.8848 +trainer/V1 Predictions Std 16.59 +trainer/V1 Predictions Max -1.42125 +trainer/V1 Predictions Min -86.786 +trainer/VF Loss 0.0543445 +expl/num steps total 513000 +expl/num paths total 640 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0979498 +expl/Actions Std 0.823631 +expl/Actions Max 2.39885 +expl/Actions Min -2.15815 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 480217 +eval/num paths total 513 +eval/path length Mean 870 +eval/path length Std 0 +eval/path length Max 870 +eval/path length Min 870 +eval/Rewards Mean 0.00114943 +eval/Rewards Std 0.0338837 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0284753 +eval/Actions Std 0.744182 +eval/Actions Max 0.999948 +eval/Actions Min -0.999633 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 9.3542e-06 +time/evaluation sampling (s) 4.68835 +time/exploration sampling (s) 6.83679 +time/logging (s) 0.010306 +time/saving (s) 0.0156777 +time/training (s) 18.8237 +time/epoch (s) 30.3748 +time/total (s) 12045 +Epoch -488 +------------------------------ ---------------- +2022-05-15 21:23:43.301212 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -487 finished +------------------------------ ---------------- +epoch -487 +replay_buffer/size 999047 +trainer/num train calls 514000 +trainer/QF1 Loss 0.868488 +trainer/QF2 Loss 0.929787 +trainer/Policy Loss 75.5171 +trainer/Q1 Predictions Mean -71.2055 +trainer/Q1 Predictions Std 20.7189 +trainer/Q1 Predictions Max -0.617662 +trainer/Q1 Predictions Min -86.8093 +trainer/Q2 Predictions Mean -71.1584 +trainer/Q2 Predictions Std 20.7791 +trainer/Q2 Predictions Max -0.232186 +trainer/Q2 Predictions Min -86.1614 +trainer/Q Targets Mean -71.58 +trainer/Q Targets Std 21.0678 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9011 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0125481 +trainer/policy/mean Std 0.722709 +trainer/policy/mean Max 0.998398 +trainer/policy/mean Min -0.998264 +trainer/policy/std Mean 0.420605 +trainer/policy/std Std 0.0207559 +trainer/policy/std Max 0.444311 +trainer/policy/std Min 0.385976 +trainer/Advantage Weights Mean 16.4325 +trainer/Advantage Weights Std 32.4235 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.86271e-18 +trainer/Advantage Score Mean -0.208748 +trainer/Advantage Score Std 0.837137 +trainer/Advantage Score Max 2.30904 +trainer/Advantage Score Min -4.03948 +trainer/V1 Predictions Mean -71.3738 +trainer/V1 Predictions Std 21.0895 +trainer/V1 Predictions Max -0.124808 +trainer/V1 Predictions Min -86.8937 +trainer/VF Loss 0.146596 +expl/num steps total 514000 +expl/num paths total 642 +expl/path length Mean 500 +expl/path length Std 285 +expl/path length Max 785 +expl/path length Min 215 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0340403 +expl/Actions Std 0.836423 +expl/Actions Max 2.67841 +expl/Actions Min -2.29701 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 481000 +eval/num paths total 514 +eval/path length Mean 783 +eval/path length Std 0 +eval/path length Max 783 +eval/path length Min 783 +eval/Rewards Mean 0.00127714 +eval/Rewards Std 0.0357143 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0504399 +eval/Actions Std 0.758829 +eval/Actions Max 0.999606 +eval/Actions Min -0.998946 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 6.38794e-06 +time/evaluation sampling (s) 5.301 +time/exploration sampling (s) 6.48731 +time/logging (s) 0.00766539 +time/saving (s) 0.011251 +time/training (s) 18.8627 +time/epoch (s) 30.6699 +time/total (s) 12075.7 +Epoch -487 +------------------------------ ---------------- +2022-05-15 21:24:14.332362 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -486 finished +------------------------------ ---------------- +epoch -486 +replay_buffer/size 999047 +trainer/num train calls 515000 +trainer/QF1 Loss 0.869537 +trainer/QF2 Loss 0.943392 +trainer/Policy Loss 16.1701 +trainer/Q1 Predictions Mean -73.8731 +trainer/Q1 Predictions Std 18.0547 +trainer/Q1 Predictions Max -3.22071 +trainer/Q1 Predictions Min -87.1293 +trainer/Q2 Predictions Mean -73.8932 +trainer/Q2 Predictions Std 18.0547 +trainer/Q2 Predictions Max -3.52961 +trainer/Q2 Predictions Min -87.3208 +trainer/Q Targets Mean -73.636 +trainer/Q Targets Std 17.9216 +trainer/Q Targets Max -3.61536 +trainer/Q Targets Min -86.7528 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0103865 +trainer/policy/mean Std 0.724904 +trainer/policy/mean Max 0.997861 +trainer/policy/mean Min -0.998661 +trainer/policy/std Mean 0.418942 +trainer/policy/std Std 0.020541 +trainer/policy/std Max 0.441508 +trainer/policy/std Min 0.382536 +trainer/Advantage Weights Mean 4.92112 +trainer/Advantage Weights Std 19.7253 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.88454e-12 +trainer/Advantage Score Mean -0.32839 +trainer/Advantage Score Std 0.494322 +trainer/Advantage Score Max 1.83379 +trainer/Advantage Score Min -2.65717 +trainer/V1 Predictions Mean -73.4938 +trainer/V1 Predictions Std 17.8638 +trainer/V1 Predictions Max -3.39251 +trainer/V1 Predictions Min -86.8698 +trainer/VF Loss 0.076464 +expl/num steps total 515000 +expl/num paths total 644 +expl/path length Mean 500 +expl/path length Std 276 +expl/path length Max 776 +expl/path length Min 224 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0335198 +expl/Actions Std 0.827572 +expl/Actions Max 2.60758 +expl/Actions Min -2.21794 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 482000 +eval/num paths total 515 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.18171 +eval/Actions Std 0.752741 +eval/Actions Max 0.999921 +eval/Actions Min -0.999395 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.3053e-05 +time/evaluation sampling (s) 4.96769 +time/exploration sampling (s) 6.86937 +time/logging (s) 0.0128955 +time/saving (s) 0.0199487 +time/training (s) 19.1566 +time/epoch (s) 31.0265 +time/total (s) 12106.7 +Epoch -486 +------------------------------ ---------------- +2022-05-15 21:24:45.295968 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -485 finished +------------------------------ ---------------- +epoch -485 +replay_buffer/size 999047 +trainer/num train calls 516000 +trainer/QF1 Loss 1.40946 +trainer/QF2 Loss 1.40193 +trainer/Policy Loss 4.68334 +trainer/Q1 Predictions Mean -72.8347 +trainer/Q1 Predictions Std 20.1296 +trainer/Q1 Predictions Max -0.511102 +trainer/Q1 Predictions Min -86.7155 +trainer/Q2 Predictions Mean -72.9106 +trainer/Q2 Predictions Std 20.0218 +trainer/Q2 Predictions Max -0.503717 +trainer/Q2 Predictions Min -87.2901 +trainer/Q Targets Mean -72.2803 +trainer/Q Targets Std 19.9344 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.1358 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0178576 +trainer/policy/mean Std 0.730155 +trainer/policy/mean Max 0.999479 +trainer/policy/mean Min -0.998456 +trainer/policy/std Mean 0.418948 +trainer/policy/std Std 0.0193586 +trainer/policy/std Max 0.441558 +trainer/policy/std Min 0.388161 +trainer/Advantage Weights Mean 0.776585 +trainer/Advantage Weights Std 6.6218 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.90242e-17 +trainer/Advantage Score Mean -0.637006 +trainer/Advantage Score Std 0.611928 +trainer/Advantage Score Max 1.56269 +trainer/Advantage Score Min -3.73686 +trainer/V1 Predictions Mean -72.0224 +trainer/V1 Predictions Std 19.9526 +trainer/V1 Predictions Max 0.125268 +trainer/V1 Predictions Min -86.353 +trainer/VF Loss 0.0868733 +expl/num steps total 516000 +expl/num paths total 645 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0515608 +expl/Actions Std 0.8175 +expl/Actions Max 2.18736 +expl/Actions Min -2.56991 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 482745 +eval/num paths total 516 +eval/path length Mean 745 +eval/path length Std 0 +eval/path length Max 745 +eval/path length Min 745 +eval/Rewards Mean 0.00134228 +eval/Rewards Std 0.0366126 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.032393 +eval/Actions Std 0.72216 +eval/Actions Max 0.99965 +eval/Actions Min -0.999582 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.11924e-06 +time/evaluation sampling (s) 5.16378 +time/exploration sampling (s) 6.22358 +time/logging (s) 0.0114103 +time/saving (s) 0.0183675 +time/training (s) 19.5279 +time/epoch (s) 30.945 +time/total (s) 12137.7 +Epoch -485 +------------------------------ ---------------- +2022-05-15 21:25:16.105662 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -484 finished +------------------------------ ---------------- +epoch -484 +replay_buffer/size 999047 +trainer/num train calls 517000 +trainer/QF1 Loss 1.22897 +trainer/QF2 Loss 1.42154 +trainer/Policy Loss 11.3396 +trainer/Q1 Predictions Mean -73.495 +trainer/Q1 Predictions Std 20.3672 +trainer/Q1 Predictions Max -0.694417 +trainer/Q1 Predictions Min -86.9562 +trainer/Q2 Predictions Mean -73.4726 +trainer/Q2 Predictions Std 20.4523 +trainer/Q2 Predictions Max -0.368198 +trainer/Q2 Predictions Min -86.7372 +trainer/Q Targets Mean -73.4435 +trainer/Q Targets Std 20.1482 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8608 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0104793 +trainer/policy/mean Std 0.72627 +trainer/policy/mean Max 0.998685 +trainer/policy/mean Min -0.99764 +trainer/policy/std Mean 0.416799 +trainer/policy/std Std 0.0205546 +trainer/policy/std Max 0.437844 +trainer/policy/std Min 0.384012 +trainer/Advantage Weights Mean 2.51786 +trainer/Advantage Weights Std 14.4009 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.51516e-12 +trainer/Advantage Score Mean -0.544357 +trainer/Advantage Score Std 0.500993 +trainer/Advantage Score Max 1.4484 +trainer/Advantage Score Min -2.61236 +trainer/V1 Predictions Mean -73.2251 +trainer/V1 Predictions Std 20.0635 +trainer/V1 Predictions Max -0.0884365 +trainer/V1 Predictions Min -86.646 +trainer/VF Loss 0.0780235 +expl/num steps total 517000 +expl/num paths total 646 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00608667 +expl/Actions Std 0.859967 +expl/Actions Max 2.59842 +expl/Actions Min -2.13695 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 483366 +eval/num paths total 517 +eval/path length Mean 621 +eval/path length Std 0 +eval/path length Max 621 +eval/path length Min 621 +eval/Rewards Mean 0.00161031 +eval/Rewards Std 0.0400963 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0205751 +eval/Actions Std 0.751884 +eval/Actions Max 0.999809 +eval/Actions Min -0.998803 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.25752e-05 +time/evaluation sampling (s) 4.63555 +time/exploration sampling (s) 7.25255 +time/logging (s) 0.0106615 +time/saving (s) 0.0163088 +time/training (s) 18.8737 +time/epoch (s) 30.7888 +time/total (s) 12168.4 +Epoch -484 +------------------------------ ---------------- +2022-05-15 21:25:46.972649 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -483 finished +------------------------------ ---------------- +epoch -483 +replay_buffer/size 999047 +trainer/num train calls 518000 +trainer/QF1 Loss 0.599329 +trainer/QF2 Loss 0.683388 +trainer/Policy Loss 24.6422 +trainer/Q1 Predictions Mean -73.4195 +trainer/Q1 Predictions Std 17.4218 +trainer/Q1 Predictions Max -0.244266 +trainer/Q1 Predictions Min -87.9151 +trainer/Q2 Predictions Mean -73.4303 +trainer/Q2 Predictions Std 17.4851 +trainer/Q2 Predictions Max -0.438674 +trainer/Q2 Predictions Min -87.9886 +trainer/Q Targets Mean -73.3135 +trainer/Q Targets Std 17.2851 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4384 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0129252 +trainer/policy/mean Std 0.726126 +trainer/policy/mean Max 0.997942 +trainer/policy/mean Min -0.999043 +trainer/policy/std Mean 0.416761 +trainer/policy/std Std 0.0203128 +trainer/policy/std Max 0.441402 +trainer/policy/std Min 0.383016 +trainer/Advantage Weights Mean 6.25409 +trainer/Advantage Weights Std 21.4879 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.97013e-14 +trainer/Advantage Score Mean -0.275391 +trainer/Advantage Score Std 0.518294 +trainer/Advantage Score Max 1.05972 +trainer/Advantage Score Min -3.08574 +trainer/V1 Predictions Mean -73.1399 +trainer/V1 Predictions Std 17.333 +trainer/V1 Predictions Max -0.271157 +trainer/V1 Predictions Min -87.3185 +trainer/VF Loss 0.0598653 +expl/num steps total 518000 +expl/num paths total 647 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00730624 +expl/Actions Std 0.80699 +expl/Actions Max 2.11711 +expl/Actions Min -2.39674 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 484366 +eval/num paths total 518 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.049839 +eval/Actions Std 0.753748 +eval/Actions Max 0.999942 +eval/Actions Min -0.999428 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.601e-06 +time/evaluation sampling (s) 5.01157 +time/exploration sampling (s) 6.40916 +time/logging (s) 0.0123615 +time/saving (s) 0.0181403 +time/training (s) 19.4012 +time/epoch (s) 30.8524 +time/total (s) 12199.3 +Epoch -483 +------------------------------ ---------------- +2022-05-15 21:26:17.420584 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -482 finished +------------------------------ ---------------- +epoch -482 +replay_buffer/size 999047 +trainer/num train calls 519000 +trainer/QF1 Loss 1.07058 +trainer/QF2 Loss 0.895763 +trainer/Policy Loss 7.75337 +trainer/Q1 Predictions Mean -72.4405 +trainer/Q1 Predictions Std 19.3721 +trainer/Q1 Predictions Max -0.544928 +trainer/Q1 Predictions Min -87.0513 +trainer/Q2 Predictions Mean -72.5243 +trainer/Q2 Predictions Std 19.2793 +trainer/Q2 Predictions Max -1.09321 +trainer/Q2 Predictions Min -86.9961 +trainer/Q Targets Mean -72.447 +trainer/Q Targets Std 19.3788 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.989 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00450104 +trainer/policy/mean Std 0.714954 +trainer/policy/mean Max 0.997856 +trainer/policy/mean Min -0.998714 +trainer/policy/std Mean 0.418005 +trainer/policy/std Std 0.0203705 +trainer/policy/std Max 0.441979 +trainer/policy/std Min 0.384455 +trainer/Advantage Weights Mean 2.76392 +trainer/Advantage Weights Std 14.4888 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.44505e-14 +trainer/Advantage Score Mean -0.390715 +trainer/Advantage Score Std 0.536575 +trainer/Advantage Score Max 1.80092 +trainer/Advantage Score Min -3.07444 +trainer/V1 Predictions Mean -72.1476 +trainer/V1 Predictions Std 19.5163 +trainer/V1 Predictions Max -0.278958 +trainer/V1 Predictions Min -86.8801 +trainer/VF Loss 0.0638628 +expl/num steps total 519000 +expl/num paths total 648 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0444969 +expl/Actions Std 0.804832 +expl/Actions Max 2.35292 +expl/Actions Min -2.14028 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 485366 +eval/num paths total 519 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0981009 +eval/Actions Std 0.701467 +eval/Actions Max 0.999899 +eval/Actions Min -0.999484 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0673e-05 +time/evaluation sampling (s) 5.20185 +time/exploration sampling (s) 6.88781 +time/logging (s) 0.0126371 +time/saving (s) 0.0190244 +time/training (s) 18.3101 +time/epoch (s) 30.4314 +time/total (s) 12229.7 +Epoch -482 +------------------------------ ---------------- +2022-05-15 21:26:49.096046 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -481 finished +------------------------------ ---------------- +epoch -481 +replay_buffer/size 999047 +trainer/num train calls 520000 +trainer/QF1 Loss 3.1686 +trainer/QF2 Loss 2.94766 +trainer/Policy Loss 7.48668 +trainer/Q1 Predictions Mean -73.4144 +trainer/Q1 Predictions Std 17.1147 +trainer/Q1 Predictions Max -0.284323 +trainer/Q1 Predictions Min -86.6252 +trainer/Q2 Predictions Mean -73.4028 +trainer/Q2 Predictions Std 17.0938 +trainer/Q2 Predictions Max -0.422737 +trainer/Q2 Predictions Min -86.3415 +trainer/Q Targets Mean -73.1625 +trainer/Q Targets Std 17.3388 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6725 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0288715 +trainer/policy/mean Std 0.734259 +trainer/policy/mean Max 0.999821 +trainer/policy/mean Min -0.999141 +trainer/policy/std Mean 0.417116 +trainer/policy/std Std 0.0195294 +trainer/policy/std Max 0.440453 +trainer/policy/std Min 0.385589 +trainer/Advantage Weights Mean 3.15223 +trainer/Advantage Weights Std 13.093 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.05488e-15 +trainer/Advantage Score Mean -0.406547 +trainer/Advantage Score Std 0.602217 +trainer/Advantage Score Max 2.26438 +trainer/Advantage Score Min -3.23355 +trainer/V1 Predictions Mean -73.0523 +trainer/V1 Predictions Std 17.3734 +trainer/V1 Predictions Max -0.381021 +trainer/V1 Predictions Min -86.522 +trainer/VF Loss 0.076338 +expl/num steps total 520000 +expl/num paths total 650 +expl/path length Mean 500 +expl/path length Std 314 +expl/path length Max 814 +expl/path length Min 186 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.00462173 +expl/Actions Std 0.841769 +expl/Actions Max 2.38298 +expl/Actions Min -2.30525 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 486366 +eval/num paths total 520 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.108989 +eval/Actions Std 0.638255 +eval/Actions Max 0.999777 +eval/Actions Min -0.99906 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.28211e-05 +time/evaluation sampling (s) 5.63044 +time/exploration sampling (s) 6.9517 +time/logging (s) 0.0094996 +time/saving (s) 0.015133 +time/training (s) 19.0473 +time/epoch (s) 31.6541 +time/total (s) 12261.4 +Epoch -481 +------------------------------ ---------------- +2022-05-15 21:27:20.485073 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -480 finished +------------------------------ ---------------- +epoch -480 +replay_buffer/size 999047 +trainer/num train calls 521000 +trainer/QF1 Loss 0.995962 +trainer/QF2 Loss 0.953161 +trainer/Policy Loss 42.2169 +trainer/Q1 Predictions Mean -71.9042 +trainer/Q1 Predictions Std 18.987 +trainer/Q1 Predictions Max -1.26518 +trainer/Q1 Predictions Min -86.9181 +trainer/Q2 Predictions Mean -71.8929 +trainer/Q2 Predictions Std 19.0115 +trainer/Q2 Predictions Max -1.28094 +trainer/Q2 Predictions Min -86.8442 +trainer/Q Targets Mean -72.1802 +trainer/Q Targets Std 19.3375 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1022 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0135689 +trainer/policy/mean Std 0.731082 +trainer/policy/mean Max 0.999313 +trainer/policy/mean Min -0.9979 +trainer/policy/std Mean 0.418729 +trainer/policy/std Std 0.01969 +trainer/policy/std Max 0.441464 +trainer/policy/std Min 0.387547 +trainer/Advantage Weights Mean 8.31654 +trainer/Advantage Weights Std 21.7209 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.06102e-20 +trainer/Advantage Score Mean -0.23953 +trainer/Advantage Score Std 0.736191 +trainer/Advantage Score Max 2.74187 +trainer/Advantage Score Min -4.59925 +trainer/V1 Predictions Mean -71.9444 +trainer/V1 Predictions Std 19.4412 +trainer/V1 Predictions Max -0.299773 +trainer/V1 Predictions Min -86.9657 +trainer/VF Loss 0.107362 +expl/num steps total 521000 +expl/num paths total 652 +expl/path length Mean 500 +expl/path length Std 314 +expl/path length Max 814 +expl/path length Min 186 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.018569 +expl/Actions Std 0.832918 +expl/Actions Max 2.22025 +expl/Actions Min -2.27515 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 487366 +eval/num paths total 521 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.119396 +eval/Actions Std 0.641102 +eval/Actions Max 0.999853 +eval/Actions Min -0.998883 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.17812e-05 +time/evaluation sampling (s) 5.21881 +time/exploration sampling (s) 6.69943 +time/logging (s) 0.00957449 +time/saving (s) 0.0144313 +time/training (s) 19.4354 +time/epoch (s) 31.3777 +time/total (s) 12292.8 +Epoch -480 +------------------------------ ---------------- +2022-05-15 21:27:50.782890 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -479 finished +------------------------------ ---------------- +epoch -479 +replay_buffer/size 999047 +trainer/num train calls 522000 +trainer/QF1 Loss 1.3017 +trainer/QF2 Loss 1.28524 +trainer/Policy Loss 27.3443 +trainer/Q1 Predictions Mean -72.7382 +trainer/Q1 Predictions Std 18.1265 +trainer/Q1 Predictions Max -0.761811 +trainer/Q1 Predictions Min -87.01 +trainer/Q2 Predictions Mean -72.7423 +trainer/Q2 Predictions Std 18.1267 +trainer/Q2 Predictions Max -0.184398 +trainer/Q2 Predictions Min -86.7306 +trainer/Q Targets Mean -73.0153 +trainer/Q Targets Std 18.2057 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8341 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0139723 +trainer/policy/mean Std 0.723088 +trainer/policy/mean Max 0.999753 +trainer/policy/mean Min -0.99929 +trainer/policy/std Mean 0.417952 +trainer/policy/std Std 0.0181387 +trainer/policy/std Max 0.439445 +trainer/policy/std Min 0.390176 +trainer/Advantage Weights Mean 6.3691 +trainer/Advantage Weights Std 19.463 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.83747e-25 +trainer/Advantage Score Mean -0.373926 +trainer/Advantage Score Std 0.76935 +trainer/Advantage Score Max 1.80893 +trainer/Advantage Score Min -5.59882 +trainer/V1 Predictions Mean -72.7664 +trainer/V1 Predictions Std 18.2501 +trainer/V1 Predictions Max -0.68344 +trainer/V1 Predictions Min -86.7444 +trainer/VF Loss 0.10844 +expl/num steps total 522000 +expl/num paths total 654 +expl/path length Mean 500 +expl/path length Std 200 +expl/path length Max 700 +expl/path length Min 300 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0355758 +expl/Actions Std 0.824112 +expl/Actions Max 2.55129 +expl/Actions Min -2.57576 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 488366 +eval/num paths total 522 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.171424 +eval/Actions Std 0.694864 +eval/Actions Max 0.999822 +eval/Actions Min -0.999815 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.9116e-06 +time/evaluation sampling (s) 4.74559 +time/exploration sampling (s) 6.34396 +time/logging (s) 0.0104657 +time/saving (s) 0.0151409 +time/training (s) 19.1696 +time/epoch (s) 30.2847 +time/total (s) 12323.1 +Epoch -479 +------------------------------ ---------------- +2022-05-15 21:28:20.991350 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -478 finished +------------------------------ ---------------- +epoch -478 +replay_buffer/size 999047 +trainer/num train calls 523000 +trainer/QF1 Loss 1.06663 +trainer/QF2 Loss 1.13035 +trainer/Policy Loss 33.4501 +trainer/Q1 Predictions Mean -73.2612 +trainer/Q1 Predictions Std 18.9526 +trainer/Q1 Predictions Max -1.98046 +trainer/Q1 Predictions Min -86.9026 +trainer/Q2 Predictions Mean -73.2115 +trainer/Q2 Predictions Std 18.9602 +trainer/Q2 Predictions Max -1.73603 +trainer/Q2 Predictions Min -86.6281 +trainer/Q Targets Mean -73.2948 +trainer/Q Targets Std 18.7614 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9013 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0118886 +trainer/policy/mean Std 0.711786 +trainer/policy/mean Max 0.999487 +trainer/policy/mean Min -0.99799 +trainer/policy/std Mean 0.418543 +trainer/policy/std Std 0.0193945 +trainer/policy/std Max 0.442504 +trainer/policy/std Min 0.389119 +trainer/Advantage Weights Mean 6.47209 +trainer/Advantage Weights Std 22.4232 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.90546e-16 +trainer/Advantage Score Mean -0.303281 +trainer/Advantage Score Std 0.59603 +trainer/Advantage Score Max 2.44873 +trainer/Advantage Score Min -3.47738 +trainer/V1 Predictions Mean -72.9316 +trainer/V1 Predictions Std 19.0076 +trainer/V1 Predictions Max -0.853884 +trainer/V1 Predictions Min -86.6992 +trainer/VF Loss 0.0951649 +expl/num steps total 523000 +expl/num paths total 655 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0238375 +expl/Actions Std 0.840007 +expl/Actions Max 2.21641 +expl/Actions Min -2.41753 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 488989 +eval/num paths total 523 +eval/path length Mean 623 +eval/path length Std 0 +eval/path length Max 623 +eval/path length Min 623 +eval/Rewards Mean 0.00160514 +eval/Rewards Std 0.040032 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0453186 +eval/Actions Std 0.712689 +eval/Actions Max 0.999657 +eval/Actions Min -0.998846 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 6.41588e-06 +time/evaluation sampling (s) 4.66272 +time/exploration sampling (s) 6.89726 +time/logging (s) 0.0105149 +time/saving (s) 0.0172698 +time/training (s) 18.6062 +time/epoch (s) 30.1939 +time/total (s) 12353.3 +Epoch -478 +------------------------------ ---------------- +2022-05-15 21:28:51.223739 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -477 finished +------------------------------ ---------------- +epoch -477 +replay_buffer/size 999047 +trainer/num train calls 524000 +trainer/QF1 Loss 0.922115 +trainer/QF2 Loss 0.861826 +trainer/Policy Loss 12.2232 +trainer/Q1 Predictions Mean -73.2169 +trainer/Q1 Predictions Std 18.9151 +trainer/Q1 Predictions Max -0.895426 +trainer/Q1 Predictions Min -87.4731 +trainer/Q2 Predictions Mean -73.1387 +trainer/Q2 Predictions Std 18.9234 +trainer/Q2 Predictions Max -0.886317 +trainer/Q2 Predictions Min -87.2617 +trainer/Q Targets Mean -72.998 +trainer/Q Targets Std 18.827 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9425 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00730062 +trainer/policy/mean Std 0.712112 +trainer/policy/mean Max 0.999004 +trainer/policy/mean Min -0.996375 +trainer/policy/std Mean 0.419049 +trainer/policy/std Std 0.019465 +trainer/policy/std Max 0.443798 +trainer/policy/std Min 0.38725 +trainer/Advantage Weights Mean 2.25994 +trainer/Advantage Weights Std 12.5918 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.58653e-12 +trainer/Advantage Score Mean -0.503444 +trainer/Advantage Score Std 0.538616 +trainer/Advantage Score Max 1.97985 +trainer/Advantage Score Min -2.63538 +trainer/V1 Predictions Mean -72.7061 +trainer/V1 Predictions Std 19.0443 +trainer/V1 Predictions Max -0.0419177 +trainer/V1 Predictions Min -86.8491 +trainer/VF Loss 0.0710305 +expl/num steps total 524000 +expl/num paths total 656 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0306304 +expl/Actions Std 0.786343 +expl/Actions Max 2.28227 +expl/Actions Min -2.31328 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 489989 +eval/num paths total 524 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.148193 +eval/Actions Std 0.796993 +eval/Actions Max 0.999955 +eval/Actions Min -0.999505 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.78867e-06 +time/evaluation sampling (s) 4.78856 +time/exploration sampling (s) 6.77951 +time/logging (s) 0.0125519 +time/saving (s) 0.0198423 +time/training (s) 18.6177 +time/epoch (s) 30.2181 +time/total (s) 12383.5 +Epoch -477 +------------------------------ ---------------- +2022-05-15 21:29:22.493917 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -476 finished +------------------------------ ---------------- +epoch -476 +replay_buffer/size 999047 +trainer/num train calls 525000 +trainer/QF1 Loss 8.56208 +trainer/QF2 Loss 8.79799 +trainer/Policy Loss 22.3621 +trainer/Q1 Predictions Mean -73.1744 +trainer/Q1 Predictions Std 18.6372 +trainer/Q1 Predictions Max -1.01317 +trainer/Q1 Predictions Min -86.2985 +trainer/Q2 Predictions Mean -73.1609 +trainer/Q2 Predictions Std 18.5799 +trainer/Q2 Predictions Max -2.22342 +trainer/Q2 Predictions Min -86.2091 +trainer/Q Targets Mean -73.386 +trainer/Q Targets Std 18.7453 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5313 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0113829 +trainer/policy/mean Std 0.717567 +trainer/policy/mean Max 0.999928 +trainer/policy/mean Min -0.996212 +trainer/policy/std Mean 0.418677 +trainer/policy/std Std 0.0197187 +trainer/policy/std Max 0.440829 +trainer/policy/std Min 0.384568 +trainer/Advantage Weights Mean 4.46764 +trainer/Advantage Weights Std 15.1493 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.07587e-11 +trainer/Advantage Score Mean -0.283894 +trainer/Advantage Score Std 0.481535 +trainer/Advantage Score Max 0.63723 +trainer/Advantage Score Min -2.45981 +trainer/V1 Predictions Mean -73.2959 +trainer/V1 Predictions Std 18.6671 +trainer/V1 Predictions Max -0.398853 +trainer/V1 Predictions Min -86.7044 +trainer/VF Loss 0.043752 +expl/num steps total 525000 +expl/num paths total 657 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.188093 +expl/Actions Std 0.816514 +expl/Actions Max 2.29679 +expl/Actions Min -2.37682 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 490989 +eval/num paths total 525 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.390525 +eval/Actions Std 0.647941 +eval/Actions Max 0.999829 +eval/Actions Min -0.997032 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.5547e-05 +time/evaluation sampling (s) 5.27363 +time/exploration sampling (s) 7.16256 +time/logging (s) 0.00918886 +time/saving (s) 0.0164538 +time/training (s) 18.7884 +time/epoch (s) 31.2502 +time/total (s) 12414.8 +Epoch -476 +------------------------------ ---------------- +2022-05-15 21:29:53.362504 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -475 finished +------------------------------ ---------------- +epoch -475 +replay_buffer/size 999047 +trainer/num train calls 526000 +trainer/QF1 Loss 1.06673 +trainer/QF2 Loss 0.94091 +trainer/Policy Loss 1.75716 +trainer/Q1 Predictions Mean -75.8503 +trainer/Q1 Predictions Std 15.2704 +trainer/Q1 Predictions Max -2.42826 +trainer/Q1 Predictions Min -87.3514 +trainer/Q2 Predictions Mean -75.8451 +trainer/Q2 Predictions Std 15.2199 +trainer/Q2 Predictions Max -2.45851 +trainer/Q2 Predictions Min -87.1754 +trainer/Q Targets Mean -75.527 +trainer/Q Targets Std 15.0962 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7425 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0239134 +trainer/policy/mean Std 0.739356 +trainer/policy/mean Max 0.999284 +trainer/policy/mean Min -0.999122 +trainer/policy/std Mean 0.417984 +trainer/policy/std Std 0.0201747 +trainer/policy/std Max 0.441441 +trainer/policy/std Min 0.386344 +trainer/Advantage Weights Mean 0.76449 +trainer/Advantage Weights Std 6.65421 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.29786e-13 +trainer/Advantage Score Mean -0.575624 +trainer/Advantage Score Std 0.493452 +trainer/Advantage Score Max 1.34576 +trainer/Advantage Score Min -2.96729 +trainer/V1 Predictions Mean -75.2526 +trainer/V1 Predictions Std 15.2593 +trainer/V1 Predictions Max -0.684959 +trainer/V1 Predictions Min -86.4839 +trainer/VF Loss 0.0642088 +expl/num steps total 526000 +expl/num paths total 659 +expl/path length Mean 500 +expl/path length Std 376 +expl/path length Max 876 +expl/path length Min 124 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0449817 +expl/Actions Std 0.834792 +expl/Actions Max 2.57158 +expl/Actions Min -2.27503 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 491989 +eval/num paths total 526 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0367938 +eval/Actions Std 0.809746 +eval/Actions Max 0.999655 +eval/Actions Min -0.998879 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.38567e-05 +time/evaluation sampling (s) 5.27605 +time/exploration sampling (s) 6.72894 +time/logging (s) 0.0113478 +time/saving (s) 0.0156113 +time/training (s) 18.8259 +time/epoch (s) 30.8579 +time/total (s) 12445.6 +Epoch -475 +------------------------------ ---------------- +2022-05-15 21:30:23.948265 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -474 finished +------------------------------ ---------------- +epoch -474 +replay_buffer/size 999047 +trainer/num train calls 527000 +trainer/QF1 Loss 1.12096 +trainer/QF2 Loss 1.16881 +trainer/Policy Loss 5.17542 +trainer/Q1 Predictions Mean -73.0026 +trainer/Q1 Predictions Std 17.8449 +trainer/Q1 Predictions Max -4.51151 +trainer/Q1 Predictions Min -87.0038 +trainer/Q2 Predictions Mean -72.9823 +trainer/Q2 Predictions Std 17.8899 +trainer/Q2 Predictions Max -4.55476 +trainer/Q2 Predictions Min -86.8218 +trainer/Q Targets Mean -73.0167 +trainer/Q Targets Std 17.6648 +trainer/Q Targets Max -5.06938 +trainer/Q Targets Min -86.8564 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00912413 +trainer/policy/mean Std 0.725534 +trainer/policy/mean Max 0.99936 +trainer/policy/mean Min -0.997191 +trainer/policy/std Mean 0.418904 +trainer/policy/std Std 0.0197064 +trainer/policy/std Max 0.442076 +trainer/policy/std Min 0.387478 +trainer/Advantage Weights Mean 1.13744 +trainer/Advantage Weights Std 8.92388 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.20279e-15 +trainer/Advantage Score Mean -0.533752 +trainer/Advantage Score Std 0.507597 +trainer/Advantage Score Max 1.19968 +trainer/Advantage Score Min -3.43541 +trainer/V1 Predictions Mean -72.6794 +trainer/V1 Predictions Std 17.8784 +trainer/V1 Predictions Max -5.0095 +trainer/V1 Predictions Min -86.7419 +trainer/VF Loss 0.0617088 +expl/num steps total 527000 +expl/num paths total 661 +expl/path length Mean 500 +expl/path length Std 14 +expl/path length Max 514 +expl/path length Min 486 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0313484 +expl/Actions Std 0.840036 +expl/Actions Max 2.3818 +expl/Actions Min -2.35792 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 492649 +eval/num paths total 527 +eval/path length Mean 660 +eval/path length Std 0 +eval/path length Max 660 +eval/path length Min 660 +eval/Rewards Mean 0.00151515 +eval/Rewards Std 0.0388954 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0376476 +eval/Actions Std 0.746434 +eval/Actions Max 0.999568 +eval/Actions Min -0.99923 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.06269e-05 +time/evaluation sampling (s) 4.66438 +time/exploration sampling (s) 6.85571 +time/logging (s) 0.00820556 +time/saving (s) 0.0113123 +time/training (s) 19.0324 +time/epoch (s) 30.572 +time/total (s) 12476.2 +Epoch -474 +------------------------------ ---------------- +2022-05-15 21:30:54.398048 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -473 finished +------------------------------ ---------------- +epoch -473 +replay_buffer/size 999047 +trainer/num train calls 528000 +trainer/QF1 Loss 0.462967 +trainer/QF2 Loss 0.424841 +trainer/Policy Loss 16.9257 +trainer/Q1 Predictions Mean -75.0722 +trainer/Q1 Predictions Std 15.5507 +trainer/Q1 Predictions Max -0.60333 +trainer/Q1 Predictions Min -87.5704 +trainer/Q2 Predictions Mean -75.0467 +trainer/Q2 Predictions Std 15.5298 +trainer/Q2 Predictions Max -0.15715 +trainer/Q2 Predictions Min -87.4614 +trainer/Q Targets Mean -75.0327 +trainer/Q Targets Std 15.3943 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2756 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00392146 +trainer/policy/mean Std 0.724816 +trainer/policy/mean Max 0.999711 +trainer/policy/mean Min -0.998811 +trainer/policy/std Mean 0.41806 +trainer/policy/std Std 0.0200963 +trainer/policy/std Max 0.439933 +trainer/policy/std Min 0.385975 +trainer/Advantage Weights Mean 4.39688 +trainer/Advantage Weights Std 17.5731 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.87114e-09 +trainer/Advantage Score Mean -0.300278 +trainer/Advantage Score Std 0.459295 +trainer/Advantage Score Max 2.65577 +trainer/Advantage Score Min -1.91399 +trainer/V1 Predictions Mean -74.7839 +trainer/V1 Predictions Std 15.4993 +trainer/V1 Predictions Max -0.967086 +trainer/V1 Predictions Min -87.197 +trainer/VF Loss 0.0625113 +expl/num steps total 528000 +expl/num paths total 663 +expl/path length Mean 500 +expl/path length Std 366 +expl/path length Max 866 +expl/path length Min 134 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0228097 +expl/Actions Std 0.830867 +expl/Actions Max 2.88813 +expl/Actions Min -2.40937 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 493649 +eval/num paths total 528 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0830971 +eval/Actions Std 0.611636 +eval/Actions Max 0.99965 +eval/Actions Min -0.999572 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.40793e-05 +time/evaluation sampling (s) 4.71282 +time/exploration sampling (s) 6.41303 +time/logging (s) 0.0108969 +time/saving (s) 0.0155794 +time/training (s) 19.2859 +time/epoch (s) 30.4383 +time/total (s) 12506.6 +Epoch -473 +------------------------------ ---------------- +2022-05-15 21:31:24.492771 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -472 finished +------------------------------ ---------------- +epoch -472 +replay_buffer/size 999047 +trainer/num train calls 529000 +trainer/QF1 Loss 1.13843 +trainer/QF2 Loss 1.13699 +trainer/Policy Loss 14.3372 +trainer/Q1 Predictions Mean -74.1549 +trainer/Q1 Predictions Std 16.7078 +trainer/Q1 Predictions Max -3.16977 +trainer/Q1 Predictions Min -87.1508 +trainer/Q2 Predictions Mean -74.1221 +trainer/Q2 Predictions Std 16.6375 +trainer/Q2 Predictions Max -3.07712 +trainer/Q2 Predictions Min -86.9253 +trainer/Q Targets Mean -73.9241 +trainer/Q Targets Std 16.638 +trainer/Q Targets Max -3.39048 +trainer/Q Targets Min -86.8351 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.036965 +trainer/policy/mean Std 0.727446 +trainer/policy/mean Max 0.999549 +trainer/policy/mean Min -0.997248 +trainer/policy/std Mean 0.416956 +trainer/policy/std Std 0.0208766 +trainer/policy/std Max 0.440679 +trainer/policy/std Min 0.382682 +trainer/Advantage Weights Mean 3.29712 +trainer/Advantage Weights Std 15.2109 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.40714e-15 +trainer/Advantage Score Mean -0.478002 +trainer/Advantage Score Std 0.508818 +trainer/Advantage Score Max 1.79809 +trainer/Advantage Score Min -3.41972 +trainer/V1 Predictions Mean -73.7609 +trainer/V1 Predictions Std 16.791 +trainer/V1 Predictions Max -2.25688 +trainer/V1 Predictions Min -86.203 +trainer/VF Loss 0.0659519 +expl/num steps total 529000 +expl/num paths total 664 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00307064 +expl/Actions Std 0.81043 +expl/Actions Max 2.66888 +expl/Actions Min -2.23338 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 494649 +eval/num paths total 529 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.013119 +eval/Actions Std 0.711942 +eval/Actions Max 0.999911 +eval/Actions Min -0.999562 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.16988e-05 +time/evaluation sampling (s) 4.79267 +time/exploration sampling (s) 6.44418 +time/logging (s) 0.00913607 +time/saving (s) 0.0185283 +time/training (s) 18.8186 +time/epoch (s) 30.0831 +time/total (s) 12536.7 +Epoch -472 +------------------------------ ---------------- +2022-05-15 21:31:54.990635 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -471 finished +------------------------------ ---------------- +epoch -471 +replay_buffer/size 999047 +trainer/num train calls 530000 +trainer/QF1 Loss 0.617252 +trainer/QF2 Loss 0.508658 +trainer/Policy Loss 5.96432 +trainer/Q1 Predictions Mean -72.3036 +trainer/Q1 Predictions Std 17.9315 +trainer/Q1 Predictions Max -1.68899 +trainer/Q1 Predictions Min -87.4579 +trainer/Q2 Predictions Mean -72.2928 +trainer/Q2 Predictions Std 17.961 +trainer/Q2 Predictions Max -1.14562 +trainer/Q2 Predictions Min -87.5143 +trainer/Q Targets Mean -72.1814 +trainer/Q Targets Std 18.0334 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8214 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.026729 +trainer/policy/mean Std 0.723284 +trainer/policy/mean Max 0.999433 +trainer/policy/mean Min -0.996235 +trainer/policy/std Mean 0.41835 +trainer/policy/std Std 0.0196958 +trainer/policy/std Max 0.440612 +trainer/policy/std Min 0.388426 +trainer/Advantage Weights Mean 1.3508 +trainer/Advantage Weights Std 9.48839 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.91159e-16 +trainer/Advantage Score Mean -0.596945 +trainer/Advantage Score Std 0.571958 +trainer/Advantage Score Max 1.22554 +trainer/Advantage Score Min -3.49082 +trainer/V1 Predictions Mean -71.8242 +trainer/V1 Predictions Std 18.2891 +trainer/V1 Predictions Max -0.564776 +trainer/V1 Predictions Min -86.7912 +trainer/VF Loss 0.0754198 +expl/num steps total 530000 +expl/num paths total 666 +expl/path length Mean 500 +expl/path length Std 211 +expl/path length Max 711 +expl/path length Min 289 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0316673 +expl/Actions Std 0.835389 +expl/Actions Max 2.28916 +expl/Actions Min -2.21731 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 495137 +eval/num paths total 530 +eval/path length Mean 488 +eval/path length Std 0 +eval/path length Max 488 +eval/path length Min 488 +eval/Rewards Mean 0.00204918 +eval/Rewards Std 0.0452215 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0229369 +eval/Actions Std 0.738313 +eval/Actions Max 0.999094 +eval/Actions Min -0.9989 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.31787e-05 +time/evaluation sampling (s) 4.79801 +time/exploration sampling (s) 6.33276 +time/logging (s) 0.0103725 +time/saving (s) 0.0221547 +time/training (s) 19.3246 +time/epoch (s) 30.4879 +time/total (s) 12567.2 +Epoch -471 +------------------------------ ---------------- +2022-05-15 21:32:25.349149 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -470 finished +------------------------------ ---------------- +epoch -470 +replay_buffer/size 999047 +trainer/num train calls 531000 +trainer/QF1 Loss 0.517015 +trainer/QF2 Loss 0.489832 +trainer/Policy Loss 5.38848 +trainer/Q1 Predictions Mean -75.161 +trainer/Q1 Predictions Std 16.2475 +trainer/Q1 Predictions Max -0.242373 +trainer/Q1 Predictions Min -86.5204 +trainer/Q2 Predictions Mean -75.1615 +trainer/Q2 Predictions Std 16.1564 +trainer/Q2 Predictions Max -0.405702 +trainer/Q2 Predictions Min -86.575 +trainer/Q Targets Mean -74.8798 +trainer/Q Targets Std 16.3233 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5011 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.011683 +trainer/policy/mean Std 0.732595 +trainer/policy/mean Max 0.999756 +trainer/policy/mean Min -0.998299 +trainer/policy/std Mean 0.416004 +trainer/policy/std Std 0.0191962 +trainer/policy/std Max 0.437943 +trainer/policy/std Min 0.387823 +trainer/Advantage Weights Mean 1.00205 +trainer/Advantage Weights Std 7.44899 +trainer/Advantage Weights Max 82.3778 +trainer/Advantage Weights Min 1.16894e-16 +trainer/Advantage Score Mean -0.618113 +trainer/Advantage Score Std 0.590416 +trainer/Advantage Score Max 0.441132 +trainer/Advantage Score Min -3.66853 +trainer/V1 Predictions Mean -74.6689 +trainer/V1 Predictions Std 16.4463 +trainer/V1 Predictions Max 1.28015 +trainer/V1 Predictions Min -86.4651 +trainer/VF Loss 0.0751556 +expl/num steps total 531000 +expl/num paths total 668 +expl/path length Mean 500 +expl/path length Std 235 +expl/path length Max 735 +expl/path length Min 265 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0294864 +expl/Actions Std 0.828933 +expl/Actions Max 2.52045 +expl/Actions Min -2.31882 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 496137 +eval/num paths total 531 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.212912 +eval/Actions Std 0.634409 +eval/Actions Max 0.999892 +eval/Actions Min -0.999144 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.66278e-05 +time/evaluation sampling (s) 4.3254 +time/exploration sampling (s) 7.08588 +time/logging (s) 0.00824499 +time/saving (s) 0.0114143 +time/training (s) 18.9086 +time/epoch (s) 30.3395 +time/total (s) 12597.6 +Epoch -470 +------------------------------ ---------------- +2022-05-15 21:32:55.674670 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -469 finished +------------------------------ ---------------- +epoch -469 +replay_buffer/size 999047 +trainer/num train calls 532000 +trainer/QF1 Loss 0.509794 +trainer/QF2 Loss 0.481158 +trainer/Policy Loss 13.8481 +trainer/Q1 Predictions Mean -74.1468 +trainer/Q1 Predictions Std 17.1565 +trainer/Q1 Predictions Max -0.0381036 +trainer/Q1 Predictions Min -86.7035 +trainer/Q2 Predictions Mean -74.165 +trainer/Q2 Predictions Std 17.1991 +trainer/Q2 Predictions Max 0.53489 +trainer/Q2 Predictions Min -86.7957 +trainer/Q Targets Mean -74.0376 +trainer/Q Targets Std 17.5069 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8518 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0137124 +trainer/policy/mean Std 0.721311 +trainer/policy/mean Max 0.999671 +trainer/policy/mean Min -0.999108 +trainer/policy/std Mean 0.417249 +trainer/policy/std Std 0.0202529 +trainer/policy/std Max 0.443271 +trainer/policy/std Min 0.385663 +trainer/Advantage Weights Mean 2.81783 +trainer/Advantage Weights Std 12.0374 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.33214e-14 +trainer/Advantage Score Mean -0.374313 +trainer/Advantage Score Std 0.572084 +trainer/Advantage Score Max 0.933701 +trainer/Advantage Score Min -3.05624 +trainer/V1 Predictions Mean -73.8065 +trainer/V1 Predictions Std 17.5287 +trainer/V1 Predictions Max 0.665215 +trainer/V1 Predictions Min -86.7202 +trainer/VF Loss 0.0582156 +expl/num steps total 532000 +expl/num paths total 669 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0577839 +expl/Actions Std 0.822554 +expl/Actions Max 2.51154 +expl/Actions Min -2.40929 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 497137 +eval/num paths total 532 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0535638 +eval/Actions Std 0.712279 +eval/Actions Max 0.999917 +eval/Actions Min -0.999524 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.14299e-06 +time/evaluation sampling (s) 4.55473 +time/exploration sampling (s) 7.21385 +time/logging (s) 0.0134903 +time/saving (s) 0.0185073 +time/training (s) 18.5183 +time/epoch (s) 30.3188 +time/total (s) 12627.9 +Epoch -469 +------------------------------ ---------------- +2022-05-15 21:33:25.751377 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -468 finished +------------------------------ ---------------- +epoch -468 +replay_buffer/size 999047 +trainer/num train calls 533000 +trainer/QF1 Loss 0.729279 +trainer/QF2 Loss 0.631239 +trainer/Policy Loss 23.4625 +trainer/Q1 Predictions Mean -73.1781 +trainer/Q1 Predictions Std 18.3803 +trainer/Q1 Predictions Max -1.08389 +trainer/Q1 Predictions Min -86.8359 +trainer/Q2 Predictions Mean -73.1357 +trainer/Q2 Predictions Std 18.4357 +trainer/Q2 Predictions Max -0.864507 +trainer/Q2 Predictions Min -87.0228 +trainer/Q Targets Mean -73.1342 +trainer/Q Targets Std 18.4957 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9791 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00721429 +trainer/policy/mean Std 0.717534 +trainer/policy/mean Max 0.999611 +trainer/policy/mean Min -0.999189 +trainer/policy/std Mean 0.41648 +trainer/policy/std Std 0.0197558 +trainer/policy/std Max 0.440193 +trainer/policy/std Min 0.38486 +trainer/Advantage Weights Mean 4.80884 +trainer/Advantage Weights Std 18.3567 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.7873e-09 +trainer/Advantage Score Mean -0.302277 +trainer/Advantage Score Std 0.478286 +trainer/Advantage Score Max 1.38729 +trainer/Advantage Score Min -1.88082 +trainer/V1 Predictions Mean -72.9411 +trainer/V1 Predictions Std 18.3744 +trainer/V1 Predictions Max 0.138761 +trainer/V1 Predictions Min -86.6153 +trainer/VF Loss 0.0595704 +expl/num steps total 533000 +expl/num paths total 671 +expl/path length Mean 500 +expl/path length Std 178 +expl/path length Max 678 +expl/path length Min 322 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00181591 +expl/Actions Std 0.842414 +expl/Actions Max 2.52438 +expl/Actions Min -2.49713 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 498137 +eval/num paths total 533 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0547357 +eval/Actions Std 0.749575 +eval/Actions Max 0.999189 +eval/Actions Min -0.998741 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.31452e-05 +time/evaluation sampling (s) 5.11888 +time/exploration sampling (s) 6.56359 +time/logging (s) 0.0113425 +time/saving (s) 0.0131344 +time/training (s) 18.351 +time/epoch (s) 30.058 +time/total (s) 12658 +Epoch -468 +------------------------------ ---------------- +2022-05-15 21:33:56.512189 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -467 finished +------------------------------ ---------------- +epoch -467 +replay_buffer/size 999047 +trainer/num train calls 534000 +trainer/QF1 Loss 1.02418 +trainer/QF2 Loss 1.27076 +trainer/Policy Loss 12.0636 +trainer/Q1 Predictions Mean -73.4904 +trainer/Q1 Predictions Std 16.7122 +trainer/Q1 Predictions Max -3.14324 +trainer/Q1 Predictions Min -86.7513 +trainer/Q2 Predictions Mean -73.4223 +trainer/Q2 Predictions Std 16.7321 +trainer/Q2 Predictions Max -2.07544 +trainer/Q2 Predictions Min -86.5623 +trainer/Q Targets Mean -73.7457 +trainer/Q Targets Std 16.5142 +trainer/Q Targets Max -3.71954 +trainer/Q Targets Min -86.1232 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0137042 +trainer/policy/mean Std 0.728769 +trainer/policy/mean Max 0.999494 +trainer/policy/mean Min -0.998093 +trainer/policy/std Mean 0.417612 +trainer/policy/std Std 0.0204814 +trainer/policy/std Max 0.43939 +trainer/policy/std Min 0.382795 +trainer/Advantage Weights Mean 2.89334 +trainer/Advantage Weights Std 14.1495 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.55628e-15 +trainer/Advantage Score Mean -0.39849 +trainer/Advantage Score Std 0.564688 +trainer/Advantage Score Max 1.24933 +trainer/Advantage Score Min -3.30223 +trainer/V1 Predictions Mean -73.3918 +trainer/V1 Predictions Std 16.7628 +trainer/V1 Predictions Max -1.34085 +trainer/V1 Predictions Min -86.1473 +trainer/VF Loss 0.0625706 +expl/num steps total 534000 +expl/num paths total 672 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00801778 +expl/Actions Std 0.809213 +expl/Actions Max 2.23746 +expl/Actions Min -2.48633 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 499137 +eval/num paths total 534 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.244695 +eval/Actions Std 0.692335 +eval/Actions Max 0.999688 +eval/Actions Min -0.996901 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.22809e-05 +time/evaluation sampling (s) 5.35017 +time/exploration sampling (s) 7.17358 +time/logging (s) 0.00938043 +time/saving (s) 0.0116141 +time/training (s) 18.1983 +time/epoch (s) 30.743 +time/total (s) 12688.7 +Epoch -467 +------------------------------ ---------------- +2022-05-15 21:34:27.064606 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -466 finished +------------------------------ ---------------- +epoch -466 +replay_buffer/size 999047 +trainer/num train calls 535000 +trainer/QF1 Loss 3.28949 +trainer/QF2 Loss 3.20516 +trainer/Policy Loss 12.7769 +trainer/Q1 Predictions Mean -72.2602 +trainer/Q1 Predictions Std 17.581 +trainer/Q1 Predictions Max -2.17972 +trainer/Q1 Predictions Min -86.579 +trainer/Q2 Predictions Mean -72.2443 +trainer/Q2 Predictions Std 17.6368 +trainer/Q2 Predictions Max -1.72064 +trainer/Q2 Predictions Min -86.5213 +trainer/Q Targets Mean -72.2914 +trainer/Q Targets Std 17.8828 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5602 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0168534 +trainer/policy/mean Std 0.725711 +trainer/policy/mean Max 0.997567 +trainer/policy/mean Min -0.997675 +trainer/policy/std Mean 0.418512 +trainer/policy/std Std 0.0212566 +trainer/policy/std Max 0.440287 +trainer/policy/std Min 0.383604 +trainer/Advantage Weights Mean 2.53667 +trainer/Advantage Weights Std 13.6587 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.34957e-20 +trainer/Advantage Score Mean -0.533467 +trainer/Advantage Score Std 0.578638 +trainer/Advantage Score Max 1.04242 +trainer/Advantage Score Min -4.51975 +trainer/V1 Predictions Mean -72.1415 +trainer/V1 Predictions Std 17.6437 +trainer/V1 Predictions Max -2.58857 +trainer/V1 Predictions Min -86.422 +trainer/VF Loss 0.0708602 +expl/num steps total 535000 +expl/num paths total 674 +expl/path length Mean 500 +expl/path length Std 447 +expl/path length Max 947 +expl/path length Min 53 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.030559 +expl/Actions Std 0.837277 +expl/Actions Max 2.56973 +expl/Actions Min -2.18982 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 500137 +eval/num paths total 535 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0491739 +eval/Actions Std 0.740981 +eval/Actions Max 0.999872 +eval/Actions Min -0.99851 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.70177e-06 +time/evaluation sampling (s) 5.06584 +time/exploration sampling (s) 6.78559 +time/logging (s) 0.0123477 +time/saving (s) 0.0177845 +time/training (s) 18.6601 +time/epoch (s) 30.5417 +time/total (s) 12719.3 +Epoch -466 +------------------------------ ---------------- +2022-05-15 21:34:57.394067 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -465 finished +------------------------------ ---------------- +epoch -465 +replay_buffer/size 999047 +trainer/num train calls 536000 +trainer/QF1 Loss 1.3054 +trainer/QF2 Loss 1.20311 +trainer/Policy Loss 30.2624 +trainer/Q1 Predictions Mean -72.569 +trainer/Q1 Predictions Std 18.9125 +trainer/Q1 Predictions Max -0.490514 +trainer/Q1 Predictions Min -86.8345 +trainer/Q2 Predictions Mean -72.5694 +trainer/Q2 Predictions Std 18.9323 +trainer/Q2 Predictions Max 0.839106 +trainer/Q2 Predictions Min -86.6674 +trainer/Q Targets Mean -72.9548 +trainer/Q Targets Std 18.5469 +trainer/Q Targets Max -1.20625 +trainer/Q Targets Min -86.7303 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00429258 +trainer/policy/mean Std 0.729142 +trainer/policy/mean Max 0.99959 +trainer/policy/mean Min -0.998587 +trainer/policy/std Mean 0.417427 +trainer/policy/std Std 0.0208433 +trainer/policy/std Max 0.441233 +trainer/policy/std Min 0.385037 +trainer/Advantage Weights Mean 7.75422 +trainer/Advantage Weights Std 24.6602 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.92152e-13 +trainer/Advantage Score Mean -0.232431 +trainer/Advantage Score Std 0.616561 +trainer/Advantage Score Max 4.47328 +trainer/Advantage Score Min -2.92805 +trainer/V1 Predictions Mean -72.6352 +trainer/V1 Predictions Std 18.7627 +trainer/V1 Predictions Max 0.158924 +trainer/V1 Predictions Min -86.5986 +trainer/VF Loss 0.165701 +expl/num steps total 536000 +expl/num paths total 675 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0150049 +expl/Actions Std 0.834578 +expl/Actions Max 2.27957 +expl/Actions Min -2.40078 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 501095 +eval/num paths total 536 +eval/path length Mean 958 +eval/path length Std 0 +eval/path length Max 958 +eval/path length Min 958 +eval/Rewards Mean 0.00104384 +eval/Rewards Std 0.0322917 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0416622 +eval/Actions Std 0.733898 +eval/Actions Max 0.999996 +eval/Actions Min -0.99952 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 9.00915e-06 +time/evaluation sampling (s) 5.0339 +time/exploration sampling (s) 6.22571 +time/logging (s) 0.0122087 +time/saving (s) 0.0181676 +time/training (s) 19.0223 +time/epoch (s) 30.3123 +time/total (s) 12749.6 +Epoch -465 +------------------------------ ---------------- +2022-05-15 21:35:28.294358 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -464 finished +------------------------------ ---------------- +epoch -464 +replay_buffer/size 999047 +trainer/num train calls 537000 +trainer/QF1 Loss 0.443736 +trainer/QF2 Loss 0.535119 +trainer/Policy Loss 24.0534 +trainer/Q1 Predictions Mean -74.6266 +trainer/Q1 Predictions Std 15.8034 +trainer/Q1 Predictions Max -1.45425 +trainer/Q1 Predictions Min -86.7623 +trainer/Q2 Predictions Mean -74.5211 +trainer/Q2 Predictions Std 15.7559 +trainer/Q2 Predictions Max -1.69804 +trainer/Q2 Predictions Min -86.9023 +trainer/Q Targets Mean -74.7278 +trainer/Q Targets Std 15.8459 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8289 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00225414 +trainer/policy/mean Std 0.716664 +trainer/policy/mean Max 0.997868 +trainer/policy/mean Min -0.998756 +trainer/policy/std Mean 0.417027 +trainer/policy/std Std 0.0207533 +trainer/policy/std Max 0.439495 +trainer/policy/std Min 0.385421 +trainer/Advantage Weights Mean 5.34514 +trainer/Advantage Weights Std 18.7105 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.00887e-14 +trainer/Advantage Score Mean -0.196606 +trainer/Advantage Score Std 0.557911 +trainer/Advantage Score Max 3.57985 +trainer/Advantage Score Min -3.22274 +trainer/V1 Predictions Mean -74.56 +trainer/V1 Predictions Std 15.8047 +trainer/V1 Predictions Max -0.387961 +trainer/V1 Predictions Min -86.9323 +trainer/VF Loss 0.111992 +expl/num steps total 537000 +expl/num paths total 676 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.120771 +expl/Actions Std 0.844018 +expl/Actions Max 2.37745 +expl/Actions Min -2.36931 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 502095 +eval/num paths total 537 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0373828 +eval/Actions Std 0.763079 +eval/Actions Max 0.999882 +eval/Actions Min -0.999602 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.02494e-06 +time/evaluation sampling (s) 5.00933 +time/exploration sampling (s) 6.84948 +time/logging (s) 0.00818952 +time/saving (s) 0.0130604 +time/training (s) 18.9992 +time/epoch (s) 30.8793 +time/total (s) 12780.5 +Epoch -464 +------------------------------ ---------------- +2022-05-15 21:35:59.488494 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -463 finished +------------------------------ ---------------- +epoch -463 +replay_buffer/size 999047 +trainer/num train calls 538000 +trainer/QF1 Loss 1.27838 +trainer/QF2 Loss 1.36339 +trainer/Policy Loss 6.14187 +trainer/Q1 Predictions Mean -70.8297 +trainer/Q1 Predictions Std 20.3977 +trainer/Q1 Predictions Max -1.21976 +trainer/Q1 Predictions Min -86.6057 +trainer/Q2 Predictions Mean -70.8523 +trainer/Q2 Predictions Std 20.3652 +trainer/Q2 Predictions Max -0.532546 +trainer/Q2 Predictions Min -86.6659 +trainer/Q Targets Mean -70.6415 +trainer/Q Targets Std 20.6276 +trainer/Q Targets Max 0.310402 +trainer/Q Targets Min -87.104 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00597029 +trainer/policy/mean Std 0.717626 +trainer/policy/mean Max 0.999743 +trainer/policy/mean Min -0.999581 +trainer/policy/std Mean 0.416984 +trainer/policy/std Std 0.0209878 +trainer/policy/std Max 0.440296 +trainer/policy/std Min 0.382795 +trainer/Advantage Weights Mean 1.68837 +trainer/Advantage Weights Std 11.2652 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.17082e-16 +trainer/Advantage Score Mean -0.661278 +trainer/Advantage Score Std 0.62287 +trainer/Advantage Score Max 1.20052 +trainer/Advantage Score Min -3.56874 +trainer/V1 Predictions Mean -70.3685 +trainer/V1 Predictions Std 20.6651 +trainer/V1 Predictions Max 1.02053 +trainer/V1 Predictions Min -86.7456 +trainer/VF Loss 0.0921646 +expl/num steps total 538000 +expl/num paths total 677 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.304873 +expl/Actions Std 0.754726 +expl/Actions Max 2.45998 +expl/Actions Min -2.26774 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 503095 +eval/num paths total 538 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0755308 +eval/Actions Std 0.733528 +eval/Actions Max 0.99956 +eval/Actions Min -0.999938 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.31782e-05 +time/evaluation sampling (s) 5.11638 +time/exploration sampling (s) 6.76662 +time/logging (s) 0.00970218 +time/saving (s) 0.0135392 +time/training (s) 19.2763 +time/epoch (s) 31.1826 +time/total (s) 12811.7 +Epoch -463 +------------------------------ ---------------- +2022-05-15 21:36:29.492706 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -462 finished +------------------------------ ---------------- +epoch -462 +replay_buffer/size 999047 +trainer/num train calls 539000 +trainer/QF1 Loss 0.893298 +trainer/QF2 Loss 0.983402 +trainer/Policy Loss 18.0287 +trainer/Q1 Predictions Mean -74.8722 +trainer/Q1 Predictions Std 15.9973 +trainer/Q1 Predictions Max -0.286445 +trainer/Q1 Predictions Min -86.8357 +trainer/Q2 Predictions Mean -74.8397 +trainer/Q2 Predictions Std 15.9264 +trainer/Q2 Predictions Max -0.567475 +trainer/Q2 Predictions Min -86.7069 +trainer/Q Targets Mean -74.755 +trainer/Q Targets Std 15.5944 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8394 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0103287 +trainer/policy/mean Std 0.710637 +trainer/policy/mean Max 0.999587 +trainer/policy/mean Min -0.999465 +trainer/policy/std Mean 0.417122 +trainer/policy/std Std 0.0212981 +trainer/policy/std Max 0.44067 +trainer/policy/std Min 0.385028 +trainer/Advantage Weights Mean 4.36672 +trainer/Advantage Weights Std 17.8287 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.53897e-13 +trainer/Advantage Score Mean -0.446106 +trainer/Advantage Score Std 0.52804 +trainer/Advantage Score Max 1.28108 +trainer/Advantage Score Min -2.90018 +trainer/V1 Predictions Mean -74.405 +trainer/V1 Predictions Std 15.8957 +trainer/V1 Predictions Max -1.84589 +trainer/V1 Predictions Min -86.7179 +trainer/VF Loss 0.0623263 +expl/num steps total 539000 +expl/num paths total 679 +expl/path length Mean 500 +expl/path length Std 442 +expl/path length Max 942 +expl/path length Min 58 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0320104 +expl/Actions Std 0.838991 +expl/Actions Max 2.32845 +expl/Actions Min -2.30713 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 503456 +eval/num paths total 539 +eval/path length Mean 361 +eval/path length Std 0 +eval/path length Max 361 +eval/path length Min 361 +eval/Rewards Mean 0.00277008 +eval/Rewards Std 0.0525586 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0210248 +eval/Actions Std 0.749379 +eval/Actions Max 0.999834 +eval/Actions Min -0.998474 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.52017e-06 +time/evaluation sampling (s) 4.65332 +time/exploration sampling (s) 6.25421 +time/logging (s) 0.00961915 +time/saving (s) 0.0183053 +time/training (s) 19.0571 +time/epoch (s) 29.9925 +time/total (s) 12841.7 +Epoch -462 +------------------------------ ---------------- +2022-05-15 21:36:59.840479 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -461 finished +------------------------------ ---------------- +epoch -461 +replay_buffer/size 999047 +trainer/num train calls 540000 +trainer/QF1 Loss 1.37423 +trainer/QF2 Loss 1.61398 +trainer/Policy Loss 2.81988 +trainer/Q1 Predictions Mean -73.3317 +trainer/Q1 Predictions Std 17.6349 +trainer/Q1 Predictions Max -0.683567 +trainer/Q1 Predictions Min -86.5223 +trainer/Q2 Predictions Mean -73.2649 +trainer/Q2 Predictions Std 17.5726 +trainer/Q2 Predictions Max -0.905315 +trainer/Q2 Predictions Min -86.4657 +trainer/Q Targets Mean -73.1384 +trainer/Q Targets Std 18.1696 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6123 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0205672 +trainer/policy/mean Std 0.734199 +trainer/policy/mean Max 0.999602 +trainer/policy/mean Min -0.999479 +trainer/policy/std Mean 0.416143 +trainer/policy/std Std 0.0210555 +trainer/policy/std Max 0.442552 +trainer/policy/std Min 0.383794 +trainer/Advantage Weights Mean 1.42552 +trainer/Advantage Weights Std 9.22425 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.27572e-33 +trainer/Advantage Score Mean -0.536216 +trainer/Advantage Score Std 0.80411 +trainer/Advantage Score Max 0.801186 +trainer/Advantage Score Min -7.45324 +trainer/V1 Predictions Mean -72.9163 +trainer/V1 Predictions Std 18.1332 +trainer/V1 Predictions Max 0.122846 +trainer/V1 Predictions Min -86.4566 +trainer/VF Loss 0.0986548 +expl/num steps total 540000 +expl/num paths total 681 +expl/path length Mean 500 +expl/path length Std 478 +expl/path length Max 978 +expl/path length Min 22 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0409866 +expl/Actions Std 0.806539 +expl/Actions Max 2.45851 +expl/Actions Min -2.15049 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 504456 +eval/num paths total 540 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0746605 +eval/Actions Std 0.743877 +eval/Actions Max 0.999861 +eval/Actions Min -0.999344 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.07689e-05 +time/evaluation sampling (s) 4.68752 +time/exploration sampling (s) 7.40171 +time/logging (s) 0.0123876 +time/saving (s) 0.0160069 +time/training (s) 18.2163 +time/epoch (s) 30.3339 +time/total (s) 12872 +Epoch -461 +------------------------------ ---------------- +2022-05-15 21:37:30.959639 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -460 finished +------------------------------ ---------------- +epoch -460 +replay_buffer/size 999047 +trainer/num train calls 541000 +trainer/QF1 Loss 0.727939 +trainer/QF2 Loss 0.559493 +trainer/Policy Loss 32.305 +trainer/Q1 Predictions Mean -72.8498 +trainer/Q1 Predictions Std 19.3143 +trainer/Q1 Predictions Max -0.411951 +trainer/Q1 Predictions Min -87.0406 +trainer/Q2 Predictions Mean -72.8786 +trainer/Q2 Predictions Std 19.1815 +trainer/Q2 Predictions Max -0.513793 +trainer/Q2 Predictions Min -87.1711 +trainer/Q Targets Mean -72.8629 +trainer/Q Targets Std 19.0532 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9043 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0258162 +trainer/policy/mean Std 0.71993 +trainer/policy/mean Max 0.999051 +trainer/policy/mean Min -0.998931 +trainer/policy/std Mean 0.415697 +trainer/policy/std Std 0.0210514 +trainer/policy/std Max 0.441083 +trainer/policy/std Min 0.384409 +trainer/Advantage Weights Mean 4.9585 +trainer/Advantage Weights Std 18.975 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.48908e-15 +trainer/Advantage Score Mean -0.361088 +trainer/Advantage Score Std 0.577679 +trainer/Advantage Score Max 1.71828 +trainer/Advantage Score Min -3.36269 +trainer/V1 Predictions Mean -72.5483 +trainer/V1 Predictions Std 19.3059 +trainer/V1 Predictions Max 0.0297306 +trainer/V1 Predictions Min -86.7721 +trainer/VF Loss 0.0692564 +expl/num steps total 541000 +expl/num paths total 682 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00476604 +expl/Actions Std 0.807822 +expl/Actions Max 2.25493 +expl/Actions Min -2.60494 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 505456 +eval/num paths total 541 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0857888 +eval/Actions Std 0.638828 +eval/Actions Max 0.999914 +eval/Actions Min -0.999866 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.40212e-06 +time/evaluation sampling (s) 5.26157 +time/exploration sampling (s) 7.12037 +time/logging (s) 0.0123886 +time/saving (s) 0.0178826 +time/training (s) 18.6862 +time/epoch (s) 31.0984 +time/total (s) 12903.1 +Epoch -460 +------------------------------ ---------------- +2022-05-15 21:38:02.169909 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -459 finished +------------------------------ ---------------- +epoch -459 +replay_buffer/size 999047 +trainer/num train calls 542000 +trainer/QF1 Loss 1.64785 +trainer/QF2 Loss 1.49388 +trainer/Policy Loss 25.7857 +trainer/Q1 Predictions Mean -73.1121 +trainer/Q1 Predictions Std 19.4343 +trainer/Q1 Predictions Max -0.479171 +trainer/Q1 Predictions Min -87.2227 +trainer/Q2 Predictions Mean -73.1038 +trainer/Q2 Predictions Std 19.4318 +trainer/Q2 Predictions Max -0.459287 +trainer/Q2 Predictions Min -87.1449 +trainer/Q Targets Mean -73.2771 +trainer/Q Targets Std 19.1668 +trainer/Q Targets Max 0.659437 +trainer/Q Targets Min -86.8754 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0108014 +trainer/policy/mean Std 0.727016 +trainer/policy/mean Max 0.998898 +trainer/policy/mean Min -0.999673 +trainer/policy/std Mean 0.415599 +trainer/policy/std Std 0.0212988 +trainer/policy/std Max 0.4422 +trainer/policy/std Min 0.38331 +trainer/Advantage Weights Mean 5.82819 +trainer/Advantage Weights Std 21.7635 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.89066e-15 +trainer/Advantage Score Mean -0.339097 +trainer/Advantage Score Std 0.583643 +trainer/Advantage Score Max 2.00591 +trainer/Advantage Score Min -3.31802 +trainer/V1 Predictions Mean -72.8901 +trainer/V1 Predictions Std 19.5377 +trainer/V1 Predictions Max 1.69802 +trainer/V1 Predictions Min -86.9099 +trainer/VF Loss 0.0969179 +expl/num steps total 542000 +expl/num paths total 684 +expl/path length Mean 500 +expl/path length Std 414 +expl/path length Max 914 +expl/path length Min 86 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0703751 +expl/Actions Std 0.837594 +expl/Actions Max 2.17512 +expl/Actions Min -2.26961 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 506456 +eval/num paths total 542 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.191267 +eval/Actions Std 0.788004 +eval/Actions Max 0.999409 +eval/Actions Min -0.99898 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.31899e-05 +time/evaluation sampling (s) 4.92957 +time/exploration sampling (s) 6.81521 +time/logging (s) 0.0127441 +time/saving (s) 0.0165997 +time/training (s) 19.42 +time/epoch (s) 31.1941 +time/total (s) 12934.3 +Epoch -459 +------------------------------ ---------------- +2022-05-15 21:38:31.754085 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -458 finished +------------------------------ ---------------- +epoch -458 +replay_buffer/size 999047 +trainer/num train calls 543000 +trainer/QF1 Loss 0.530253 +trainer/QF2 Loss 0.442968 +trainer/Policy Loss 29.1021 +trainer/Q1 Predictions Mean -74.0904 +trainer/Q1 Predictions Std 16.1996 +trainer/Q1 Predictions Max -0.551425 +trainer/Q1 Predictions Min -87.4368 +trainer/Q2 Predictions Mean -73.9991 +trainer/Q2 Predictions Std 16.2525 +trainer/Q2 Predictions Max -0.0480581 +trainer/Q2 Predictions Min -87.5378 +trainer/Q Targets Mean -74.0542 +trainer/Q Targets Std 16.2464 +trainer/Q Targets Max -1.05721 +trainer/Q Targets Min -87.4794 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00518154 +trainer/policy/mean Std 0.713413 +trainer/policy/mean Max 0.999459 +trainer/policy/mean Min -0.99929 +trainer/policy/std Mean 0.415554 +trainer/policy/std Std 0.0198857 +trainer/policy/std Max 0.440349 +trainer/policy/std Min 0.383803 +trainer/Advantage Weights Mean 4.63764 +trainer/Advantage Weights Std 17.8532 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.12563e-19 +trainer/Advantage Score Mean -0.394251 +trainer/Advantage Score Std 0.703761 +trainer/Advantage Score Max 0.917255 +trainer/Advantage Score Min -4.26095 +trainer/V1 Predictions Mean -73.6888 +trainer/V1 Predictions Std 16.5309 +trainer/V1 Predictions Max 0.301468 +trainer/V1 Predictions Min -87.3548 +trainer/VF Loss 0.0801504 +expl/num steps total 543000 +expl/num paths total 685 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.119028 +expl/Actions Std 0.799228 +expl/Actions Max 2.59164 +expl/Actions Min -2.34201 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 507224 +eval/num paths total 544 +eval/path length Mean 384 +eval/path length Std 46 +eval/path length Max 430 +eval/path length Min 338 +eval/Rewards Mean 0.00260417 +eval/Rewards Std 0.0509645 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0163892 +eval/Actions Std 0.732357 +eval/Actions Max 0.99993 +eval/Actions Min -0.99963 +eval/Num Paths 2 +eval/Average Returns 1 +time/data storing (s) 4.63007e-06 +time/evaluation sampling (s) 4.88982 +time/exploration sampling (s) 6.27769 +time/logging (s) 0.0112391 +time/saving (s) 0.0187346 +time/training (s) 18.3687 +time/epoch (s) 29.5661 +time/total (s) 12963.9 +Epoch -458 +------------------------------ ---------------- +2022-05-15 21:39:02.598828 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -457 finished +------------------------------ ---------------- +epoch -457 +replay_buffer/size 999047 +trainer/num train calls 544000 +trainer/QF1 Loss 1.45687 +trainer/QF2 Loss 1.35015 +trainer/Policy Loss 43.6913 +trainer/Q1 Predictions Mean -73.5388 +trainer/Q1 Predictions Std 17.4516 +trainer/Q1 Predictions Max -0.816632 +trainer/Q1 Predictions Min -87.2827 +trainer/Q2 Predictions Mean -73.4646 +trainer/Q2 Predictions Std 17.5634 +trainer/Q2 Predictions Max -0.51524 +trainer/Q2 Predictions Min -87.4724 +trainer/Q Targets Mean -73.5754 +trainer/Q Targets Std 17.7576 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3003 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.026032 +trainer/policy/mean Std 0.73003 +trainer/policy/mean Max 0.999926 +trainer/policy/mean Min -0.998995 +trainer/policy/std Mean 0.415451 +trainer/policy/std Std 0.0211419 +trainer/policy/std Max 0.440546 +trainer/policy/std Min 0.385337 +trainer/Advantage Weights Mean 9.15852 +trainer/Advantage Weights Std 23.4097 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.44178e-22 +trainer/Advantage Score Mean -0.186433 +trainer/Advantage Score Std 0.687739 +trainer/Advantage Score Max 1.38833 +trainer/Advantage Score Min -4.97641 +trainer/V1 Predictions Mean -73.392 +trainer/V1 Predictions Std 17.9297 +trainer/V1 Predictions Max -1.47209 +trainer/V1 Predictions Min -87.2158 +trainer/VF Loss 0.088721 +expl/num steps total 544000 +expl/num paths total 687 +expl/path length Mean 500 +expl/path length Std 230 +expl/path length Max 730 +expl/path length Min 270 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0302656 +expl/Actions Std 0.837691 +expl/Actions Max 2.18988 +expl/Actions Min -2.17327 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 508224 +eval/num paths total 545 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0148166 +eval/Actions Std 0.717288 +eval/Actions Max 0.99998 +eval/Actions Min -0.999311 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.3215e-05 +time/evaluation sampling (s) 5.0135 +time/exploration sampling (s) 6.85431 +time/logging (s) 0.0091144 +time/saving (s) 0.0122364 +time/training (s) 18.9365 +time/epoch (s) 30.8256 +time/total (s) 12994.7 +Epoch -457 +------------------------------ ---------------- +2022-05-15 21:39:33.133693 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -456 finished +------------------------------ ---------------- +epoch -456 +replay_buffer/size 999047 +trainer/num train calls 545000 +trainer/QF1 Loss 0.982604 +trainer/QF2 Loss 0.921563 +trainer/Policy Loss 35.6594 +trainer/Q1 Predictions Mean -72.7716 +trainer/Q1 Predictions Std 18.0166 +trainer/Q1 Predictions Max -1.46767 +trainer/Q1 Predictions Min -87.5946 +trainer/Q2 Predictions Mean -72.7973 +trainer/Q2 Predictions Std 17.9267 +trainer/Q2 Predictions Max -2.60508 +trainer/Q2 Predictions Min -87.9495 +trainer/Q Targets Mean -73.0046 +trainer/Q Targets Std 17.8303 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5743 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00472232 +trainer/policy/mean Std 0.72718 +trainer/policy/mean Max 0.99924 +trainer/policy/mean Min -0.998322 +trainer/policy/std Mean 0.41591 +trainer/policy/std Std 0.0208541 +trainer/policy/std Max 0.440933 +trainer/policy/std Min 0.383825 +trainer/Advantage Weights Mean 8.43923 +trainer/Advantage Weights Std 23.535 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.24849e-15 +trainer/Advantage Score Mean -0.20706 +trainer/Advantage Score Std 0.579535 +trainer/Advantage Score Max 2.08541 +trainer/Advantage Score Min -3.2558 +trainer/V1 Predictions Mean -72.8149 +trainer/V1 Predictions Std 17.8573 +trainer/V1 Predictions Max -1.98679 +trainer/V1 Predictions Min -87.3089 +trainer/VF Loss 0.0908147 +expl/num steps total 545000 +expl/num paths total 688 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0297786 +expl/Actions Std 0.848971 +expl/Actions Max 2.39364 +expl/Actions Min -2.15085 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 508853 +eval/num paths total 546 +eval/path length Mean 629 +eval/path length Std 0 +eval/path length Max 629 +eval/path length Min 629 +eval/Rewards Mean 0.00158983 +eval/Rewards Std 0.0398409 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0478291 +eval/Actions Std 0.751534 +eval/Actions Max 0.999929 +eval/Actions Min -0.999624 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.58986e-05 +time/evaluation sampling (s) 4.97147 +time/exploration sampling (s) 6.29265 +time/logging (s) 0.00791644 +time/saving (s) 0.0144689 +time/training (s) 19.233 +time/epoch (s) 30.5195 +time/total (s) 13025.2 +Epoch -456 +------------------------------ ---------------- +2022-05-15 21:40:04.030502 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -455 finished +------------------------------ ---------------- +epoch -455 +replay_buffer/size 999047 +trainer/num train calls 546000 +trainer/QF1 Loss 2.21482 +trainer/QF2 Loss 2.00913 +trainer/Policy Loss 5.70657 +trainer/Q1 Predictions Mean -75.4667 +trainer/Q1 Predictions Std 16.5468 +trainer/Q1 Predictions Max -1.51504 +trainer/Q1 Predictions Min -87.2208 +trainer/Q2 Predictions Mean -75.4342 +trainer/Q2 Predictions Std 16.5768 +trainer/Q2 Predictions Max -2.06475 +trainer/Q2 Predictions Min -87.2229 +trainer/Q Targets Mean -75.0024 +trainer/Q Targets Std 16.9158 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9724 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0096149 +trainer/policy/mean Std 0.72587 +trainer/policy/mean Max 0.999105 +trainer/policy/mean Min -0.999079 +trainer/policy/std Mean 0.417173 +trainer/policy/std Std 0.0205918 +trainer/policy/std Max 0.443241 +trainer/policy/std Min 0.387049 +trainer/Advantage Weights Mean 1.12468 +trainer/Advantage Weights Std 7.73032 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.26652e-20 +trainer/Advantage Score Mean -0.53166 +trainer/Advantage Score Std 0.595443 +trainer/Advantage Score Max 0.503876 +trainer/Advantage Score Min -4.42165 +trainer/V1 Predictions Mean -74.8444 +trainer/V1 Predictions Std 16.7657 +trainer/V1 Predictions Max 0.0610828 +trainer/V1 Predictions Min -87.0256 +trainer/VF Loss 0.0665014 +expl/num steps total 546000 +expl/num paths total 689 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.189152 +expl/Actions Std 0.833562 +expl/Actions Max 2.48499 +expl/Actions Min -2.23905 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 509853 +eval/num paths total 547 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0207752 +eval/Actions Std 0.789199 +eval/Actions Max 0.999853 +eval/Actions Min -0.999239 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.28848e-05 +time/evaluation sampling (s) 4.93941 +time/exploration sampling (s) 6.85185 +time/logging (s) 0.00761691 +time/saving (s) 0.0143827 +time/training (s) 19.0718 +time/epoch (s) 30.8851 +time/total (s) 13056.1 +Epoch -455 +------------------------------ ---------------- +2022-05-15 21:40:35.357983 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -454 finished +------------------------------ ---------------- +epoch -454 +replay_buffer/size 999047 +trainer/num train calls 547000 +trainer/QF1 Loss 0.871363 +trainer/QF2 Loss 0.946182 +trainer/Policy Loss 9.61008 +trainer/Q1 Predictions Mean -74.8192 +trainer/Q1 Predictions Std 16.0974 +trainer/Q1 Predictions Max -2.24059 +trainer/Q1 Predictions Min -87.3689 +trainer/Q2 Predictions Mean -74.8289 +trainer/Q2 Predictions Std 16.0846 +trainer/Q2 Predictions Max -1.14223 +trainer/Q2 Predictions Min -87.5437 +trainer/Q Targets Mean -74.3502 +trainer/Q Targets Std 16.2855 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3293 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00747849 +trainer/policy/mean Std 0.726052 +trainer/policy/mean Max 0.999715 +trainer/policy/mean Min -0.998875 +trainer/policy/std Mean 0.4166 +trainer/policy/std Std 0.0204319 +trainer/policy/std Max 0.439937 +trainer/policy/std Min 0.38672 +trainer/Advantage Weights Mean 3.63233 +trainer/Advantage Weights Std 16.4155 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.42527e-17 +trainer/Advantage Score Mean -0.403489 +trainer/Advantage Score Std 0.56622 +trainer/Advantage Score Max 0.763229 +trainer/Advantage Score Min -3.70127 +trainer/V1 Predictions Mean -74.2033 +trainer/V1 Predictions Std 16.1413 +trainer/V1 Predictions Max -0.502979 +trainer/V1 Predictions Min -87.022 +trainer/VF Loss 0.0575913 +expl/num steps total 547000 +expl/num paths total 690 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.222727 +expl/Actions Std 0.834641 +expl/Actions Max 2.36685 +expl/Actions Min -2.11073 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 510853 +eval/num paths total 548 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.142682 +eval/Actions Std 0.761587 +eval/Actions Max 0.999686 +eval/Actions Min -0.998895 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.29752e-05 +time/evaluation sampling (s) 4.66158 +time/exploration sampling (s) 7.62888 +time/logging (s) 0.0124265 +time/saving (s) 0.0171377 +time/training (s) 19.0011 +time/epoch (s) 31.3211 +time/total (s) 13087.5 +Epoch -454 +------------------------------ ---------------- +2022-05-15 21:41:05.810636 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -453 finished +------------------------------ ---------------- +epoch -453 +replay_buffer/size 999047 +trainer/num train calls 548000 +trainer/QF1 Loss 1.14224 +trainer/QF2 Loss 1.23223 +trainer/Policy Loss 12.6237 +trainer/Q1 Predictions Mean -72.7196 +trainer/Q1 Predictions Std 18.6641 +trainer/Q1 Predictions Max -2.42969 +trainer/Q1 Predictions Min -86.3993 +trainer/Q2 Predictions Mean -72.6433 +trainer/Q2 Predictions Std 18.5676 +trainer/Q2 Predictions Max -1.46164 +trainer/Q2 Predictions Min -86.5952 +trainer/Q Targets Mean -72.6676 +trainer/Q Targets Std 18.8583 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.05 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0191075 +trainer/policy/mean Std 0.718183 +trainer/policy/mean Max 0.999155 +trainer/policy/mean Min -0.996659 +trainer/policy/std Mean 0.415898 +trainer/policy/std Std 0.0203934 +trainer/policy/std Max 0.437904 +trainer/policy/std Min 0.385972 +trainer/Advantage Weights Mean 3.41883 +trainer/Advantage Weights Std 13.8453 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.42032e-12 +trainer/Advantage Score Mean -0.325867 +trainer/Advantage Score Std 0.499496 +trainer/Advantage Score Max 0.725563 +trainer/Advantage Score Min -2.59409 +trainer/V1 Predictions Mean -72.4849 +trainer/V1 Predictions Std 18.8585 +trainer/V1 Predictions Max -1.40125 +trainer/V1 Predictions Min -86.7253 +trainer/VF Loss 0.0456331 +expl/num steps total 548000 +expl/num paths total 692 +expl/path length Mean 500 +expl/path length Std 308 +expl/path length Max 808 +expl/path length Min 192 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0330969 +expl/Actions Std 0.825341 +expl/Actions Max 2.48001 +expl/Actions Min -2.23585 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 511458 +eval/num paths total 549 +eval/path length Mean 605 +eval/path length Std 0 +eval/path length Max 605 +eval/path length Min 605 +eval/Rewards Mean 0.00165289 +eval/Rewards Std 0.0406222 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0195365 +eval/Actions Std 0.728595 +eval/Actions Max 0.999419 +eval/Actions Min -0.999199 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.0808e-05 +time/evaluation sampling (s) 5.02639 +time/exploration sampling (s) 6.31275 +time/logging (s) 0.00776456 +time/saving (s) 0.0154146 +time/training (s) 19.0689 +time/epoch (s) 30.4313 +time/total (s) 13117.9 +Epoch -453 +------------------------------ ---------------- +2022-05-15 21:41:36.279516 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -452 finished +------------------------------ ---------------- +epoch -452 +replay_buffer/size 999047 +trainer/num train calls 549000 +trainer/QF1 Loss 0.783607 +trainer/QF2 Loss 0.834924 +trainer/Policy Loss 26.9232 +trainer/Q1 Predictions Mean -72.5668 +trainer/Q1 Predictions Std 18.6559 +trainer/Q1 Predictions Max -0.677423 +trainer/Q1 Predictions Min -87.1716 +trainer/Q2 Predictions Mean -72.6001 +trainer/Q2 Predictions Std 18.6559 +trainer/Q2 Predictions Max -0.740793 +trainer/Q2 Predictions Min -87.4902 +trainer/Q Targets Mean -72.5741 +trainer/Q Targets Std 18.8383 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4965 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0155517 +trainer/policy/mean Std 0.72823 +trainer/policy/mean Max 0.999672 +trainer/policy/mean Min -0.998952 +trainer/policy/std Mean 0.415131 +trainer/policy/std Std 0.0217867 +trainer/policy/std Max 0.438304 +trainer/policy/std Min 0.382675 +trainer/Advantage Weights Mean 6.88495 +trainer/Advantage Weights Std 22.5186 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.84857e-14 +trainer/Advantage Score Mean -0.234104 +trainer/Advantage Score Std 0.561975 +trainer/Advantage Score Max 1.74848 +trainer/Advantage Score Min -3.16218 +trainer/V1 Predictions Mean -72.4332 +trainer/V1 Predictions Std 18.7065 +trainer/V1 Predictions Max -0.662238 +trainer/V1 Predictions Min -87.3855 +trainer/VF Loss 0.0823234 +expl/num steps total 549000 +expl/num paths total 694 +expl/path length Mean 500 +expl/path length Std 101 +expl/path length Max 601 +expl/path length Min 399 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0268431 +expl/Actions Std 0.835475 +expl/Actions Max 2.24813 +expl/Actions Min -2.29283 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 512125 +eval/num paths total 550 +eval/path length Mean 667 +eval/path length Std 0 +eval/path length Max 667 +eval/path length Min 667 +eval/Rewards Mean 0.00149925 +eval/Rewards Std 0.0386911 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0271169 +eval/Actions Std 0.744945 +eval/Actions Max 0.999679 +eval/Actions Min -0.998659 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.2679e-05 +time/evaluation sampling (s) 4.76888 +time/exploration sampling (s) 6.8762 +time/logging (s) 0.00981562 +time/saving (s) 0.0147286 +time/training (s) 18.7918 +time/epoch (s) 30.4614 +time/total (s) 13148.4 +Epoch -452 +------------------------------ ---------------- +2022-05-15 21:42:07.521729 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -451 finished +------------------------------ ---------------- +epoch -451 +replay_buffer/size 999047 +trainer/num train calls 550000 +trainer/QF1 Loss 0.778016 +trainer/QF2 Loss 0.749994 +trainer/Policy Loss 17.3735 +trainer/Q1 Predictions Mean -72.133 +trainer/Q1 Predictions Std 19.8659 +trainer/Q1 Predictions Max -0.240737 +trainer/Q1 Predictions Min -86.7747 +trainer/Q2 Predictions Mean -72.0953 +trainer/Q2 Predictions Std 19.7767 +trainer/Q2 Predictions Max -0.432762 +trainer/Q2 Predictions Min -86.9178 +trainer/Q Targets Mean -71.7906 +trainer/Q Targets Std 19.8275 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.4956 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00570512 +trainer/policy/mean Std 0.713042 +trainer/policy/mean Max 0.999449 +trainer/policy/mean Min -0.999158 +trainer/policy/std Mean 0.416468 +trainer/policy/std Std 0.0215368 +trainer/policy/std Max 0.438878 +trainer/policy/std Min 0.384268 +trainer/Advantage Weights Mean 4.19345 +trainer/Advantage Weights Std 18.9598 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.08412e-15 +trainer/Advantage Score Mean -0.506681 +trainer/Advantage Score Std 0.577921 +trainer/Advantage Score Max 2.39165 +trainer/Advantage Score Min -3.4458 +trainer/V1 Predictions Mean -71.5155 +trainer/V1 Predictions Std 20.0054 +trainer/V1 Predictions Max 0.673188 +trainer/V1 Predictions Min -86.4636 +trainer/VF Loss 0.0891647 +expl/num steps total 550000 +expl/num paths total 696 +expl/path length Mean 500 +expl/path length Std 362 +expl/path length Max 862 +expl/path length Min 138 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0306428 +expl/Actions Std 0.832516 +expl/Actions Max 2.31037 +expl/Actions Min -2.53535 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 513125 +eval/num paths total 551 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.212335 +eval/Actions Std 0.727717 +eval/Actions Max 0.999921 +eval/Actions Min -0.99893 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.10833e-06 +time/evaluation sampling (s) 5.30729 +time/exploration sampling (s) 7.02193 +time/logging (s) 0.012872 +time/saving (s) 0.01749 +time/training (s) 18.8741 +time/epoch (s) 31.2337 +time/total (s) 13179.6 +Epoch -451 +------------------------------ ---------------- +2022-05-15 21:42:38.531000 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -450 finished +------------------------------ ---------------- +epoch -450 +replay_buffer/size 999047 +trainer/num train calls 551000 +trainer/QF1 Loss 0.72758 +trainer/QF2 Loss 0.737011 +trainer/Policy Loss 20.5712 +trainer/Q1 Predictions Mean -72.4463 +trainer/Q1 Predictions Std 18.6806 +trainer/Q1 Predictions Max -1.898 +trainer/Q1 Predictions Min -87.2873 +trainer/Q2 Predictions Mean -72.4291 +trainer/Q2 Predictions Std 18.611 +trainer/Q2 Predictions Max -1.7117 +trainer/Q2 Predictions Min -86.8912 +trainer/Q Targets Mean -72.5147 +trainer/Q Targets Std 19.0052 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1151 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0469983 +trainer/policy/mean Std 0.728642 +trainer/policy/mean Max 0.998935 +trainer/policy/mean Min -0.999477 +trainer/policy/std Mean 0.416317 +trainer/policy/std Std 0.0209249 +trainer/policy/std Max 0.441243 +trainer/policy/std Min 0.383255 +trainer/Advantage Weights Mean 5.00201 +trainer/Advantage Weights Std 17.6242 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.39621e-15 +trainer/Advantage Score Mean -0.259456 +trainer/Advantage Score Std 0.579626 +trainer/Advantage Score Max 1.56301 +trainer/Advantage Score Min -3.3058 +trainer/V1 Predictions Mean -72.2767 +trainer/V1 Predictions Std 18.9725 +trainer/V1 Predictions Max -1.97892 +trainer/V1 Predictions Min -87.0983 +trainer/VF Loss 0.0655695 +expl/num steps total 551000 +expl/num paths total 697 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.175069 +expl/Actions Std 0.821614 +expl/Actions Max 2.24143 +expl/Actions Min -2.18421 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 514125 +eval/num paths total 552 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0296722 +eval/Actions Std 0.743862 +eval/Actions Max 0.999756 +eval/Actions Min -0.999712 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.60892e-06 +time/evaluation sampling (s) 5.15013 +time/exploration sampling (s) 6.70995 +time/logging (s) 0.00879231 +time/saving (s) 0.0297345 +time/training (s) 19.0893 +time/epoch (s) 30.9879 +time/total (s) 13210.6 +Epoch -450 +------------------------------ ---------------- +2022-05-15 21:43:08.240341 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -449 finished +------------------------------ ---------------- +epoch -449 +replay_buffer/size 999047 +trainer/num train calls 552000 +trainer/QF1 Loss 0.79318 +trainer/QF2 Loss 0.866118 +trainer/Policy Loss 34.8075 +trainer/Q1 Predictions Mean -73.23 +trainer/Q1 Predictions Std 19.3074 +trainer/Q1 Predictions Max -0.240832 +trainer/Q1 Predictions Min -86.9801 +trainer/Q2 Predictions Mean -73.1702 +trainer/Q2 Predictions Std 19.3547 +trainer/Q2 Predictions Max -0.457279 +trainer/Q2 Predictions Min -86.6887 +trainer/Q Targets Mean -73.404 +trainer/Q Targets Std 19.6054 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0842 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0108149 +trainer/policy/mean Std 0.731607 +trainer/policy/mean Max 0.997995 +trainer/policy/mean Min -0.997668 +trainer/policy/std Mean 0.416317 +trainer/policy/std Std 0.0209732 +trainer/policy/std Max 0.440649 +trainer/policy/std Min 0.38625 +trainer/Advantage Weights Mean 8.82851 +trainer/Advantage Weights Std 22.9087 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.79412e-16 +trainer/Advantage Score Mean -0.185048 +trainer/Advantage Score Std 0.628153 +trainer/Advantage Score Max 1.77075 +trainer/Advantage Score Min -3.55079 +trainer/V1 Predictions Mean -73.1865 +trainer/V1 Predictions Std 19.606 +trainer/V1 Predictions Max -0.21303 +trainer/V1 Predictions Min -86.9718 +trainer/VF Loss 0.0853888 +expl/num steps total 552000 +expl/num paths total 699 +expl/path length Mean 500 +expl/path length Std 219 +expl/path length Max 719 +expl/path length Min 281 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0171348 +expl/Actions Std 0.820497 +expl/Actions Max 2.32502 +expl/Actions Min -2.1727 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 515125 +eval/num paths total 553 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0534979 +eval/Actions Std 0.736089 +eval/Actions Max 0.999469 +eval/Actions Min -0.999336 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.29947e-05 +time/evaluation sampling (s) 4.81158 +time/exploration sampling (s) 6.14718 +time/logging (s) 0.00777814 +time/saving (s) 0.0127834 +time/training (s) 18.7189 +time/epoch (s) 29.6983 +time/total (s) 13240.3 +Epoch -449 +------------------------------ ---------------- +2022-05-15 21:43:38.594853 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -448 finished +------------------------------ ---------------- +epoch -448 +replay_buffer/size 999047 +trainer/num train calls 553000 +trainer/QF1 Loss 0.707064 +trainer/QF2 Loss 0.717293 +trainer/Policy Loss 13.5327 +trainer/Q1 Predictions Mean -73.7524 +trainer/Q1 Predictions Std 17.8993 +trainer/Q1 Predictions Max -0.611153 +trainer/Q1 Predictions Min -86.984 +trainer/Q2 Predictions Mean -73.7258 +trainer/Q2 Predictions Std 17.9709 +trainer/Q2 Predictions Max -0.22036 +trainer/Q2 Predictions Min -86.8408 +trainer/Q Targets Mean -73.85 +trainer/Q Targets Std 17.849 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9649 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0147949 +trainer/policy/mean Std 0.718961 +trainer/policy/mean Max 0.997214 +trainer/policy/mean Min -0.997766 +trainer/policy/std Mean 0.41616 +trainer/policy/std Std 0.0208915 +trainer/policy/std Max 0.439271 +trainer/policy/std Min 0.383661 +trainer/Advantage Weights Mean 4.24911 +trainer/Advantage Weights Std 17.5614 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.04679e-10 +trainer/Advantage Score Mean -0.31475 +trainer/Advantage Score Std 0.459883 +trainer/Advantage Score Max 1.99882 +trainer/Advantage Score Min -2.08234 +trainer/V1 Predictions Mean -73.6443 +trainer/V1 Predictions Std 17.8538 +trainer/V1 Predictions Max -0.540566 +trainer/V1 Predictions Min -86.7983 +trainer/VF Loss 0.0615689 +expl/num steps total 553000 +expl/num paths total 700 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0242096 +expl/Actions Std 0.841549 +expl/Actions Max 2.19243 +expl/Actions Min -2.33224 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 516125 +eval/num paths total 554 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.230815 +eval/Actions Std 0.725554 +eval/Actions Max 0.999449 +eval/Actions Min -0.999103 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.51998e-06 +time/evaluation sampling (s) 4.96958 +time/exploration sampling (s) 6.40134 +time/logging (s) 0.00952052 +time/saving (s) 0.0164176 +time/training (s) 18.9476 +time/epoch (s) 30.3444 +time/total (s) 13270.6 +Epoch -448 +------------------------------ ---------------- +2022-05-15 21:44:08.545682 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -447 finished +------------------------------ ---------------- +epoch -447 +replay_buffer/size 999047 +trainer/num train calls 554000 +trainer/QF1 Loss 0.781221 +trainer/QF2 Loss 0.601942 +trainer/Policy Loss 36.8853 +trainer/Q1 Predictions Mean -72.7911 +trainer/Q1 Predictions Std 18.8918 +trainer/Q1 Predictions Max -0.731937 +trainer/Q1 Predictions Min -87.4741 +trainer/Q2 Predictions Mean -72.8169 +trainer/Q2 Predictions Std 18.9033 +trainer/Q2 Predictions Max -0.623877 +trainer/Q2 Predictions Min -87.4695 +trainer/Q Targets Mean -73.1156 +trainer/Q Targets Std 18.7843 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6766 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0206376 +trainer/policy/mean Std 0.721743 +trainer/policy/mean Max 0.998988 +trainer/policy/mean Min -0.998319 +trainer/policy/std Mean 0.416455 +trainer/policy/std Std 0.0213036 +trainer/policy/std Max 0.439061 +trainer/policy/std Min 0.38642 +trainer/Advantage Weights Mean 8.02358 +trainer/Advantage Weights Std 23.8549 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.10776e-17 +trainer/Advantage Score Mean -0.273557 +trainer/Advantage Score Std 0.609571 +trainer/Advantage Score Max 1.54893 +trainer/Advantage Score Min -3.70511 +trainer/V1 Predictions Mean -72.7728 +trainer/V1 Predictions Std 18.9865 +trainer/V1 Predictions Max -0.145855 +trainer/V1 Predictions Min -87.4814 +trainer/VF Loss 0.0827109 +expl/num steps total 554000 +expl/num paths total 702 +expl/path length Mean 500 +expl/path length Std 14 +expl/path length Max 514 +expl/path length Min 486 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0204612 +expl/Actions Std 0.844058 +expl/Actions Max 2.45909 +expl/Actions Min -2.37381 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 517125 +eval/num paths total 555 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0619087 +eval/Actions Std 0.697706 +eval/Actions Max 0.999071 +eval/Actions Min -0.99932 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.14222e-05 +time/evaluation sampling (s) 4.40138 +time/exploration sampling (s) 6.54514 +time/logging (s) 0.0123524 +time/saving (s) 0.0175996 +time/training (s) 18.9647 +time/epoch (s) 29.9412 +time/total (s) 13300.6 +Epoch -447 +------------------------------ ---------------- +2022-05-15 21:44:38.627504 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -446 finished +------------------------------ ---------------- +epoch -446 +replay_buffer/size 999047 +trainer/num train calls 555000 +trainer/QF1 Loss 0.777806 +trainer/QF2 Loss 0.798157 +trainer/Policy Loss 9.80266 +trainer/Q1 Predictions Mean -74.6959 +trainer/Q1 Predictions Std 16.8928 +trainer/Q1 Predictions Max -0.6536 +trainer/Q1 Predictions Min -87.7432 +trainer/Q2 Predictions Mean -74.5918 +trainer/Q2 Predictions Std 16.9587 +trainer/Q2 Predictions Max -1.09361 +trainer/Q2 Predictions Min -87.7801 +trainer/Q Targets Mean -74.5291 +trainer/Q Targets Std 16.6066 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5016 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0176629 +trainer/policy/mean Std 0.725558 +trainer/policy/mean Max 0.998001 +trainer/policy/mean Min -0.998733 +trainer/policy/std Mean 0.417945 +trainer/policy/std Std 0.0202291 +trainer/policy/std Max 0.440987 +trainer/policy/std Min 0.384754 +trainer/Advantage Weights Mean 2.01368 +trainer/Advantage Weights Std 13.398 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.09298e-17 +trainer/Advantage Score Mean -0.553832 +trainer/Advantage Score Std 0.515893 +trainer/Advantage Score Max 0.96183 +trainer/Advantage Score Min -3.80148 +trainer/V1 Predictions Mean -74.1839 +trainer/V1 Predictions Std 16.9734 +trainer/V1 Predictions Max 0.455457 +trainer/V1 Predictions Min -87.5891 +trainer/VF Loss 0.0641204 +expl/num steps total 555000 +expl/num paths total 704 +expl/path length Mean 500 +expl/path length Std 306 +expl/path length Max 806 +expl/path length Min 194 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0183482 +expl/Actions Std 0.837399 +expl/Actions Max 2.36511 +expl/Actions Min -2.16697 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 517585 +eval/num paths total 556 +eval/path length Mean 460 +eval/path length Std 0 +eval/path length Max 460 +eval/path length Min 460 +eval/Rewards Mean 0.00217391 +eval/Rewards Std 0.0465745 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0221849 +eval/Actions Std 0.762673 +eval/Actions Max 0.999826 +eval/Actions Min -0.999524 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.32532e-05 +time/evaluation sampling (s) 4.12492 +time/exploration sampling (s) 6.75379 +time/logging (s) 0.0101923 +time/saving (s) 0.0149375 +time/training (s) 19.1596 +time/epoch (s) 30.0634 +time/total (s) 13330.7 +Epoch -446 +------------------------------ ---------------- +2022-05-15 21:45:09.725572 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -445 finished +------------------------------ ---------------- +epoch -445 +replay_buffer/size 999047 +trainer/num train calls 556000 +trainer/QF1 Loss 0.827114 +trainer/QF2 Loss 0.826553 +trainer/Policy Loss 32.6564 +trainer/Q1 Predictions Mean -72.8436 +trainer/Q1 Predictions Std 19.1839 +trainer/Q1 Predictions Max -2.89877 +trainer/Q1 Predictions Min -87.7408 +trainer/Q2 Predictions Mean -72.8148 +trainer/Q2 Predictions Std 19.1659 +trainer/Q2 Predictions Max -3.10341 +trainer/Q2 Predictions Min -87.512 +trainer/Q Targets Mean -72.9326 +trainer/Q Targets Std 19.1606 +trainer/Q Targets Max -2.80837 +trainer/Q Targets Min -87.6197 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0121514 +trainer/policy/mean Std 0.727017 +trainer/policy/mean Max 0.99779 +trainer/policy/mean Min -0.999304 +trainer/policy/std Mean 0.4163 +trainer/policy/std Std 0.0207499 +trainer/policy/std Max 0.439355 +trainer/policy/std Min 0.380586 +trainer/Advantage Weights Mean 7.47218 +trainer/Advantage Weights Std 23.9372 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.45881e-14 +trainer/Advantage Score Mean -0.2814 +trainer/Advantage Score Std 0.574532 +trainer/Advantage Score Max 1.46243 +trainer/Advantage Score Min -3.07413 +trainer/V1 Predictions Mean -72.6307 +trainer/V1 Predictions Std 19.4575 +trainer/V1 Predictions Max -2.62059 +trainer/V1 Predictions Min -87.4203 +trainer/VF Loss 0.0672163 +expl/num steps total 556000 +expl/num paths total 706 +expl/path length Mean 500 +expl/path length Std 187 +expl/path length Max 687 +expl/path length Min 313 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00800091 +expl/Actions Std 0.840308 +expl/Actions Max 2.34547 +expl/Actions Min -2.1656 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 518071 +eval/num paths total 557 +eval/path length Mean 486 +eval/path length Std 0 +eval/path length Max 486 +eval/path length Min 486 +eval/Rewards Mean 0.00205761 +eval/Rewards Std 0.0453142 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0288168 +eval/Actions Std 0.733558 +eval/Actions Max 0.999791 +eval/Actions Min -0.99924 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.03661e-05 +time/evaluation sampling (s) 4.69027 +time/exploration sampling (s) 7.00119 +time/logging (s) 0.00926231 +time/saving (s) 0.0156406 +time/training (s) 19.3675 +time/epoch (s) 31.0838 +time/total (s) 13361.8 +Epoch -445 +------------------------------ ---------------- +2022-05-15 21:45:40.158796 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -444 finished +------------------------------ ---------------- +epoch -444 +replay_buffer/size 999047 +trainer/num train calls 557000 +trainer/QF1 Loss 0.799345 +trainer/QF2 Loss 0.728436 +trainer/Policy Loss 15.8387 +trainer/Q1 Predictions Mean -71.7233 +trainer/Q1 Predictions Std 19.9892 +trainer/Q1 Predictions Max -0.622398 +trainer/Q1 Predictions Min -86.6023 +trainer/Q2 Predictions Mean -71.7563 +trainer/Q2 Predictions Std 19.9395 +trainer/Q2 Predictions Max -0.867828 +trainer/Q2 Predictions Min -86.7414 +trainer/Q Targets Mean -71.739 +trainer/Q Targets Std 19.9377 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6411 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00640546 +trainer/policy/mean Std 0.714144 +trainer/policy/mean Max 0.999494 +trainer/policy/mean Min -0.998212 +trainer/policy/std Mean 0.417573 +trainer/policy/std Std 0.0200865 +trainer/policy/std Max 0.441015 +trainer/policy/std Min 0.384273 +trainer/Advantage Weights Mean 3.3669 +trainer/Advantage Weights Std 15.3451 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.2324e-15 +trainer/Advantage Score Mean -0.416438 +trainer/Advantage Score Std 0.636621 +trainer/Advantage Score Max 2.13667 +trainer/Advantage Score Min -3.2709 +trainer/V1 Predictions Mean -71.4373 +trainer/V1 Predictions Std 20.0816 +trainer/V1 Predictions Max -1.31945 +trainer/V1 Predictions Min -86.4902 +trainer/VF Loss 0.0938009 +expl/num steps total 557000 +expl/num paths total 708 +expl/path length Mean 500 +expl/path length Std 153 +expl/path length Max 653 +expl/path length Min 347 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0539686 +expl/Actions Std 0.830332 +expl/Actions Max 2.33688 +expl/Actions Min -2.44747 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 519071 +eval/num paths total 558 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0328991 +eval/Actions Std 0.692856 +eval/Actions Max 0.999621 +eval/Actions Min -0.999464 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.32681e-05 +time/evaluation sampling (s) 4.74017 +time/exploration sampling (s) 6.73674 +time/logging (s) 0.0100401 +time/saving (s) 0.0154377 +time/training (s) 18.9171 +time/epoch (s) 30.4196 +time/total (s) 13392.2 +Epoch -444 +------------------------------ ---------------- +2022-05-15 21:46:11.352028 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -443 finished +------------------------------ ---------------- +epoch -443 +replay_buffer/size 999047 +trainer/num train calls 558000 +trainer/QF1 Loss 0.659118 +trainer/QF2 Loss 0.710293 +trainer/Policy Loss 8.84726 +trainer/Q1 Predictions Mean -74.9734 +trainer/Q1 Predictions Std 17.0065 +trainer/Q1 Predictions Max -1.23281 +trainer/Q1 Predictions Min -87.0828 +trainer/Q2 Predictions Mean -75.0076 +trainer/Q2 Predictions Std 17.0259 +trainer/Q2 Predictions Max -1.06286 +trainer/Q2 Predictions Min -87.2942 +trainer/Q Targets Mean -74.7973 +trainer/Q Targets Std 17.3344 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1031 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0201123 +trainer/policy/mean Std 0.714588 +trainer/policy/mean Max 0.999405 +trainer/policy/mean Min -0.998878 +trainer/policy/std Mean 0.417186 +trainer/policy/std Std 0.0208655 +trainer/policy/std Max 0.439027 +trainer/policy/std Min 0.38277 +trainer/Advantage Weights Mean 2.80967 +trainer/Advantage Weights Std 13.2851 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.57137e-18 +trainer/Advantage Score Mean -0.466396 +trainer/Advantage Score Std 0.730753 +trainer/Advantage Score Max 1.23582 +trainer/Advantage Score Min -3.91878 +trainer/V1 Predictions Mean -74.4956 +trainer/V1 Predictions Std 17.5457 +trainer/V1 Predictions Max -0.0256324 +trainer/V1 Predictions Min -87.04 +trainer/VF Loss 0.0910334 +expl/num steps total 558000 +expl/num paths total 710 +expl/path length Mean 500 +expl/path length Std 287 +expl/path length Max 787 +expl/path length Min 213 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0319221 +expl/Actions Std 0.814455 +expl/Actions Max 2.55441 +expl/Actions Min -2.51069 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 519940 +eval/num paths total 559 +eval/path length Mean 869 +eval/path length Std 0 +eval/path length Max 869 +eval/path length Min 869 +eval/Rewards Mean 0.00115075 +eval/Rewards Std 0.0339032 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0494298 +eval/Actions Std 0.732627 +eval/Actions Max 0.999838 +eval/Actions Min -0.999316 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.08106e-06 +time/evaluation sampling (s) 4.95154 +time/exploration sampling (s) 7.3301 +time/logging (s) 0.0118757 +time/saving (s) 0.0151963 +time/training (s) 18.8721 +time/epoch (s) 31.1808 +time/total (s) 13423.4 +Epoch -443 +------------------------------ ---------------- +2022-05-15 21:46:42.192378 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -442 finished +------------------------------ ---------------- +epoch -442 +replay_buffer/size 999047 +trainer/num train calls 559000 +trainer/QF1 Loss 1.13469 +trainer/QF2 Loss 1.17524 +trainer/Policy Loss 42.7158 +trainer/Q1 Predictions Mean -71.7504 +trainer/Q1 Predictions Std 19.3872 +trainer/Q1 Predictions Max 0.293463 +trainer/Q1 Predictions Min -87.4679 +trainer/Q2 Predictions Mean -71.7322 +trainer/Q2 Predictions Std 19.3883 +trainer/Q2 Predictions Max 0.233188 +trainer/Q2 Predictions Min -87.2562 +trainer/Q Targets Mean -72.21 +trainer/Q Targets Std 19.614 +trainer/Q Targets Max 0.562286 +trainer/Q Targets Min -87.2253 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00592088 +trainer/policy/mean Std 0.708681 +trainer/policy/mean Max 0.999478 +trainer/policy/mean Min -0.998727 +trainer/policy/std Mean 0.417856 +trainer/policy/std Std 0.0198407 +trainer/policy/std Max 0.438754 +trainer/policy/std Min 0.385748 +trainer/Advantage Weights Mean 9.17879 +trainer/Advantage Weights Std 22.1757 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.10243e-28 +trainer/Advantage Score Mean -0.251499 +trainer/Advantage Score Std 0.708274 +trainer/Advantage Score Max 1.15435 +trainer/Advantage Score Min -6.26637 +trainer/V1 Predictions Mean -71.9739 +trainer/V1 Predictions Std 19.8044 +trainer/V1 Predictions Max 1.57807 +trainer/V1 Predictions Min -87.2143 +trainer/VF Loss 0.0880259 +expl/num steps total 559000 +expl/num paths total 712 +expl/path length Mean 500 +expl/path length Std 191 +expl/path length Max 691 +expl/path length Min 309 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0428625 +expl/Actions Std 0.846778 +expl/Actions Max 2.55878 +expl/Actions Min -2.27125 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 520916 +eval/num paths total 560 +eval/path length Mean 976 +eval/path length Std 0 +eval/path length Max 976 +eval/path length Min 976 +eval/Rewards Mean 0.00102459 +eval/Rewards Std 0.0319928 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0370321 +eval/Actions Std 0.748166 +eval/Actions Max 0.999739 +eval/Actions Min -0.999426 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.07028e-05 +time/evaluation sampling (s) 5.04489 +time/exploration sampling (s) 6.69085 +time/logging (s) 0.00927293 +time/saving (s) 0.0110797 +time/training (s) 19.0665 +time/epoch (s) 30.8226 +time/total (s) 13454.2 +Epoch -442 +------------------------------ ---------------- +2022-05-15 21:47:13.375872 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -441 finished +------------------------------ ---------------- +epoch -441 +replay_buffer/size 999047 +trainer/num train calls 560000 +trainer/QF1 Loss 0.495114 +trainer/QF2 Loss 0.56953 +trainer/Policy Loss 46.1162 +trainer/Q1 Predictions Mean -74.7895 +trainer/Q1 Predictions Std 15.5618 +trainer/Q1 Predictions Max -1.64481 +trainer/Q1 Predictions Min -87.1718 +trainer/Q2 Predictions Mean -74.7245 +trainer/Q2 Predictions Std 15.476 +trainer/Q2 Predictions Max -1.53279 +trainer/Q2 Predictions Min -86.9671 +trainer/Q Targets Mean -74.785 +trainer/Q Targets Std 15.4852 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0403 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0173629 +trainer/policy/mean Std 0.72862 +trainer/policy/mean Max 0.999287 +trainer/policy/mean Min -0.99974 +trainer/policy/std Mean 0.4159 +trainer/policy/std Std 0.0205353 +trainer/policy/std Max 0.438082 +trainer/policy/std Min 0.382973 +trainer/Advantage Weights Mean 9.8222 +trainer/Advantage Weights Std 24.7811 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.68732e-12 +trainer/Advantage Score Mean -0.18035 +trainer/Advantage Score Std 0.526648 +trainer/Advantage Score Max 1.0636 +trainer/Advantage Score Min -2.71079 +trainer/V1 Predictions Mean -74.4977 +trainer/V1 Predictions Std 15.7031 +trainer/V1 Predictions Max -1.66651 +trainer/V1 Predictions Min -86.9103 +trainer/VF Loss 0.0659707 +expl/num steps total 560000 +expl/num paths total 713 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.195888 +expl/Actions Std 0.832576 +expl/Actions Max 2.41399 +expl/Actions Min -2.57386 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 521916 +eval/num paths total 561 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0760846 +eval/Actions Std 0.718181 +eval/Actions Max 0.999636 +eval/Actions Min -0.998969 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.34041e-05 +time/evaluation sampling (s) 4.93446 +time/exploration sampling (s) 6.85059 +time/logging (s) 0.0122679 +time/saving (s) 0.0181572 +time/training (s) 19.3596 +time/epoch (s) 31.1751 +time/total (s) 13485.4 +Epoch -441 +------------------------------ ---------------- +2022-05-15 21:47:43.922891 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -440 finished +------------------------------ ---------------- +epoch -440 +replay_buffer/size 999047 +trainer/num train calls 561000 +trainer/QF1 Loss 0.834906 +trainer/QF2 Loss 0.823133 +trainer/Policy Loss 14.6149 +trainer/Q1 Predictions Mean -72.4967 +trainer/Q1 Predictions Std 19.6592 +trainer/Q1 Predictions Max -0.806222 +trainer/Q1 Predictions Min -87.0275 +trainer/Q2 Predictions Mean -72.4375 +trainer/Q2 Predictions Std 19.6096 +trainer/Q2 Predictions Max 0.13576 +trainer/Q2 Predictions Min -86.9669 +trainer/Q Targets Mean -72.4015 +trainer/Q Targets Std 19.7962 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3383 +trainer/rewards Mean -0.976562 +trainer/rewards Std 0.151288 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0234375 +trainer/terminals Std 0.151288 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0121105 +trainer/policy/mean Std 0.716321 +trainer/policy/mean Max 0.999723 +trainer/policy/mean Min -0.996734 +trainer/policy/std Mean 0.41395 +trainer/policy/std Std 0.0216037 +trainer/policy/std Max 0.440722 +trainer/policy/std Min 0.378471 +trainer/Advantage Weights Mean 3.88666 +trainer/Advantage Weights Std 16.0505 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.56419e-13 +trainer/Advantage Score Mean -0.354409 +trainer/Advantage Score Std 0.554318 +trainer/Advantage Score Max 1.08106 +trainer/Advantage Score Min -2.94862 +trainer/V1 Predictions Mean -72.2337 +trainer/V1 Predictions Std 19.6789 +trainer/V1 Predictions Max -1.5431 +trainer/V1 Predictions Min -86.7993 +trainer/VF Loss 0.0614696 +expl/num steps total 561000 +expl/num paths total 714 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.153823 +expl/Actions Std 0.87424 +expl/Actions Max 2.5976 +expl/Actions Min -2.38114 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 522534 +eval/num paths total 562 +eval/path length Mean 618 +eval/path length Std 0 +eval/path length Max 618 +eval/path length Min 618 +eval/Rewards Mean 0.00161812 +eval/Rewards Std 0.0401933 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.021571 +eval/Actions Std 0.740195 +eval/Actions Max 0.999582 +eval/Actions Min -0.999488 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.11214e-05 +time/evaluation sampling (s) 4.76347 +time/exploration sampling (s) 6.39005 +time/logging (s) 0.0101075 +time/saving (s) 0.0158293 +time/training (s) 19.3487 +time/epoch (s) 30.5282 +time/total (s) 13515.9 +Epoch -440 +------------------------------ ---------------- +2022-05-15 21:48:14.291806 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -439 finished +------------------------------ ---------------- +epoch -439 +replay_buffer/size 999047 +trainer/num train calls 562000 +trainer/QF1 Loss 1.02737 +trainer/QF2 Loss 1.17127 +trainer/Policy Loss 33.3569 +trainer/Q1 Predictions Mean -74.5998 +trainer/Q1 Predictions Std 15.8288 +trainer/Q1 Predictions Max -0.286649 +trainer/Q1 Predictions Min -86.8168 +trainer/Q2 Predictions Mean -74.5723 +trainer/Q2 Predictions Std 15.8596 +trainer/Q2 Predictions Max -0.902208 +trainer/Q2 Predictions Min -86.9695 +trainer/Q Targets Mean -75.1472 +trainer/Q Targets Std 15.7612 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5046 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00010886 +trainer/policy/mean Std 0.734908 +trainer/policy/mean Max 0.997413 +trainer/policy/mean Min -0.99923 +trainer/policy/std Mean 0.416689 +trainer/policy/std Std 0.020642 +trainer/policy/std Max 0.4428 +trainer/policy/std Min 0.384224 +trainer/Advantage Weights Mean 6.91034 +trainer/Advantage Weights Std 22.1759 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.77589e-18 +trainer/Advantage Score Mean -0.299217 +trainer/Advantage Score Std 0.538875 +trainer/Advantage Score Max 1.36748 +trainer/Advantage Score Min -3.93955 +trainer/V1 Predictions Mean -74.8481 +trainer/V1 Predictions Std 16.0039 +trainer/V1 Predictions Max -0.159693 +trainer/V1 Predictions Min -87.3386 +trainer/VF Loss 0.0668892 +expl/num steps total 562000 +expl/num paths total 716 +expl/path length Mean 500 +expl/path length Std 118 +expl/path length Max 618 +expl/path length Min 382 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00943394 +expl/Actions Std 0.845313 +expl/Actions Max 2.38619 +expl/Actions Min -2.64783 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 523534 +eval/num paths total 563 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0437593 +eval/Actions Std 0.742465 +eval/Actions Max 0.999409 +eval/Actions Min -0.999423 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0625e-05 +time/evaluation sampling (s) 4.6468 +time/exploration sampling (s) 6.84001 +time/logging (s) 0.0123647 +time/saving (s) 0.0176144 +time/training (s) 18.8378 +time/epoch (s) 30.3546 +time/total (s) 13546.3 +Epoch -439 +------------------------------ ---------------- +2022-05-15 21:48:44.376459 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -438 finished +------------------------------ ---------------- +epoch -438 +replay_buffer/size 999047 +trainer/num train calls 563000 +trainer/QF1 Loss 2.83525 +trainer/QF2 Loss 2.82529 +trainer/Policy Loss 13.4569 +trainer/Q1 Predictions Mean -73.8341 +trainer/Q1 Predictions Std 18.5472 +trainer/Q1 Predictions Max -1.42515 +trainer/Q1 Predictions Min -87.6135 +trainer/Q2 Predictions Mean -73.8957 +trainer/Q2 Predictions Std 18.5011 +trainer/Q2 Predictions Max -2.21586 +trainer/Q2 Predictions Min -87.2336 +trainer/Q Targets Mean -73.5258 +trainer/Q Targets Std 18.5194 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9016 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0033885 +trainer/policy/mean Std 0.728179 +trainer/policy/mean Max 0.998755 +trainer/policy/mean Min -0.997331 +trainer/policy/std Mean 0.416903 +trainer/policy/std Std 0.0207567 +trainer/policy/std Max 0.443468 +trainer/policy/std Min 0.385263 +trainer/Advantage Weights Mean 1.56887 +trainer/Advantage Weights Std 9.25419 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.06219e-15 +trainer/Advantage Score Mean -0.51877 +trainer/Advantage Score Std 0.575999 +trainer/Advantage Score Max 0.565659 +trainer/Advantage Score Min -3.44784 +trainer/V1 Predictions Mean -73.1941 +trainer/V1 Predictions Std 18.7613 +trainer/V1 Predictions Max -1.15586 +trainer/V1 Predictions Min -86.5677 +trainer/VF Loss 0.0641608 +expl/num steps total 563000 +expl/num paths total 718 +expl/path length Mean 500 +expl/path length Std 236 +expl/path length Max 736 +expl/path length Min 264 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0200624 +expl/Actions Std 0.824736 +expl/Actions Max 2.37802 +expl/Actions Min -2.61119 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 524468 +eval/num paths total 564 +eval/path length Mean 934 +eval/path length Std 0 +eval/path length Max 934 +eval/path length Min 934 +eval/Rewards Mean 0.00107066 +eval/Rewards Std 0.0327035 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0357131 +eval/Actions Std 0.712408 +eval/Actions Max 0.999914 +eval/Actions Min -0.999157 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.27442e-05 +time/evaluation sampling (s) 4.77122 +time/exploration sampling (s) 6.46606 +time/logging (s) 0.010691 +time/saving (s) 0.019157 +time/training (s) 18.7986 +time/epoch (s) 30.0657 +time/total (s) 13576.3 +Epoch -438 +------------------------------ ---------------- +2022-05-15 21:49:15.206420 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -437 finished +------------------------------ ---------------- +epoch -437 +replay_buffer/size 999047 +trainer/num train calls 564000 +trainer/QF1 Loss 21.1093 +trainer/QF2 Loss 21.1221 +trainer/Policy Loss 23.6552 +trainer/Q1 Predictions Mean -71.5771 +trainer/Q1 Predictions Std 21.0603 +trainer/Q1 Predictions Max -0.24224 +trainer/Q1 Predictions Min -87.1927 +trainer/Q2 Predictions Mean -71.6967 +trainer/Q2 Predictions Std 21.0378 +trainer/Q2 Predictions Max -0.467291 +trainer/Q2 Predictions Min -87.2341 +trainer/Q Targets Mean -71.8133 +trainer/Q Targets Std 20.6276 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7268 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0143616 +trainer/policy/mean Std 0.733831 +trainer/policy/mean Max 0.999574 +trainer/policy/mean Min -0.998808 +trainer/policy/std Mean 0.415023 +trainer/policy/std Std 0.0215429 +trainer/policy/std Max 0.439319 +trainer/policy/std Min 0.381318 +trainer/Advantage Weights Mean 6.09675 +trainer/Advantage Weights Std 19.9489 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.7234e-13 +trainer/Advantage Score Mean -0.271638 +trainer/Advantage Score Std 0.585136 +trainer/Advantage Score Max 1.17402 +trainer/Advantage Score Min -2.93893 +trainer/V1 Predictions Mean -71.3204 +trainer/V1 Predictions Std 21.048 +trainer/V1 Predictions Max -0.580621 +trainer/V1 Predictions Min -86.7694 +trainer/VF Loss 0.0624418 +expl/num steps total 564000 +expl/num paths total 719 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.118661 +expl/Actions Std 0.829815 +expl/Actions Max 2.47744 +expl/Actions Min -2.18791 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 525096 +eval/num paths total 565 +eval/path length Mean 628 +eval/path length Std 0 +eval/path length Max 628 +eval/path length Min 628 +eval/Rewards Mean 0.00159236 +eval/Rewards Std 0.0398726 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0360103 +eval/Actions Std 0.750909 +eval/Actions Max 0.999971 +eval/Actions Min -0.9995 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.78303e-06 +time/evaluation sampling (s) 5.00187 +time/exploration sampling (s) 6.90855 +time/logging (s) 0.00991949 +time/saving (s) 0.0150777 +time/training (s) 18.8778 +time/epoch (s) 30.8132 +time/total (s) 13607.2 +Epoch -437 +------------------------------ ---------------- +2022-05-15 21:49:46.161101 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -436 finished +------------------------------ ---------------- +epoch -436 +replay_buffer/size 999047 +trainer/num train calls 565000 +trainer/QF1 Loss 1.92653 +trainer/QF2 Loss 2.17848 +trainer/Policy Loss 12.5936 +trainer/Q1 Predictions Mean -71.3088 +trainer/Q1 Predictions Std 21.1943 +trainer/Q1 Predictions Max -1.06671 +trainer/Q1 Predictions Min -88.0247 +trainer/Q2 Predictions Mean -71.397 +trainer/Q2 Predictions Std 21.2354 +trainer/Q2 Predictions Max -1.41442 +trainer/Q2 Predictions Min -88.1842 +trainer/Q Targets Mean -70.7171 +trainer/Q Targets Std 21.11 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3915 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0192545 +trainer/policy/mean Std 0.726321 +trainer/policy/mean Max 0.999378 +trainer/policy/mean Min -0.999226 +trainer/policy/std Mean 0.415275 +trainer/policy/std Std 0.0211411 +trainer/policy/std Max 0.438419 +trainer/policy/std Min 0.382956 +trainer/Advantage Weights Mean 2.94439 +trainer/Advantage Weights Std 12.9109 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.33463e-14 +trainer/Advantage Score Mean -0.484498 +trainer/Advantage Score Std 0.642889 +trainer/Advantage Score Max 1.0643 +trainer/Advantage Score Min -3.19475 +trainer/V1 Predictions Mean -70.4644 +trainer/V1 Predictions Std 21.3589 +trainer/V1 Predictions Max 0.0250577 +trainer/V1 Predictions Min -87.0446 +trainer/VF Loss 0.0759114 +expl/num steps total 565000 +expl/num paths total 721 +expl/path length Mean 500 +expl/path length Std 210 +expl/path length Max 710 +expl/path length Min 290 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00893499 +expl/Actions Std 0.827652 +expl/Actions Max 2.1945 +expl/Actions Min -2.24237 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 526096 +eval/num paths total 566 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0574469 +eval/Actions Std 0.748969 +eval/Actions Max 0.999787 +eval/Actions Min -0.999232 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.1404e-05 +time/evaluation sampling (s) 4.90367 +time/exploration sampling (s) 6.81142 +time/logging (s) 0.0124573 +time/saving (s) 0.0149667 +time/training (s) 19.2006 +time/epoch (s) 30.9431 +time/total (s) 13638.1 +Epoch -436 +------------------------------ ---------------- +2022-05-15 21:50:16.274474 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -435 finished +------------------------------ ---------------- +epoch -435 +replay_buffer/size 999047 +trainer/num train calls 566000 +trainer/QF1 Loss 0.927992 +trainer/QF2 Loss 0.924225 +trainer/Policy Loss 9.99661 +trainer/Q1 Predictions Mean -73.5195 +trainer/Q1 Predictions Std 19.5539 +trainer/Q1 Predictions Max -0.557611 +trainer/Q1 Predictions Min -87.3029 +trainer/Q2 Predictions Mean -73.521 +trainer/Q2 Predictions Std 19.576 +trainer/Q2 Predictions Max -1.29082 +trainer/Q2 Predictions Min -87.2356 +trainer/Q Targets Mean -73.2677 +trainer/Q Targets Std 19.8312 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7885 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00802888 +trainer/policy/mean Std 0.712972 +trainer/policy/mean Max 0.999479 +trainer/policy/mean Min -0.999201 +trainer/policy/std Mean 0.416222 +trainer/policy/std Std 0.0204935 +trainer/policy/std Max 0.441999 +trainer/policy/std Min 0.384553 +trainer/Advantage Weights Mean 2.08651 +trainer/Advantage Weights Std 13.0196 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.14934e-20 +trainer/Advantage Score Mean -0.587506 +trainer/Advantage Score Std 0.656232 +trainer/Advantage Score Max 1.66056 +trainer/Advantage Score Min -4.46288 +trainer/V1 Predictions Mean -72.9498 +trainer/V1 Predictions Std 19.933 +trainer/V1 Predictions Max 0.138923 +trainer/V1 Predictions Min -86.7483 +trainer/VF Loss 0.0931595 +expl/num steps total 566000 +expl/num paths total 723 +expl/path length Mean 500 +expl/path length Std 252 +expl/path length Max 752 +expl/path length Min 248 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.052053 +expl/Actions Std 0.829321 +expl/Actions Max 2.23211 +expl/Actions Min -2.17712 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 527096 +eval/num paths total 567 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.258406 +eval/Actions Std 0.715784 +eval/Actions Max 0.999786 +eval/Actions Min -0.999899 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.15428e-05 +time/evaluation sampling (s) 4.71277 +time/exploration sampling (s) 6.56647 +time/logging (s) 0.0115239 +time/saving (s) 0.0155217 +time/training (s) 18.7903 +time/epoch (s) 30.0966 +time/total (s) 13668.2 +Epoch -435 +------------------------------ ---------------- +2022-05-15 21:50:46.732301 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -434 finished +------------------------------ ---------------- +epoch -434 +replay_buffer/size 999047 +trainer/num train calls 567000 +trainer/QF1 Loss 0.92837 +trainer/QF2 Loss 0.730168 +trainer/Policy Loss 18.96 +trainer/Q1 Predictions Mean -71.0131 +trainer/Q1 Predictions Std 21.2457 +trainer/Q1 Predictions Max -0.552902 +trainer/Q1 Predictions Min -87.761 +trainer/Q2 Predictions Mean -70.9808 +trainer/Q2 Predictions Std 21.3131 +trainer/Q2 Predictions Max 0.180182 +trainer/Q2 Predictions Min -87.6342 +trainer/Q Targets Mean -71.263 +trainer/Q Targets Std 21.3218 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2486 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00270439 +trainer/policy/mean Std 0.729045 +trainer/policy/mean Max 0.999026 +trainer/policy/mean Min -0.999617 +trainer/policy/std Mean 0.416166 +trainer/policy/std Std 0.0203177 +trainer/policy/std Max 0.440138 +trainer/policy/std Min 0.381856 +trainer/Advantage Weights Mean 4.9995 +trainer/Advantage Weights Std 17.5215 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.87266e-14 +trainer/Advantage Score Mean -0.250076 +trainer/Advantage Score Std 0.548883 +trainer/Advantage Score Max 1.55317 +trainer/Advantage Score Min -3.1181 +trainer/V1 Predictions Mean -71.0154 +trainer/V1 Predictions Std 21.4095 +trainer/V1 Predictions Max -0.0161346 +trainer/V1 Predictions Min -88.2781 +trainer/VF Loss 0.0628003 +expl/num steps total 567000 +expl/num paths total 724 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0128705 +expl/Actions Std 0.827181 +expl/Actions Max 2.22267 +expl/Actions Min -2.30063 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 527598 +eval/num paths total 568 +eval/path length Mean 502 +eval/path length Std 0 +eval/path length Max 502 +eval/path length Min 502 +eval/Rewards Mean 0.00199203 +eval/Rewards Std 0.0445877 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0114135 +eval/Actions Std 0.736732 +eval/Actions Max 0.999981 +eval/Actions Min -0.998467 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 6.02193e-06 +time/evaluation sampling (s) 4.87296 +time/exploration sampling (s) 6.56929 +time/logging (s) 0.00864166 +time/saving (s) 0.0164138 +time/training (s) 18.9721 +time/epoch (s) 30.4394 +time/total (s) 13698.7 +Epoch -434 +------------------------------ ---------------- +2022-05-15 21:51:16.788870 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -433 finished +------------------------------ ---------------- +epoch -433 +replay_buffer/size 999047 +trainer/num train calls 568000 +trainer/QF1 Loss 1.01031 +trainer/QF2 Loss 1.04901 +trainer/Policy Loss 12.3327 +trainer/Q1 Predictions Mean -73.5333 +trainer/Q1 Predictions Std 17.282 +trainer/Q1 Predictions Max -2.94093 +trainer/Q1 Predictions Min -87.2563 +trainer/Q2 Predictions Mean -73.4957 +trainer/Q2 Predictions Std 17.322 +trainer/Q2 Predictions Max -3.53051 +trainer/Q2 Predictions Min -87.4396 +trainer/Q Targets Mean -73.5996 +trainer/Q Targets Std 17.2658 +trainer/Q Targets Max -3.82761 +trainer/Q Targets Min -87.3456 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0197435 +trainer/policy/mean Std 0.725787 +trainer/policy/mean Max 0.998678 +trainer/policy/mean Min -0.999797 +trainer/policy/std Mean 0.415913 +trainer/policy/std Std 0.0205255 +trainer/policy/std Max 0.441357 +trainer/policy/std Min 0.383329 +trainer/Advantage Weights Mean 3.32582 +trainer/Advantage Weights Std 15.4149 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.93492e-16 +trainer/Advantage Score Mean -0.382206 +trainer/Advantage Score Std 0.63859 +trainer/Advantage Score Max 2.31802 +trainer/Advantage Score Min -3.47701 +trainer/V1 Predictions Mean -73.3042 +trainer/V1 Predictions Std 17.5202 +trainer/V1 Predictions Max -2.23406 +trainer/V1 Predictions Min -87.0546 +trainer/VF Loss 0.0905221 +expl/num steps total 568000 +expl/num paths total 726 +expl/path length Mean 500 +expl/path length Std 321 +expl/path length Max 821 +expl/path length Min 179 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00994908 +expl/Actions Std 0.834586 +expl/Actions Max 2.3009 +expl/Actions Min -2.57048 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 528598 +eval/num paths total 569 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.25578 +eval/Actions Std 0.660185 +eval/Actions Max 0.999435 +eval/Actions Min -0.998738 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.39489e-05 +time/evaluation sampling (s) 4.5184 +time/exploration sampling (s) 6.24518 +time/logging (s) 0.012559 +time/saving (s) 0.0123941 +time/training (s) 19.2571 +time/epoch (s) 30.0457 +time/total (s) 13728.7 +Epoch -433 +------------------------------ ---------------- +2022-05-15 21:51:47.886600 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -432 finished +------------------------------ ---------------- +epoch -432 +replay_buffer/size 999047 +trainer/num train calls 569000 +trainer/QF1 Loss 0.527891 +trainer/QF2 Loss 0.62525 +trainer/Policy Loss 11.9114 +trainer/Q1 Predictions Mean -73.9958 +trainer/Q1 Predictions Std 17.6106 +trainer/Q1 Predictions Max -0.516818 +trainer/Q1 Predictions Min -87.0395 +trainer/Q2 Predictions Mean -74.007 +trainer/Q2 Predictions Std 17.6106 +trainer/Q2 Predictions Max -0.51626 +trainer/Q2 Predictions Min -87.3283 +trainer/Q Targets Mean -73.7549 +trainer/Q Targets Std 17.897 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2439 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00470801 +trainer/policy/mean Std 0.724454 +trainer/policy/mean Max 0.999133 +trainer/policy/mean Min -0.997139 +trainer/policy/std Mean 0.414993 +trainer/policy/std Std 0.0200053 +trainer/policy/std Max 0.437406 +trainer/policy/std Min 0.383794 +trainer/Advantage Weights Mean 2.10418 +trainer/Advantage Weights Std 11.1145 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.73378e-23 +trainer/Advantage Score Mean -0.350935 +trainer/Advantage Score Std 0.584938 +trainer/Advantage Score Max 1.33371 +trainer/Advantage Score Min -5.19538 +trainer/V1 Predictions Mean -73.5085 +trainer/V1 Predictions Std 17.9069 +trainer/V1 Predictions Max 0.750756 +trainer/V1 Predictions Min -87.0285 +trainer/VF Loss 0.0569234 +expl/num steps total 569000 +expl/num paths total 727 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0460372 +expl/Actions Std 0.843181 +expl/Actions Max 2.5804 +expl/Actions Min -2.20387 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 529598 +eval/num paths total 570 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.159712 +eval/Actions Std 0.771397 +eval/Actions Max 0.999901 +eval/Actions Min -0.999673 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.2869e-05 +time/evaluation sampling (s) 4.77348 +time/exploration sampling (s) 6.71551 +time/logging (s) 0.0125479 +time/saving (s) 0.0183219 +time/training (s) 19.5648 +time/epoch (s) 31.0846 +time/total (s) 13759.8 +Epoch -432 +------------------------------ ---------------- +2022-05-15 21:52:17.933187 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -431 finished +------------------------------ ---------------- +epoch -431 +replay_buffer/size 999047 +trainer/num train calls 570000 +trainer/QF1 Loss 0.632474 +trainer/QF2 Loss 0.686127 +trainer/Policy Loss 38.6663 +trainer/Q1 Predictions Mean -70.3879 +trainer/Q1 Predictions Std 21.7336 +trainer/Q1 Predictions Max -0.493379 +trainer/Q1 Predictions Min -86.6207 +trainer/Q2 Predictions Mean -70.3826 +trainer/Q2 Predictions Std 21.7733 +trainer/Q2 Predictions Max 0.491127 +trainer/Q2 Predictions Min -86.3751 +trainer/Q Targets Mean -70.5306 +trainer/Q Targets Std 21.6862 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5293 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0227923 +trainer/policy/mean Std 0.735751 +trainer/policy/mean Max 0.999177 +trainer/policy/mean Min -0.997704 +trainer/policy/std Mean 0.416312 +trainer/policy/std Std 0.0195846 +trainer/policy/std Max 0.439201 +trainer/policy/std Min 0.386188 +trainer/Advantage Weights Mean 7.98666 +trainer/Advantage Weights Std 23.0292 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.33982e-14 +trainer/Advantage Score Mean -0.271966 +trainer/Advantage Score Std 0.604258 +trainer/Advantage Score Max 1.50725 +trainer/Advantage Score Min -3.10303 +trainer/V1 Predictions Mean -70.2617 +trainer/V1 Predictions Std 21.8422 +trainer/V1 Predictions Max -0.188523 +trainer/V1 Predictions Min -86.3932 +trainer/VF Loss 0.0776792 +expl/num steps total 570000 +expl/num paths total 729 +expl/path length Mean 500 +expl/path length Std 332 +expl/path length Max 832 +expl/path length Min 168 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0271833 +expl/Actions Std 0.82105 +expl/Actions Max 2.59292 +expl/Actions Min -2.25937 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 530094 +eval/num paths total 571 +eval/path length Mean 496 +eval/path length Std 0 +eval/path length Max 496 +eval/path length Min 496 +eval/Rewards Mean 0.00201613 +eval/Rewards Std 0.044856 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0194398 +eval/Actions Std 0.735741 +eval/Actions Max 0.999898 +eval/Actions Min -0.999527 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.1018e-05 +time/evaluation sampling (s) 4.50515 +time/exploration sampling (s) 6.07033 +time/logging (s) 0.00858529 +time/saving (s) 0.0164446 +time/training (s) 19.4252 +time/epoch (s) 30.0257 +time/total (s) 13789.8 +Epoch -431 +------------------------------ ---------------- +2022-05-15 21:52:47.917299 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -430 finished +------------------------------ ---------------- +epoch -430 +replay_buffer/size 999047 +trainer/num train calls 571000 +trainer/QF1 Loss 0.543199 +trainer/QF2 Loss 0.458739 +trainer/Policy Loss 18.0012 +trainer/Q1 Predictions Mean -74.6801 +trainer/Q1 Predictions Std 16.9858 +trainer/Q1 Predictions Max -0.813973 +trainer/Q1 Predictions Min -86.8821 +trainer/Q2 Predictions Mean -74.6227 +trainer/Q2 Predictions Std 17.0577 +trainer/Q2 Predictions Max -0.965776 +trainer/Q2 Predictions Min -86.7812 +trainer/Q Targets Mean -74.7674 +trainer/Q Targets Std 16.9432 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2964 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.015025 +trainer/policy/mean Std 0.724886 +trainer/policy/mean Max 0.999362 +trainer/policy/mean Min -0.999346 +trainer/policy/std Mean 0.4174 +trainer/policy/std Std 0.0189782 +trainer/policy/std Max 0.437926 +trainer/policy/std Min 0.385988 +trainer/Advantage Weights Mean 5.0507 +trainer/Advantage Weights Std 18.1625 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.03261e-12 +trainer/Advantage Score Mean -0.223806 +trainer/Advantage Score Std 0.500753 +trainer/Advantage Score Max 1.11166 +trainer/Advantage Score Min -2.69217 +trainer/V1 Predictions Mean -74.5717 +trainer/V1 Predictions Std 16.9745 +trainer/V1 Predictions Max -0.355343 +trainer/V1 Predictions Min -87.0903 +trainer/VF Loss 0.0534362 +expl/num steps total 571000 +expl/num paths total 731 +expl/path length Mean 500 +expl/path length Std 126 +expl/path length Max 626 +expl/path length Min 374 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.030086 +expl/Actions Std 0.82875 +expl/Actions Max 2.37393 +expl/Actions Min -2.19188 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 531094 +eval/num paths total 572 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0298759 +eval/Actions Std 0.740421 +eval/Actions Max 0.999612 +eval/Actions Min -0.999499 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.30739e-05 +time/evaluation sampling (s) 4.47316 +time/exploration sampling (s) 6.61228 +time/logging (s) 0.012587 +time/saving (s) 0.018197 +time/training (s) 18.8549 +time/epoch (s) 29.9711 +time/total (s) 13819.8 +Epoch -430 +------------------------------ ---------------- +2022-05-15 21:53:19.096138 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -429 finished +------------------------------ ---------------- +epoch -429 +replay_buffer/size 999047 +trainer/num train calls 572000 +trainer/QF1 Loss 1.53794 +trainer/QF2 Loss 1.73192 +trainer/Policy Loss 23.018 +trainer/Q1 Predictions Mean -73.8151 +trainer/Q1 Predictions Std 18.3775 +trainer/Q1 Predictions Max -0.565891 +trainer/Q1 Predictions Min -87.2138 +trainer/Q2 Predictions Mean -73.7111 +trainer/Q2 Predictions Std 18.3538 +trainer/Q2 Predictions Max -0.417521 +trainer/Q2 Predictions Min -86.5496 +trainer/Q Targets Mean -74.2377 +trainer/Q Targets Std 18.5792 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5359 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0539179 +trainer/policy/mean Std 0.731612 +trainer/policy/mean Max 0.999269 +trainer/policy/mean Min -0.999523 +trainer/policy/std Mean 0.416008 +trainer/policy/std Std 0.0206831 +trainer/policy/std Max 0.436975 +trainer/policy/std Min 0.383354 +trainer/Advantage Weights Mean 6.80294 +trainer/Advantage Weights Std 20.5227 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.11905e-12 +trainer/Advantage Score Mean -0.212413 +trainer/Advantage Score Std 0.492181 +trainer/Advantage Score Max 1.02534 +trainer/Advantage Score Min -2.75185 +trainer/V1 Predictions Mean -73.9543 +trainer/V1 Predictions Std 18.6062 +trainer/V1 Predictions Max 0.0637184 +trainer/V1 Predictions Min -87.1954 +trainer/VF Loss 0.0524959 +expl/num steps total 572000 +expl/num paths total 732 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0707554 +expl/Actions Std 0.907927 +expl/Actions Max 2.46631 +expl/Actions Min -2.30124 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 532094 +eval/num paths total 573 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.175869 +eval/Actions Std 0.727503 +eval/Actions Max 0.999768 +eval/Actions Min -0.999076 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.61799e-05 +time/evaluation sampling (s) 4.88058 +time/exploration sampling (s) 7.43179 +time/logging (s) 0.0112523 +time/saving (s) 0.0158668 +time/training (s) 18.8204 +time/epoch (s) 31.1599 +time/total (s) 13851 +Epoch -429 +------------------------------ ---------------- +2022-05-15 21:53:49.397768 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -428 finished +------------------------------ ---------------- +epoch -428 +replay_buffer/size 999047 +trainer/num train calls 573000 +trainer/QF1 Loss 0.671966 +trainer/QF2 Loss 0.759574 +trainer/Policy Loss 13.0631 +trainer/Q1 Predictions Mean -74.8962 +trainer/Q1 Predictions Std 16.7285 +trainer/Q1 Predictions Max -0.257859 +trainer/Q1 Predictions Min -87.0994 +trainer/Q2 Predictions Mean -74.9594 +trainer/Q2 Predictions Std 16.7198 +trainer/Q2 Predictions Max -0.470385 +trainer/Q2 Predictions Min -87.2108 +trainer/Q Targets Mean -74.5441 +trainer/Q Targets Std 16.4882 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7267 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0106956 +trainer/policy/mean Std 0.729144 +trainer/policy/mean Max 0.999773 +trainer/policy/mean Min -0.99973 +trainer/policy/std Mean 0.415939 +trainer/policy/std Std 0.0196885 +trainer/policy/std Max 0.43673 +trainer/policy/std Min 0.384746 +trainer/Advantage Weights Mean 4.41935 +trainer/Advantage Weights Std 19.4023 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.40866e-13 +trainer/Advantage Score Mean -0.392425 +trainer/Advantage Score Std 0.517799 +trainer/Advantage Score Max 1.70492 +trainer/Advantage Score Min -2.9591 +trainer/V1 Predictions Mean -74.3281 +trainer/V1 Predictions Std 16.5564 +trainer/V1 Predictions Max -0.238127 +trainer/V1 Predictions Min -86.6838 +trainer/VF Loss 0.0824469 +expl/num steps total 573000 +expl/num paths total 733 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.253023 +expl/Actions Std 0.847545 +expl/Actions Max 2.59068 +expl/Actions Min -2.59717 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 533094 +eval/num paths total 574 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0918867 +eval/Actions Std 0.727311 +eval/Actions Max 0.999936 +eval/Actions Min -0.999739 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.13274e-06 +time/evaluation sampling (s) 4.74705 +time/exploration sampling (s) 7.04403 +time/logging (s) 0.00797886 +time/saving (s) 0.0162726 +time/training (s) 18.468 +time/epoch (s) 30.2833 +time/total (s) 13881.3 +Epoch -428 +------------------------------ ---------------- +2022-05-15 21:54:20.546125 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -427 finished +------------------------------ ---------------- +epoch -427 +replay_buffer/size 999047 +trainer/num train calls 574000 +trainer/QF1 Loss 0.543745 +trainer/QF2 Loss 0.556685 +trainer/Policy Loss 27.5725 +trainer/Q1 Predictions Mean -75.1183 +trainer/Q1 Predictions Std 17.5994 +trainer/Q1 Predictions Max -0.253269 +trainer/Q1 Predictions Min -87.1474 +trainer/Q2 Predictions Mean -75.1332 +trainer/Q2 Predictions Std 17.6206 +trainer/Q2 Predictions Max -0.212606 +trainer/Q2 Predictions Min -87.229 +trainer/Q Targets Mean -74.956 +trainer/Q Targets Std 17.5671 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3577 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00792523 +trainer/policy/mean Std 0.732889 +trainer/policy/mean Max 0.999307 +trainer/policy/mean Min -0.998443 +trainer/policy/std Mean 0.416503 +trainer/policy/std Std 0.0202857 +trainer/policy/std Max 0.438406 +trainer/policy/std Min 0.385905 +trainer/Advantage Weights Mean 5.05286 +trainer/Advantage Weights Std 18.45 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.85181e-18 +trainer/Advantage Score Mean -0.292713 +trainer/Advantage Score Std 0.585084 +trainer/Advantage Score Max 2.03759 +trainer/Advantage Score Min -3.9522 +trainer/V1 Predictions Mean -74.7558 +trainer/V1 Predictions Std 17.6926 +trainer/V1 Predictions Max -0.128807 +trainer/V1 Predictions Min -87.2275 +trainer/VF Loss 0.071212 +expl/num steps total 574000 +expl/num paths total 735 +expl/path length Mean 500 +expl/path length Std 160 +expl/path length Max 660 +expl/path length Min 340 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0353438 +expl/Actions Std 0.81742 +expl/Actions Max 2.29311 +expl/Actions Min -2.71168 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 534094 +eval/num paths total 575 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.248884 +eval/Actions Std 0.780772 +eval/Actions Max 0.999635 +eval/Actions Min -0.999627 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.1126e-05 +time/evaluation sampling (s) 5.05565 +time/exploration sampling (s) 6.72547 +time/logging (s) 0.0133544 +time/saving (s) 0.0175736 +time/training (s) 19.3231 +time/epoch (s) 31.1351 +time/total (s) 13912.4 +Epoch -427 +------------------------------ ---------------- +2022-05-15 21:54:50.496086 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -426 finished +------------------------------ ---------------- +epoch -426 +replay_buffer/size 999047 +trainer/num train calls 575000 +trainer/QF1 Loss 7.85359 +trainer/QF2 Loss 7.94625 +trainer/Policy Loss 53.4941 +trainer/Q1 Predictions Mean -73.0496 +trainer/Q1 Predictions Std 17.8533 +trainer/Q1 Predictions Max -0.322102 +trainer/Q1 Predictions Min -86.4243 +trainer/Q2 Predictions Mean -73.1399 +trainer/Q2 Predictions Std 17.8711 +trainer/Q2 Predictions Max -0.357099 +trainer/Q2 Predictions Min -86.7037 +trainer/Q Targets Mean -73.1314 +trainer/Q Targets Std 18.3234 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9168 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000950439 +trainer/policy/mean Std 0.709279 +trainer/policy/mean Max 0.999944 +trainer/policy/mean Min -0.998149 +trainer/policy/std Mean 0.417369 +trainer/policy/std Std 0.0199478 +trainer/policy/std Max 0.440284 +trainer/policy/std Min 0.385746 +trainer/Advantage Weights Mean 10.1125 +trainer/Advantage Weights Std 24.6082 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.05635e-19 +trainer/Advantage Score Mean -0.211565 +trainer/Advantage Score Std 0.687366 +trainer/Advantage Score Max 1.64407 +trainer/Advantage Score Min -4.30282 +trainer/V1 Predictions Mean -73.0327 +trainer/V1 Predictions Std 18.1561 +trainer/V1 Predictions Max -0.630951 +trainer/V1 Predictions Min -86.871 +trainer/VF Loss 0.0958606 +expl/num steps total 575000 +expl/num paths total 737 +expl/path length Mean 500 +expl/path length Std 171 +expl/path length Max 671 +expl/path length Min 329 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0468061 +expl/Actions Std 0.82972 +expl/Actions Max 2.2825 +expl/Actions Min -2.31451 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 535094 +eval/num paths total 576 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0276409 +eval/Actions Std 0.798254 +eval/Actions Max 0.998912 +eval/Actions Min -0.99768 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.03931e-05 +time/evaluation sampling (s) 4.58669 +time/exploration sampling (s) 6.46104 +time/logging (s) 0.0104871 +time/saving (s) 0.015397 +time/training (s) 18.8533 +time/epoch (s) 29.9269 +time/total (s) 13942.4 +Epoch -426 +------------------------------ ---------------- +2022-05-15 21:55:21.609320 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -425 finished +------------------------------ ---------------- +epoch -425 +replay_buffer/size 999047 +trainer/num train calls 576000 +trainer/QF1 Loss 0.702833 +trainer/QF2 Loss 0.885284 +trainer/Policy Loss 4.436 +trainer/Q1 Predictions Mean -73.6589 +trainer/Q1 Predictions Std 19.1205 +trainer/Q1 Predictions Max -2.48294 +trainer/Q1 Predictions Min -87.0802 +trainer/Q2 Predictions Mean -73.638 +trainer/Q2 Predictions Std 19.1737 +trainer/Q2 Predictions Max -2.98715 +trainer/Q2 Predictions Min -87.3298 +trainer/Q Targets Mean -73.3847 +trainer/Q Targets Std 18.8822 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9057 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0166578 +trainer/policy/mean Std 0.712112 +trainer/policy/mean Max 0.999973 +trainer/policy/mean Min -0.997602 +trainer/policy/std Mean 0.419065 +trainer/policy/std Std 0.0199127 +trainer/policy/std Max 0.441023 +trainer/policy/std Min 0.387767 +trainer/Advantage Weights Mean 1.21195 +trainer/Advantage Weights Std 9.33349 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.24973e-29 +trainer/Advantage Score Mean -0.570833 +trainer/Advantage Score Std 0.61851 +trainer/Advantage Score Max 2.47286 +trainer/Advantage Score Min -6.51168 +trainer/V1 Predictions Mean -73.1454 +trainer/V1 Predictions Std 19.0566 +trainer/V1 Predictions Max -3.00888 +trainer/V1 Predictions Min -86.883 +trainer/VF Loss 0.0921375 +expl/num steps total 576000 +expl/num paths total 738 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.101477 +expl/Actions Std 0.805107 +expl/Actions Max 2.43029 +expl/Actions Min -2.39469 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 536078 +eval/num paths total 578 +eval/path length Mean 492 +eval/path length Std 88 +eval/path length Max 580 +eval/path length Min 404 +eval/Rewards Mean 0.00203252 +eval/Rewards Std 0.0450376 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0183876 +eval/Actions Std 0.735938 +eval/Actions Max 0.999842 +eval/Actions Min -0.99969 +eval/Num Paths 2 +eval/Average Returns 1 +time/data storing (s) 1.05971e-05 +time/evaluation sampling (s) 5.22269 +time/exploration sampling (s) 6.59742 +time/logging (s) 0.0105282 +time/saving (s) 0.0138215 +time/training (s) 19.253 +time/epoch (s) 31.0975 +time/total (s) 13973.5 +Epoch -425 +------------------------------ ---------------- +2022-05-15 21:55:51.627176 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -424 finished +------------------------------ ---------------- +epoch -424 +replay_buffer/size 999047 +trainer/num train calls 577000 +trainer/QF1 Loss 1.05687 +trainer/QF2 Loss 0.998358 +trainer/Policy Loss 10.3376 +trainer/Q1 Predictions Mean -73.8353 +trainer/Q1 Predictions Std 17.4692 +trainer/Q1 Predictions Max -0.55916 +trainer/Q1 Predictions Min -86.5742 +trainer/Q2 Predictions Mean -73.8745 +trainer/Q2 Predictions Std 17.3732 +trainer/Q2 Predictions Max -1.03478 +trainer/Q2 Predictions Min -86.5308 +trainer/Q Targets Mean -74.2473 +trainer/Q Targets Std 17.3951 +trainer/Q Targets Max -1.55434 +trainer/Q Targets Min -87.0382 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0141622 +trainer/policy/mean Std 0.723175 +trainer/policy/mean Max 0.999356 +trainer/policy/mean Min -0.999532 +trainer/policy/std Mean 0.417613 +trainer/policy/std Std 0.0194271 +trainer/policy/std Max 0.438999 +trainer/policy/std Min 0.38626 +trainer/Advantage Weights Mean 3.24622 +trainer/Advantage Weights Std 14.931 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.8942e-12 +trainer/Advantage Score Mean -0.356575 +trainer/Advantage Score Std 0.4829 +trainer/Advantage Score Max 0.879189 +trainer/Advantage Score Min -2.69922 +trainer/V1 Predictions Mean -73.896 +trainer/V1 Predictions Std 17.7348 +trainer/V1 Predictions Max -0.230246 +trainer/V1 Predictions Min -87.0254 +trainer/VF Loss 0.047196 +expl/num steps total 577000 +expl/num paths total 739 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0460621 +expl/Actions Std 0.818235 +expl/Actions Max 2.1984 +expl/Actions Min -2.39024 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 537078 +eval/num paths total 579 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0760094 +eval/Actions Std 0.73053 +eval/Actions Max 0.999893 +eval/Actions Min -0.999754 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.16411e-05 +time/evaluation sampling (s) 4.75817 +time/exploration sampling (s) 6.07388 +time/logging (s) 0.012443 +time/saving (s) 0.0156361 +time/training (s) 19.1428 +time/epoch (s) 30.003 +time/total (s) 14003.5 +Epoch -424 +------------------------------ ---------------- +2022-05-15 21:56:22.958728 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -423 finished +------------------------------ ---------------- +epoch -423 +replay_buffer/size 999047 +trainer/num train calls 578000 +trainer/QF1 Loss 0.525273 +trainer/QF2 Loss 0.552188 +trainer/Policy Loss 25.0957 +trainer/Q1 Predictions Mean -75.3719 +trainer/Q1 Predictions Std 17.0665 +trainer/Q1 Predictions Max -0.644574 +trainer/Q1 Predictions Min -87.3284 +trainer/Q2 Predictions Mean -75.3802 +trainer/Q2 Predictions Std 16.984 +trainer/Q2 Predictions Max -1.00835 +trainer/Q2 Predictions Min -87.0498 +trainer/Q Targets Mean -75.4044 +trainer/Q Targets Std 17.0065 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0844 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0242884 +trainer/policy/mean Std 0.727711 +trainer/policy/mean Max 0.999823 +trainer/policy/mean Min -0.99937 +trainer/policy/std Mean 0.416153 +trainer/policy/std Std 0.0192273 +trainer/policy/std Max 0.434984 +trainer/policy/std Min 0.385086 +trainer/Advantage Weights Mean 5.92935 +trainer/Advantage Weights Std 18.499 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.45366e-16 +trainer/Advantage Score Mean -0.280144 +trainer/Advantage Score Std 0.646234 +trainer/Advantage Score Max 1.50698 +trainer/Advantage Score Min -3.64673 +trainer/V1 Predictions Mean -75.1519 +trainer/V1 Predictions Std 17.1366 +trainer/V1 Predictions Max -0.066905 +trainer/V1 Predictions Min -87.0802 +trainer/VF Loss 0.0747021 +expl/num steps total 578000 +expl/num paths total 741 +expl/path length Mean 500 +expl/path length Std 410 +expl/path length Max 910 +expl/path length Min 90 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00793504 +expl/Actions Std 0.830681 +expl/Actions Max 2.23123 +expl/Actions Min -2.17065 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 538078 +eval/num paths total 580 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.264817 +eval/Actions Std 0.656232 +eval/Actions Max 0.999693 +eval/Actions Min -0.999441 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.06199e-05 +time/evaluation sampling (s) 4.51792 +time/exploration sampling (s) 7.51556 +time/logging (s) 0.0119515 +time/saving (s) 0.0144976 +time/training (s) 19.2562 +time/epoch (s) 31.3162 +time/total (s) 14034.8 +Epoch -423 +------------------------------ ---------------- +2022-05-15 21:56:53.356070 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -422 finished +------------------------------ ---------------- +epoch -422 +replay_buffer/size 999047 +trainer/num train calls 579000 +trainer/QF1 Loss 1.34102 +trainer/QF2 Loss 1.65673 +trainer/Policy Loss 33.4224 +trainer/Q1 Predictions Mean -72.2327 +trainer/Q1 Predictions Std 20.3506 +trainer/Q1 Predictions Max -0.449609 +trainer/Q1 Predictions Min -87.3833 +trainer/Q2 Predictions Mean -72.1071 +trainer/Q2 Predictions Std 20.5086 +trainer/Q2 Predictions Max -1.03266 +trainer/Q2 Predictions Min -87.4924 +trainer/Q Targets Mean -72.1358 +trainer/Q Targets Std 20.0621 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9202 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00114851 +trainer/policy/mean Std 0.72812 +trainer/policy/mean Max 0.999953 +trainer/policy/mean Min -0.997384 +trainer/policy/std Mean 0.416464 +trainer/policy/std Std 0.0199059 +trainer/policy/std Max 0.438181 +trainer/policy/std Min 0.387163 +trainer/Advantage Weights Mean 6.31687 +trainer/Advantage Weights Std 22.9579 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.10771e-18 +trainer/Advantage Score Mean -0.35643 +trainer/Advantage Score Std 0.534558 +trainer/Advantage Score Max 1.11027 +trainer/Advantage Score Min -4.03126 +trainer/V1 Predictions Mean -71.9754 +trainer/V1 Predictions Std 20.0418 +trainer/V1 Predictions Max -1.46225 +trainer/V1 Predictions Min -86.8451 +trainer/VF Loss 0.0676499 +expl/num steps total 579000 +expl/num paths total 742 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.14141 +expl/Actions Std 0.811518 +expl/Actions Max 2.33754 +expl/Actions Min -2.35236 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 539078 +eval/num paths total 581 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.013191 +eval/Actions Std 0.735686 +eval/Actions Max 0.999679 +eval/Actions Min -0.999706 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.43802e-06 +time/evaluation sampling (s) 4.82011 +time/exploration sampling (s) 6.46882 +time/logging (s) 0.0111057 +time/saving (s) 0.0166613 +time/training (s) 19.0658 +time/epoch (s) 30.3825 +time/total (s) 14065.2 +Epoch -422 +------------------------------ ---------------- +2022-05-15 21:57:24.401021 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -421 finished +------------------------------ ---------------- +epoch -421 +replay_buffer/size 999047 +trainer/num train calls 580000 +trainer/QF1 Loss 0.606258 +trainer/QF2 Loss 0.586528 +trainer/Policy Loss 11.5344 +trainer/Q1 Predictions Mean -74.2638 +trainer/Q1 Predictions Std 17.2309 +trainer/Q1 Predictions Max -2.09209 +trainer/Q1 Predictions Min -87.879 +trainer/Q2 Predictions Mean -74.1998 +trainer/Q2 Predictions Std 17.2488 +trainer/Q2 Predictions Max -2.26793 +trainer/Q2 Predictions Min -87.6949 +trainer/Q Targets Mean -74.0172 +trainer/Q Targets Std 17.0586 +trainer/Q Targets Max -0.908478 +trainer/Q Targets Min -87.6467 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0150793 +trainer/policy/mean Std 0.721771 +trainer/policy/mean Max 0.999463 +trainer/policy/mean Min -0.99949 +trainer/policy/std Mean 0.416965 +trainer/policy/std Std 0.0194698 +trainer/policy/std Max 0.439421 +trainer/policy/std Min 0.38485 +trainer/Advantage Weights Mean 3.18155 +trainer/Advantage Weights Std 16.1008 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.7561e-12 +trainer/Advantage Score Mean -0.387107 +trainer/Advantage Score Std 0.497681 +trainer/Advantage Score Max 1.27067 +trainer/Advantage Score Min -2.66172 +trainer/V1 Predictions Mean -73.687 +trainer/V1 Predictions Std 17.2999 +trainer/V1 Predictions Max -1.50294 +trainer/V1 Predictions Min -87.4182 +trainer/VF Loss 0.0535909 +expl/num steps total 580000 +expl/num paths total 744 +expl/path length Mean 500 +expl/path length Std 239 +expl/path length Max 739 +expl/path length Min 261 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0193713 +expl/Actions Std 0.823325 +expl/Actions Max 2.41338 +expl/Actions Min -2.26276 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 540078 +eval/num paths total 582 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.165022 +eval/Actions Std 0.6736 +eval/Actions Max 0.999857 +eval/Actions Min -0.99967 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.03768e-05 +time/evaluation sampling (s) 4.99625 +time/exploration sampling (s) 6.68883 +time/logging (s) 0.0120419 +time/saving (s) 0.0164792 +time/training (s) 19.3132 +time/epoch (s) 31.0268 +time/total (s) 14096.2 +Epoch -421 +------------------------------ ---------------- +2022-05-15 21:57:55.305804 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -420 finished +------------------------------ ---------------- +epoch -420 +replay_buffer/size 999047 +trainer/num train calls 581000 +trainer/QF1 Loss 0.684698 +trainer/QF2 Loss 0.708256 +trainer/Policy Loss 23.5275 +trainer/Q1 Predictions Mean -73.3501 +trainer/Q1 Predictions Std 16.9623 +trainer/Q1 Predictions Max -2.31786 +trainer/Q1 Predictions Min -87.9901 +trainer/Q2 Predictions Mean -73.4371 +trainer/Q2 Predictions Std 17.0281 +trainer/Q2 Predictions Max -2.18459 +trainer/Q2 Predictions Min -87.9668 +trainer/Q Targets Mean -73.4346 +trainer/Q Targets Std 16.9803 +trainer/Q Targets Max -3.77869 +trainer/Q Targets Min -87.8742 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00570565 +trainer/policy/mean Std 0.733196 +trainer/policy/mean Max 0.99885 +trainer/policy/mean Min -0.999344 +trainer/policy/std Mean 0.417719 +trainer/policy/std Std 0.0196309 +trainer/policy/std Max 0.438293 +trainer/policy/std Min 0.385967 +trainer/Advantage Weights Mean 5.73237 +trainer/Advantage Weights Std 20.0944 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.41046e-16 +trainer/Advantage Score Mean -0.328846 +trainer/Advantage Score Std 0.656077 +trainer/Advantage Score Max 2.56491 +trainer/Advantage Score Min -3.5153 +trainer/V1 Predictions Mean -73.1931 +trainer/V1 Predictions Std 17.105 +trainer/V1 Predictions Max -2.4795 +trainer/V1 Predictions Min -87.6142 +trainer/VF Loss 0.0934186 +expl/num steps total 581000 +expl/num paths total 746 +expl/path length Mean 500 +expl/path length Std 362 +expl/path length Max 862 +expl/path length Min 138 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0311186 +expl/Actions Std 0.835615 +expl/Actions Max 2.36682 +expl/Actions Min -2.391 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 541078 +eval/num paths total 583 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0576334 +eval/Actions Std 0.715736 +eval/Actions Max 0.999901 +eval/Actions Min -0.999269 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.3392e-05 +time/evaluation sampling (s) 5.03949 +time/exploration sampling (s) 6.30638 +time/logging (s) 0.00801842 +time/saving (s) 0.017754 +time/training (s) 19.5145 +time/epoch (s) 30.8861 +time/total (s) 14127.1 +Epoch -420 +------------------------------ ---------------- +2022-05-15 21:58:25.721952 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -419 finished +------------------------------ ---------------- +epoch -419 +replay_buffer/size 999047 +trainer/num train calls 582000 +trainer/QF1 Loss 1.19242 +trainer/QF2 Loss 1.19969 +trainer/Policy Loss 19.0723 +trainer/Q1 Predictions Mean -73.5357 +trainer/Q1 Predictions Std 17.7478 +trainer/Q1 Predictions Max -0.496334 +trainer/Q1 Predictions Min -87.3698 +trainer/Q2 Predictions Mean -73.5242 +trainer/Q2 Predictions Std 17.7431 +trainer/Q2 Predictions Max 0.778137 +trainer/Q2 Predictions Min -87.1979 +trainer/Q Targets Mean -73.5645 +trainer/Q Targets Std 17.7361 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2053 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0132972 +trainer/policy/mean Std 0.731124 +trainer/policy/mean Max 0.999981 +trainer/policy/mean Min -0.999716 +trainer/policy/std Mean 0.416125 +trainer/policy/std Std 0.0198508 +trainer/policy/std Max 0.436559 +trainer/policy/std Min 0.382642 +trainer/Advantage Weights Mean 6.01485 +trainer/Advantage Weights Std 20.7935 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.09045e-16 +trainer/Advantage Score Mean -0.296951 +trainer/Advantage Score Std 0.640186 +trainer/Advantage Score Max 4.18178 +trainer/Advantage Score Min -3.5713 +trainer/V1 Predictions Mean -73.2858 +trainer/V1 Predictions Std 17.7197 +trainer/V1 Predictions Max -0.782603 +trainer/V1 Predictions Min -87.0601 +trainer/VF Loss 0.123658 +expl/num steps total 582000 +expl/num paths total 747 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0166968 +expl/Actions Std 0.831839 +expl/Actions Max 2.58282 +expl/Actions Min -2.40174 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 541563 +eval/num paths total 584 +eval/path length Mean 485 +eval/path length Std 0 +eval/path length Max 485 +eval/path length Min 485 +eval/Rewards Mean 0.00206186 +eval/Rewards Std 0.0453608 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0341924 +eval/Actions Std 0.73919 +eval/Actions Max 0.999693 +eval/Actions Min -0.9994 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.33137e-05 +time/evaluation sampling (s) 4.55275 +time/exploration sampling (s) 6.18834 +time/logging (s) 0.00854115 +time/saving (s) 0.0156117 +time/training (s) 19.6382 +time/epoch (s) 30.4035 +time/total (s) 14157.5 +Epoch -419 +------------------------------ ---------------- +2022-05-15 21:58:55.518274 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -418 finished +------------------------------ ---------------- +epoch -418 +replay_buffer/size 999047 +trainer/num train calls 583000 +trainer/QF1 Loss 0.996969 +trainer/QF2 Loss 1.0143 +trainer/Policy Loss 14.2111 +trainer/Q1 Predictions Mean -73.6559 +trainer/Q1 Predictions Std 19.4215 +trainer/Q1 Predictions Max -0.706279 +trainer/Q1 Predictions Min -87.3087 +trainer/Q2 Predictions Mean -73.706 +trainer/Q2 Predictions Std 19.405 +trainer/Q2 Predictions Max -1.02574 +trainer/Q2 Predictions Min -87.2394 +trainer/Q Targets Mean -73.3644 +trainer/Q Targets Std 19.5652 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7571 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.037767 +trainer/policy/mean Std 0.734358 +trainer/policy/mean Max 0.999142 +trainer/policy/mean Min -0.999317 +trainer/policy/std Mean 0.417141 +trainer/policy/std Std 0.0196491 +trainer/policy/std Max 0.438111 +trainer/policy/std Min 0.384876 +trainer/Advantage Weights Mean 3.75645 +trainer/Advantage Weights Std 16.253 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.65375e-17 +trainer/Advantage Score Mean -0.318597 +trainer/Advantage Score Std 0.501689 +trainer/Advantage Score Max 0.90235 +trainer/Advantage Score Min -3.8168 +trainer/V1 Predictions Mean -73.169 +trainer/V1 Predictions Std 19.4559 +trainer/V1 Predictions Max -1.05672 +trainer/V1 Predictions Min -86.4462 +trainer/VF Loss 0.0464936 +expl/num steps total 583000 +expl/num paths total 748 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0154289 +expl/Actions Std 0.808424 +expl/Actions Max 2.43862 +expl/Actions Min -2.21168 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 542057 +eval/num paths total 585 +eval/path length Mean 494 +eval/path length Std 0 +eval/path length Max 494 +eval/path length Min 494 +eval/Rewards Mean 0.00202429 +eval/Rewards Std 0.0449466 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0284101 +eval/Actions Std 0.745174 +eval/Actions Max 0.999795 +eval/Actions Min -0.999419 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.1339e-06 +time/evaluation sampling (s) 4.12377 +time/exploration sampling (s) 6.74643 +time/logging (s) 0.00974452 +time/saving (s) 0.0168047 +time/training (s) 18.8859 +time/epoch (s) 29.7827 +time/total (s) 14187.3 +Epoch -418 +------------------------------ ---------------- +2022-05-15 21:59:25.829585 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -417 finished +------------------------------ ---------------- +epoch -417 +replay_buffer/size 999047 +trainer/num train calls 584000 +trainer/QF1 Loss 1.07859 +trainer/QF2 Loss 0.929959 +trainer/Policy Loss 15.5917 +trainer/Q1 Predictions Mean -75.7475 +trainer/Q1 Predictions Std 14.8554 +trainer/Q1 Predictions Max -4.53201 +trainer/Q1 Predictions Min -87.1684 +trainer/Q2 Predictions Mean -75.7857 +trainer/Q2 Predictions Std 14.7413 +trainer/Q2 Predictions Max -5.10777 +trainer/Q2 Predictions Min -86.618 +trainer/Q Targets Mean -75.6429 +trainer/Q Targets Std 14.3912 +trainer/Q Targets Max -5.36777 +trainer/Q Targets Min -86.7599 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0181619 +trainer/policy/mean Std 0.728846 +trainer/policy/mean Max 0.999849 +trainer/policy/mean Min -0.997584 +trainer/policy/std Mean 0.417817 +trainer/policy/std Std 0.0206891 +trainer/policy/std Max 0.441498 +trainer/policy/std Min 0.385373 +trainer/Advantage Weights Mean 3.70872 +trainer/Advantage Weights Std 17.5835 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.11627e-15 +trainer/Advantage Score Mean -0.427313 +trainer/Advantage Score Std 0.491277 +trainer/Advantage Score Max 1.28507 +trainer/Advantage Score Min -3.44288 +trainer/V1 Predictions Mean -75.2779 +trainer/V1 Predictions Std 14.8118 +trainer/V1 Predictions Max -4.09564 +trainer/V1 Predictions Min -86.5027 +trainer/VF Loss 0.057334 +expl/num steps total 584000 +expl/num paths total 750 +expl/path length Mean 500 +expl/path length Std 202 +expl/path length Max 702 +expl/path length Min 298 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.0088208 +expl/Actions Std 0.844242 +expl/Actions Max 2.20962 +expl/Actions Min -2.7293 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 543057 +eval/num paths total 586 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.243352 +eval/Actions Std 0.684827 +eval/Actions Max 0.999805 +eval/Actions Min -0.999309 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.2835e-05 +time/evaluation sampling (s) 4.41485 +time/exploration sampling (s) 6.69133 +time/logging (s) 0.0116386 +time/saving (s) 0.0150292 +time/training (s) 19.1635 +time/epoch (s) 30.2963 +time/total (s) 14217.6 +Epoch -417 +------------------------------ ---------------- +2022-05-15 21:59:56.176818 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -416 finished +------------------------------ ---------------- +epoch -416 +replay_buffer/size 999047 +trainer/num train calls 585000 +trainer/QF1 Loss 0.778533 +trainer/QF2 Loss 0.769187 +trainer/Policy Loss 14.8198 +trainer/Q1 Predictions Mean -75.2297 +trainer/Q1 Predictions Std 16.514 +trainer/Q1 Predictions Max -0.928788 +trainer/Q1 Predictions Min -87.9716 +trainer/Q2 Predictions Mean -75.2125 +trainer/Q2 Predictions Std 16.5128 +trainer/Q2 Predictions Max -0.52557 +trainer/Q2 Predictions Min -87.4351 +trainer/Q Targets Mean -75.246 +trainer/Q Targets Std 16.9361 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0461 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00710208 +trainer/policy/mean Std 0.724581 +trainer/policy/mean Max 0.998834 +trainer/policy/mean Min -0.999236 +trainer/policy/std Mean 0.415537 +trainer/policy/std Std 0.0199225 +trainer/policy/std Max 0.439878 +trainer/policy/std Min 0.387003 +trainer/Advantage Weights Mean 2.67856 +trainer/Advantage Weights Std 14.2356 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.6211e-17 +trainer/Advantage Score Mean -0.397762 +trainer/Advantage Score Std 0.557983 +trainer/Advantage Score Max 1.32511 +trainer/Advantage Score Min -3.81804 +trainer/V1 Predictions Mean -75.0798 +trainer/V1 Predictions Std 16.8294 +trainer/V1 Predictions Max -2.00464 +trainer/V1 Predictions Min -88.1013 +trainer/VF Loss 0.0636754 +expl/num steps total 585000 +expl/num paths total 751 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0406457 +expl/Actions Std 0.833388 +expl/Actions Max 2.2908 +expl/Actions Min -2.17006 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 544057 +eval/num paths total 587 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.172725 +eval/Actions Std 0.670071 +eval/Actions Max 0.999981 +eval/Actions Min -0.999417 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.30679e-05 +time/evaluation sampling (s) 4.54961 +time/exploration sampling (s) 6.92238 +time/logging (s) 0.0125438 +time/saving (s) 0.0193304 +time/training (s) 18.8248 +time/epoch (s) 30.3287 +time/total (s) 14247.9 +Epoch -416 +------------------------------ ---------------- +2022-05-15 22:00:26.963513 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -415 finished +------------------------------ ---------------- +epoch -415 +replay_buffer/size 999047 +trainer/num train calls 586000 +trainer/QF1 Loss 0.503389 +trainer/QF2 Loss 0.47505 +trainer/Policy Loss 18.4253 +trainer/Q1 Predictions Mean -75.687 +trainer/Q1 Predictions Std 13.7566 +trainer/Q1 Predictions Max -6.12273 +trainer/Q1 Predictions Min -86.3521 +trainer/Q2 Predictions Mean -75.7293 +trainer/Q2 Predictions Std 13.8746 +trainer/Q2 Predictions Max -6.82497 +trainer/Q2 Predictions Min -86.4415 +trainer/Q Targets Mean -75.9396 +trainer/Q Targets Std 13.6467 +trainer/Q Targets Max -8.79859 +trainer/Q Targets Min -86.5825 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00637813 +trainer/policy/mean Std 0.715717 +trainer/policy/mean Max 0.999169 +trainer/policy/mean Min -0.997906 +trainer/policy/std Mean 0.416776 +trainer/policy/std Std 0.019599 +trainer/policy/std Max 0.440424 +trainer/policy/std Min 0.387222 +trainer/Advantage Weights Mean 4.0468 +trainer/Advantage Weights Std 17.4301 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.09014e-13 +trainer/Advantage Score Mean -0.376136 +trainer/Advantage Score Std 0.515949 +trainer/Advantage Score Max 0.825862 +trainer/Advantage Score Min -2.91964 +trainer/V1 Predictions Mean -75.6137 +trainer/V1 Predictions Std 14.0045 +trainer/V1 Predictions Max -7.3633 +trainer/V1 Predictions Min -86.4956 +trainer/VF Loss 0.0550797 +expl/num steps total 586000 +expl/num paths total 752 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.341057 +expl/Actions Std 0.828118 +expl/Actions Max 2.23647 +expl/Actions Min -2.50085 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 544689 +eval/num paths total 588 +eval/path length Mean 632 +eval/path length Std 0 +eval/path length Max 632 +eval/path length Min 632 +eval/Rewards Mean 0.00158228 +eval/Rewards Std 0.0397464 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0290714 +eval/Actions Std 0.722331 +eval/Actions Max 0.999691 +eval/Actions Min -0.999006 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.04029e-05 +time/evaluation sampling (s) 4.89019 +time/exploration sampling (s) 6.65881 +time/logging (s) 0.0103876 +time/saving (s) 0.012605 +time/training (s) 19.1948 +time/epoch (s) 30.7668 +time/total (s) 14278.7 +Epoch -415 +------------------------------ ---------------- +2022-05-15 22:00:58.116922 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -414 finished +------------------------------ ---------------- +epoch -414 +replay_buffer/size 999047 +trainer/num train calls 587000 +trainer/QF1 Loss 0.902809 +trainer/QF2 Loss 0.993064 +trainer/Policy Loss 2.63729 +trainer/Q1 Predictions Mean -74.8763 +trainer/Q1 Predictions Std 18.8281 +trainer/Q1 Predictions Max -0.596516 +trainer/Q1 Predictions Min -87.4503 +trainer/Q2 Predictions Mean -74.8602 +trainer/Q2 Predictions Std 18.8369 +trainer/Q2 Predictions Max -0.246167 +trainer/Q2 Predictions Min -87.5195 +trainer/Q Targets Mean -74.3862 +trainer/Q Targets Std 18.5926 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7952 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0109091 +trainer/policy/mean Std 0.732539 +trainer/policy/mean Max 0.999432 +trainer/policy/mean Min -0.999092 +trainer/policy/std Mean 0.416543 +trainer/policy/std Std 0.021268 +trainer/policy/std Max 0.44311 +trainer/policy/std Min 0.381541 +trainer/Advantage Weights Mean 0.700767 +trainer/Advantage Weights Std 6.3091 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.68089e-15 +trainer/Advantage Score Mean -0.480891 +trainer/Advantage Score Std 0.448761 +trainer/Advantage Score Max 0.487899 +trainer/Advantage Score Min -3.35526 +trainer/V1 Predictions Mean -74.1181 +trainer/V1 Predictions Std 18.7987 +trainer/V1 Predictions Max -0.439185 +trainer/V1 Predictions Min -86.8169 +trainer/VF Loss 0.0448292 +expl/num steps total 587000 +expl/num paths total 753 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0673547 +expl/Actions Std 0.832687 +expl/Actions Max 2.68804 +expl/Actions Min -2.29192 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 545560 +eval/num paths total 589 +eval/path length Mean 871 +eval/path length Std 0 +eval/path length Max 871 +eval/path length Min 871 +eval/Rewards Mean 0.00114811 +eval/Rewards Std 0.0338643 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0496067 +eval/Actions Std 0.73547 +eval/Actions Max 0.999982 +eval/Actions Min -0.999576 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.05207e-05 +time/evaluation sampling (s) 4.63201 +time/exploration sampling (s) 6.87061 +time/logging (s) 0.0112856 +time/saving (s) 0.0119624 +time/training (s) 19.6126 +time/epoch (s) 31.1385 +time/total (s) 14309.9 +Epoch -414 +------------------------------ ---------------- +2022-05-15 22:01:28.295889 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -413 finished +------------------------------ ---------------- +epoch -413 +replay_buffer/size 999047 +trainer/num train calls 588000 +trainer/QF1 Loss 0.516598 +trainer/QF2 Loss 0.632588 +trainer/Policy Loss 10.4818 +trainer/Q1 Predictions Mean -75.1102 +trainer/Q1 Predictions Std 16.1422 +trainer/Q1 Predictions Max -0.624887 +trainer/Q1 Predictions Min -88.1274 +trainer/Q2 Predictions Mean -75.1819 +trainer/Q2 Predictions Std 16.1687 +trainer/Q2 Predictions Max -0.354711 +trainer/Q2 Predictions Min -88.049 +trainer/Q Targets Mean -74.8599 +trainer/Q Targets Std 16.1625 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6128 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0248103 +trainer/policy/mean Std 0.726126 +trainer/policy/mean Max 0.998822 +trainer/policy/mean Min -0.998066 +trainer/policy/std Mean 0.417134 +trainer/policy/std Std 0.0203952 +trainer/policy/std Max 0.439722 +trainer/policy/std Min 0.384567 +trainer/Advantage Weights Mean 1.9635 +trainer/Advantage Weights Std 11.3655 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.43705e-13 +trainer/Advantage Score Mean -0.360211 +trainer/Advantage Score Std 0.430334 +trainer/Advantage Score Max 1.8939 +trainer/Advantage Score Min -2.90428 +trainer/V1 Predictions Mean -74.6739 +trainer/V1 Predictions Std 16.1796 +trainer/V1 Predictions Max -0.639563 +trainer/V1 Predictions Min -87.5047 +trainer/VF Loss 0.0489501 +expl/num steps total 588000 +expl/num paths total 754 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.023815 +expl/Actions Std 0.824188 +expl/Actions Max 2.65051 +expl/Actions Min -2.66006 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 546277 +eval/num paths total 590 +eval/path length Mean 717 +eval/path length Std 0 +eval/path length Max 717 +eval/path length Min 717 +eval/Rewards Mean 0.0013947 +eval/Rewards Std 0.0373196 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0303118 +eval/Actions Std 0.724183 +eval/Actions Max 0.999852 +eval/Actions Min -0.999916 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 9.62336e-06 +time/evaluation sampling (s) 4.66506 +time/exploration sampling (s) 6.81096 +time/logging (s) 0.0110895 +time/saving (s) 0.0179458 +time/training (s) 18.6593 +time/epoch (s) 30.1644 +time/total (s) 14340 +Epoch -413 +------------------------------ ---------------- +2022-05-15 22:01:59.257128 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -412 finished +------------------------------ ---------------- +epoch -412 +replay_buffer/size 999047 +trainer/num train calls 589000 +trainer/QF1 Loss 0.884547 +trainer/QF2 Loss 0.883695 +trainer/Policy Loss 20.2531 +trainer/Q1 Predictions Mean -72.1038 +trainer/Q1 Predictions Std 20.0361 +trainer/Q1 Predictions Max -1.13368 +trainer/Q1 Predictions Min -87.6869 +trainer/Q2 Predictions Mean -72.0856 +trainer/Q2 Predictions Std 20.0494 +trainer/Q2 Predictions Max 0.134329 +trainer/Q2 Predictions Min -87.3133 +trainer/Q Targets Mean -72.1677 +trainer/Q Targets Std 19.5039 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0657 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00158989 +trainer/policy/mean Std 0.714858 +trainer/policy/mean Max 0.99935 +trainer/policy/mean Min -0.999383 +trainer/policy/std Mean 0.416361 +trainer/policy/std Std 0.0189884 +trainer/policy/std Max 0.439198 +trainer/policy/std Min 0.384621 +trainer/Advantage Weights Mean 6.22435 +trainer/Advantage Weights Std 21.7196 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.58116e-14 +trainer/Advantage Score Mean -0.310708 +trainer/Advantage Score Std 0.605417 +trainer/Advantage Score Max 1.75914 +trainer/Advantage Score Min -3.1778 +trainer/V1 Predictions Mean -71.8307 +trainer/V1 Predictions Std 19.8478 +trainer/V1 Predictions Max -0.926921 +trainer/V1 Predictions Min -86.9439 +trainer/VF Loss 0.0890465 +expl/num steps total 589000 +expl/num paths total 755 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.134442 +expl/Actions Std 0.882991 +expl/Actions Max 2.72391 +expl/Actions Min -2.40822 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 547277 +eval/num paths total 591 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0122157 +eval/Actions Std 0.66377 +eval/Actions Max 0.999615 +eval/Actions Min -0.998183 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.64402e-06 +time/evaluation sampling (s) 4.53159 +time/exploration sampling (s) 7.08489 +time/logging (s) 0.0104929 +time/saving (s) 0.0149774 +time/training (s) 19.3021 +time/epoch (s) 30.944 +time/total (s) 14371 +Epoch -412 +------------------------------ ---------------- +2022-05-15 22:02:30.440584 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -411 finished +------------------------------ ---------------- +epoch -411 +replay_buffer/size 999047 +trainer/num train calls 590000 +trainer/QF1 Loss 1.72112 +trainer/QF2 Loss 1.8485 +trainer/Policy Loss 21.8544 +trainer/Q1 Predictions Mean -72.8388 +trainer/Q1 Predictions Std 18.1963 +trainer/Q1 Predictions Max -1.6356 +trainer/Q1 Predictions Min -86.743 +trainer/Q2 Predictions Mean -72.8648 +trainer/Q2 Predictions Std 18.23 +trainer/Q2 Predictions Max -2.08727 +trainer/Q2 Predictions Min -87.0097 +trainer/Q Targets Mean -73.0212 +trainer/Q Targets Std 18.4477 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7458 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00388409 +trainer/policy/mean Std 0.730854 +trainer/policy/mean Max 0.998358 +trainer/policy/mean Min -0.998108 +trainer/policy/std Mean 0.413295 +trainer/policy/std Std 0.0197562 +trainer/policy/std Max 0.434975 +trainer/policy/std Min 0.384013 +trainer/Advantage Weights Mean 5.01178 +trainer/Advantage Weights Std 18.7759 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.42984e-15 +trainer/Advantage Score Mean -0.326349 +trainer/Advantage Score Std 0.563006 +trainer/Advantage Score Max 1.94815 +trainer/Advantage Score Min -3.41812 +trainer/V1 Predictions Mean -72.8768 +trainer/V1 Predictions Std 18.4084 +trainer/V1 Predictions Max -1.18927 +trainer/V1 Predictions Min -87.0058 +trainer/VF Loss 0.0685449 +expl/num steps total 590000 +expl/num paths total 757 +expl/path length Mean 500 +expl/path length Std 353 +expl/path length Max 853 +expl/path length Min 147 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0385131 +expl/Actions Std 0.832845 +expl/Actions Max 2.38444 +expl/Actions Min -2.53522 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 548277 +eval/num paths total 592 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.084509 +eval/Actions Std 0.696707 +eval/Actions Max 0.999737 +eval/Actions Min -0.999138 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.09142e-05 +time/evaluation sampling (s) 4.9413 +time/exploration sampling (s) 6.93204 +time/logging (s) 0.0111408 +time/saving (s) 0.0138822 +time/training (s) 19.2698 +time/epoch (s) 31.1681 +time/total (s) 14402.2 +Epoch -411 +------------------------------ ---------------- +2022-05-15 22:03:00.896314 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -410 finished +------------------------------ ---------------- +epoch -410 +replay_buffer/size 999047 +trainer/num train calls 591000 +trainer/QF1 Loss 0.525546 +trainer/QF2 Loss 0.550812 +trainer/Policy Loss 19.1377 +trainer/Q1 Predictions Mean -74.1828 +trainer/Q1 Predictions Std 17.6011 +trainer/Q1 Predictions Max -1.59569 +trainer/Q1 Predictions Min -88.1168 +trainer/Q2 Predictions Mean -74.2096 +trainer/Q2 Predictions Std 17.5537 +trainer/Q2 Predictions Max -2.33261 +trainer/Q2 Predictions Min -88.032 +trainer/Q Targets Mean -73.8602 +trainer/Q Targets Std 17.6811 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4419 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.023589 +trainer/policy/mean Std 0.726087 +trainer/policy/mean Max 0.999726 +trainer/policy/mean Min -0.999269 +trainer/policy/std Mean 0.414553 +trainer/policy/std Std 0.020564 +trainer/policy/std Max 0.437103 +trainer/policy/std Min 0.384074 +trainer/Advantage Weights Mean 4.10234 +trainer/Advantage Weights Std 17.7871 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.33381e-13 +trainer/Advantage Score Mean -0.397978 +trainer/Advantage Score Std 0.561682 +trainer/Advantage Score Max 2.57244 +trainer/Advantage Score Min -2.78133 +trainer/V1 Predictions Mean -73.6617 +trainer/V1 Predictions Std 17.6437 +trainer/V1 Predictions Max -1.46697 +trainer/V1 Predictions Min -87.3849 +trainer/VF Loss 0.088145 +expl/num steps total 591000 +expl/num paths total 758 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.08607 +expl/Actions Std 0.799551 +expl/Actions Max 2.42802 +expl/Actions Min -2.28374 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 549277 +eval/num paths total 593 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0885159 +eval/Actions Std 0.740844 +eval/Actions Max 0.999963 +eval/Actions Min -0.999907 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.30329e-05 +time/evaluation sampling (s) 4.77277 +time/exploration sampling (s) 6.51576 +time/logging (s) 0.00802908 +time/saving (s) 0.0137832 +time/training (s) 19.1287 +time/epoch (s) 30.4391 +time/total (s) 14432.6 +Epoch -410 +------------------------------ ---------------- +2022-05-15 22:03:31.652668 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -409 finished +------------------------------ ---------------- +epoch -409 +replay_buffer/size 999047 +trainer/num train calls 592000 +trainer/QF1 Loss 1.34145 +trainer/QF2 Loss 1.44787 +trainer/Policy Loss 3.73872 +trainer/Q1 Predictions Mean -75.0226 +trainer/Q1 Predictions Std 17.9046 +trainer/Q1 Predictions Max -0.797506 +trainer/Q1 Predictions Min -87.5913 +trainer/Q2 Predictions Mean -74.9873 +trainer/Q2 Predictions Std 17.943 +trainer/Q2 Predictions Max -0.763449 +trainer/Q2 Predictions Min -87.6415 +trainer/Q Targets Mean -74.4356 +trainer/Q Targets Std 17.4312 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1313 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0221058 +trainer/policy/mean Std 0.730415 +trainer/policy/mean Max 0.999834 +trainer/policy/mean Min -0.999794 +trainer/policy/std Mean 0.416575 +trainer/policy/std Std 0.0214777 +trainer/policy/std Max 0.440776 +trainer/policy/std Min 0.382641 +trainer/Advantage Weights Mean 0.765727 +trainer/Advantage Weights Std 7.34118 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.76913e-14 +trainer/Advantage Score Mean -0.757375 +trainer/Advantage Score Std 0.594753 +trainer/Advantage Score Max 1.23322 +trainer/Advantage Score Min -3.09093 +trainer/V1 Predictions Mean -74.2148 +trainer/V1 Predictions Std 17.7477 +trainer/V1 Predictions Max -0.862771 +trainer/V1 Predictions Min -87.057 +trainer/VF Loss 0.098337 +expl/num steps total 592000 +expl/num paths total 759 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0498595 +expl/Actions Std 0.852093 +expl/Actions Max 2.52229 +expl/Actions Min -2.45853 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 550277 +eval/num paths total 594 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0939365 +eval/Actions Std 0.691413 +eval/Actions Max 0.999849 +eval/Actions Min -0.998464 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.38809e-05 +time/evaluation sampling (s) 4.63927 +time/exploration sampling (s) 6.57542 +time/logging (s) 0.0129128 +time/saving (s) 0.0166295 +time/training (s) 19.5053 +time/epoch (s) 30.7496 +time/total (s) 14463.4 +Epoch -409 +------------------------------ ---------------- +2022-05-15 22:04:02.585463 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -408 finished +------------------------------ ---------------- +epoch -408 +replay_buffer/size 999047 +trainer/num train calls 593000 +trainer/QF1 Loss 0.570391 +trainer/QF2 Loss 0.65407 +trainer/Policy Loss 28.8932 +trainer/Q1 Predictions Mean -73.1088 +trainer/Q1 Predictions Std 18.4199 +trainer/Q1 Predictions Max -0.38715 +trainer/Q1 Predictions Min -87.3191 +trainer/Q2 Predictions Mean -73.1716 +trainer/Q2 Predictions Std 18.4174 +trainer/Q2 Predictions Max -0.670552 +trainer/Q2 Predictions Min -87.5138 +trainer/Q Targets Mean -73.3062 +trainer/Q Targets Std 18.4284 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4406 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00600109 +trainer/policy/mean Std 0.728925 +trainer/policy/mean Max 0.999953 +trainer/policy/mean Min -0.9997 +trainer/policy/std Mean 0.41623 +trainer/policy/std Std 0.0208073 +trainer/policy/std Max 0.436279 +trainer/policy/std Min 0.383298 +trainer/Advantage Weights Mean 5.19789 +trainer/Advantage Weights Std 18.3227 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.95323e-16 +trainer/Advantage Score Mean -0.320287 +trainer/Advantage Score Std 0.58657 +trainer/Advantage Score Max 1.28835 +trainer/Advantage Score Min -3.61719 +trainer/V1 Predictions Mean -72.9625 +trainer/V1 Predictions Std 18.5758 +trainer/V1 Predictions Max 0.181124 +trainer/V1 Predictions Min -87.3391 +trainer/VF Loss 0.0676648 +expl/num steps total 593000 +expl/num paths total 760 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00947012 +expl/Actions Std 0.819743 +expl/Actions Max 2.06477 +expl/Actions Min -2.57629 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 551115 +eval/num paths total 595 +eval/path length Mean 838 +eval/path length Std 0 +eval/path length Max 838 +eval/path length Min 838 +eval/Rewards Mean 0.00119332 +eval/Rewards Std 0.0345238 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00751784 +eval/Actions Std 0.759552 +eval/Actions Max 0.999796 +eval/Actions Min -0.999646 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.08918e-05 +time/evaluation sampling (s) 4.69472 +time/exploration sampling (s) 7.03077 +time/logging (s) 0.00712016 +time/saving (s) 0.0105258 +time/training (s) 19.1663 +time/epoch (s) 30.9095 +time/total (s) 14494.3 +Epoch -408 +------------------------------ ---------------- +2022-05-15 22:04:32.589240 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -407 finished +------------------------------ ---------------- +epoch -407 +replay_buffer/size 999047 +trainer/num train calls 594000 +trainer/QF1 Loss 0.716404 +trainer/QF2 Loss 0.721829 +trainer/Policy Loss 6.29882 +trainer/Q1 Predictions Mean -74.4976 +trainer/Q1 Predictions Std 17.6352 +trainer/Q1 Predictions Max -0.30465 +trainer/Q1 Predictions Min -87.1617 +trainer/Q2 Predictions Mean -74.5447 +trainer/Q2 Predictions Std 17.6225 +trainer/Q2 Predictions Max 0.909234 +trainer/Q2 Predictions Min -87.3864 +trainer/Q Targets Mean -74.6768 +trainer/Q Targets Std 17.4812 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6526 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000990877 +trainer/policy/mean Std 0.725323 +trainer/policy/mean Max 0.999567 +trainer/policy/mean Min -0.998725 +trainer/policy/std Mean 0.416113 +trainer/policy/std Std 0.021562 +trainer/policy/std Max 0.440229 +trainer/policy/std Min 0.382753 +trainer/Advantage Weights Mean 1.51533 +trainer/Advantage Weights Std 9.36069 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.2347e-11 +trainer/Advantage Score Mean -0.475141 +trainer/Advantage Score Std 0.507165 +trainer/Advantage Score Max 1.37605 +trainer/Advantage Score Min -2.51176 +trainer/V1 Predictions Mean -74.4062 +trainer/V1 Predictions Std 17.6493 +trainer/V1 Predictions Max 0.517513 +trainer/V1 Predictions Min -87.4347 +trainer/VF Loss 0.0569452 +expl/num steps total 594000 +expl/num paths total 761 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0846458 +expl/Actions Std 0.852049 +expl/Actions Max 2.23234 +expl/Actions Min -2.33122 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 552115 +eval/num paths total 596 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.181264 +eval/Actions Std 0.744762 +eval/Actions Max 0.999891 +eval/Actions Min -0.999856 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.33649e-05 +time/evaluation sampling (s) 4.44377 +time/exploration sampling (s) 6.53932 +time/logging (s) 0.011276 +time/saving (s) 0.0166772 +time/training (s) 18.9862 +time/epoch (s) 29.9972 +time/total (s) 14524.3 +Epoch -407 +------------------------------ ---------------- +2022-05-15 22:05:02.967118 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -406 finished +------------------------------ ---------------- +epoch -406 +replay_buffer/size 999047 +trainer/num train calls 595000 +trainer/QF1 Loss 1.05193 +trainer/QF2 Loss 1.00298 +trainer/Policy Loss 12.7866 +trainer/Q1 Predictions Mean -71.0063 +trainer/Q1 Predictions Std 21.3687 +trainer/Q1 Predictions Max -0.540852 +trainer/Q1 Predictions Min -87.1958 +trainer/Q2 Predictions Mean -71.1047 +trainer/Q2 Predictions Std 21.4013 +trainer/Q2 Predictions Max -0.262933 +trainer/Q2 Predictions Min -86.795 +trainer/Q Targets Mean -70.8653 +trainer/Q Targets Std 21.381 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5414 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0112028 +trainer/policy/mean Std 0.727432 +trainer/policy/mean Max 0.999651 +trainer/policy/mean Min -0.999589 +trainer/policy/std Mean 0.417167 +trainer/policy/std Std 0.0197773 +trainer/policy/std Max 0.439181 +trainer/policy/std Min 0.387219 +trainer/Advantage Weights Mean 4.39464 +trainer/Advantage Weights Std 18.1626 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.72487e-23 +trainer/Advantage Score Mean -0.561651 +trainer/Advantage Score Std 0.836209 +trainer/Advantage Score Max 2.75245 +trainer/Advantage Score Min -5.0915 +trainer/V1 Predictions Mean -70.5786 +trainer/V1 Predictions Std 21.6926 +trainer/V1 Predictions Max -0.681111 +trainer/V1 Predictions Min -86.7541 +trainer/VF Loss 0.147078 +expl/num steps total 595000 +expl/num paths total 763 +expl/path length Mean 500 +expl/path length Std 175 +expl/path length Max 675 +expl/path length Min 325 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00864557 +expl/Actions Std 0.826048 +expl/Actions Max 2.24718 +expl/Actions Min -2.38799 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 553115 +eval/num paths total 597 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.142239 +eval/Actions Std 0.573989 +eval/Actions Max 0.999845 +eval/Actions Min -0.998221 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.06683e-05 +time/evaluation sampling (s) 4.67131 +time/exploration sampling (s) 6.64236 +time/logging (s) 0.0127701 +time/saving (s) 0.0189028 +time/training (s) 19.0192 +time/epoch (s) 30.3645 +time/total (s) 14554.6 +Epoch -406 +------------------------------ ---------------- +2022-05-15 22:05:33.763841 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -405 finished +------------------------------ ---------------- +epoch -405 +replay_buffer/size 999047 +trainer/num train calls 596000 +trainer/QF1 Loss 1.04186 +trainer/QF2 Loss 0.94302 +trainer/Policy Loss 29.7467 +trainer/Q1 Predictions Mean -72.6479 +trainer/Q1 Predictions Std 19.078 +trainer/Q1 Predictions Max -0.990683 +trainer/Q1 Predictions Min -86.7497 +trainer/Q2 Predictions Mean -72.7299 +trainer/Q2 Predictions Std 19.1322 +trainer/Q2 Predictions Max 0.0324778 +trainer/Q2 Predictions Min -86.7733 +trainer/Q Targets Mean -72.9427 +trainer/Q Targets Std 18.5776 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.775 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00687373 +trainer/policy/mean Std 0.729974 +trainer/policy/mean Max 0.999885 +trainer/policy/mean Min -0.999416 +trainer/policy/std Mean 0.41592 +trainer/policy/std Std 0.0211816 +trainer/policy/std Max 0.438753 +trainer/policy/std Min 0.381509 +trainer/Advantage Weights Mean 5.7473 +trainer/Advantage Weights Std 20.3999 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.83501e-12 +trainer/Advantage Score Mean -0.361425 +trainer/Advantage Score Std 0.545855 +trainer/Advantage Score Max 2.20136 +trainer/Advantage Score Min -2.62868 +trainer/V1 Predictions Mean -72.6831 +trainer/V1 Predictions Std 18.766 +trainer/V1 Predictions Max -0.546797 +trainer/V1 Predictions Min -86.6406 +trainer/VF Loss 0.0855903 +expl/num steps total 596000 +expl/num paths total 765 +expl/path length Mean 500 +expl/path length Std 350 +expl/path length Max 850 +expl/path length Min 150 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0386906 +expl/Actions Std 0.8324 +expl/Actions Max 2.45175 +expl/Actions Min -2.39599 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 553983 +eval/num paths total 598 +eval/path length Mean 868 +eval/path length Std 0 +eval/path length Max 868 +eval/path length Min 868 +eval/Rewards Mean 0.00115207 +eval/Rewards Std 0.0339227 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.0126585 +eval/Actions Std 0.727461 +eval/Actions Max 0.999933 +eval/Actions Min -0.999259 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.99606e-06 +time/evaluation sampling (s) 5.05635 +time/exploration sampling (s) 6.29897 +time/logging (s) 0.00828001 +time/saving (s) 0.012393 +time/training (s) 19.3951 +time/epoch (s) 30.7711 +time/total (s) 14585.4 +Epoch -405 +------------------------------ ---------------- +2022-05-15 22:06:03.943702 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -404 finished +------------------------------ ---------------- +epoch -404 +replay_buffer/size 999047 +trainer/num train calls 597000 +trainer/QF1 Loss 0.822357 +trainer/QF2 Loss 0.65427 +trainer/Policy Loss 14.8249 +trainer/Q1 Predictions Mean -73.7049 +trainer/Q1 Predictions Std 18.5587 +trainer/Q1 Predictions Max -1.43241 +trainer/Q1 Predictions Min -86.7349 +trainer/Q2 Predictions Mean -73.6908 +trainer/Q2 Predictions Std 18.6459 +trainer/Q2 Predictions Max -1.14942 +trainer/Q2 Predictions Min -86.8095 +trainer/Q Targets Mean -73.5629 +trainer/Q Targets Std 18.5916 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5755 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0254628 +trainer/policy/mean Std 0.721827 +trainer/policy/mean Max 0.999194 +trainer/policy/mean Min -0.999111 +trainer/policy/std Mean 0.417661 +trainer/policy/std Std 0.020684 +trainer/policy/std Max 0.440024 +trainer/policy/std Min 0.384627 +trainer/Advantage Weights Mean 4.32056 +trainer/Advantage Weights Std 18.1389 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.73407e-19 +trainer/Advantage Score Mean -0.392366 +trainer/Advantage Score Std 0.669994 +trainer/Advantage Score Max 1.14904 +trainer/Advantage Score Min -4.31986 +trainer/V1 Predictions Mean -73.3011 +trainer/V1 Predictions Std 18.7737 +trainer/V1 Predictions Max -0.540339 +trainer/V1 Predictions Min -86.4043 +trainer/VF Loss 0.078476 +expl/num steps total 597000 +expl/num paths total 767 +expl/path length Mean 500 +expl/path length Std 208 +expl/path length Max 708 +expl/path length Min 292 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0370987 +expl/Actions Std 0.816665 +expl/Actions Max 2.52486 +expl/Actions Min -2.34045 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 554983 +eval/num paths total 599 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0255762 +eval/Actions Std 0.757356 +eval/Actions Max 0.999369 +eval/Actions Min -0.998964 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.39491e-06 +time/evaluation sampling (s) 4.92269 +time/exploration sampling (s) 5.76419 +time/logging (s) 0.00898349 +time/saving (s) 0.0121087 +time/training (s) 19.4607 +time/epoch (s) 30.1687 +time/total (s) 14615.6 +Epoch -404 +------------------------------ ---------------- +2022-05-15 22:06:34.502796 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -403 finished +------------------------------ ---------------- +epoch -403 +replay_buffer/size 999047 +trainer/num train calls 598000 +trainer/QF1 Loss 1.71434 +trainer/QF2 Loss 1.96041 +trainer/Policy Loss 89.1248 +trainer/Q1 Predictions Mean -73.3498 +trainer/Q1 Predictions Std 19.2857 +trainer/Q1 Predictions Max -1.55843 +trainer/Q1 Predictions Min -87.0526 +trainer/Q2 Predictions Mean -73.2476 +trainer/Q2 Predictions Std 19.2661 +trainer/Q2 Predictions Max -1.7177 +trainer/Q2 Predictions Min -87.2933 +trainer/Q Targets Mean -74.1817 +trainer/Q Targets Std 19.182 +trainer/Q Targets Max -2.64468 +trainer/Q Targets Min -87.6653 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0230935 +trainer/policy/mean Std 0.718062 +trainer/policy/mean Max 0.999798 +trainer/policy/mean Min -0.998392 +trainer/policy/std Mean 0.4153 +trainer/policy/std Std 0.0193885 +trainer/policy/std Max 0.434946 +trainer/policy/std Min 0.383548 +trainer/Advantage Weights Mean 18.9855 +trainer/Advantage Weights Std 31.509 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.45147e-12 +trainer/Advantage Score Mean 0.021019 +trainer/Advantage Score Std 0.504724 +trainer/Advantage Score Max 1.61955 +trainer/Advantage Score Min -2.72584 +trainer/V1 Predictions Mean -73.9365 +trainer/V1 Predictions Std 19.4779 +trainer/V1 Predictions Max -1.92272 +trainer/V1 Predictions Min -87.6988 +trainer/VF Loss 0.0997733 +expl/num steps total 598000 +expl/num paths total 768 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0334839 +expl/Actions Std 0.817991 +expl/Actions Max 2.16894 +expl/Actions Min -2.214 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 555983 +eval/num paths total 600 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.179822 +eval/Actions Std 0.660568 +eval/Actions Max 0.999631 +eval/Actions Min -0.999243 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.55184e-06 +time/evaluation sampling (s) 4.65708 +time/exploration sampling (s) 6.30534 +time/logging (s) 0.0117491 +time/saving (s) 0.0155775 +time/training (s) 19.5598 +time/epoch (s) 30.5495 +time/total (s) 14646.1 +Epoch -403 +------------------------------ ---------------- +2022-05-15 22:07:04.499549 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -402 finished +------------------------------ ---------------- +epoch -402 +replay_buffer/size 999047 +trainer/num train calls 599000 +trainer/QF1 Loss 0.872816 +trainer/QF2 Loss 0.971742 +trainer/Policy Loss 18.3896 +trainer/Q1 Predictions Mean -73.4967 +trainer/Q1 Predictions Std 17.9595 +trainer/Q1 Predictions Max -0.400997 +trainer/Q1 Predictions Min -87.946 +trainer/Q2 Predictions Mean -73.5168 +trainer/Q2 Predictions Std 17.9292 +trainer/Q2 Predictions Max -0.26929 +trainer/Q2 Predictions Min -87.4512 +trainer/Q Targets Mean -73.4906 +trainer/Q Targets Std 18.4159 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6994 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00661621 +trainer/policy/mean Std 0.72174 +trainer/policy/mean Max 0.999839 +trainer/policy/mean Min -0.998245 +trainer/policy/std Mean 0.415856 +trainer/policy/std Std 0.0191697 +trainer/policy/std Max 0.436359 +trainer/policy/std Min 0.386128 +trainer/Advantage Weights Mean 5.98208 +trainer/Advantage Weights Std 20.8132 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.49423e-19 +trainer/Advantage Score Mean -0.364403 +trainer/Advantage Score Std 0.797188 +trainer/Advantage Score Max 1.3537 +trainer/Advantage Score Min -4.33475 +trainer/V1 Predictions Mean -73.2389 +trainer/V1 Predictions Std 18.5543 +trainer/V1 Predictions Max 1.15657 +trainer/V1 Predictions Min -87.3398 +trainer/VF Loss 0.106995 +expl/num steps total 599000 +expl/num paths total 770 +expl/path length Mean 500 +expl/path length Std 400 +expl/path length Max 900 +expl/path length Min 100 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0427769 +expl/Actions Std 0.821113 +expl/Actions Max 2.25445 +expl/Actions Min -2.4541 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 556983 +eval/num paths total 601 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0359002 +eval/Actions Std 0.771939 +eval/Actions Max 0.999186 +eval/Actions Min -0.998987 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.1012e-05 +time/evaluation sampling (s) 4.49619 +time/exploration sampling (s) 5.88462 +time/logging (s) 0.0119688 +time/saving (s) 0.0163385 +time/training (s) 19.5671 +time/epoch (s) 29.9762 +time/total (s) 14676.1 +Epoch -402 +------------------------------ ---------------- +2022-05-15 22:07:34.510336 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -401 finished +------------------------------ ---------------- +epoch -401 +replay_buffer/size 999047 +trainer/num train calls 600000 +trainer/QF1 Loss 1.01433 +trainer/QF2 Loss 0.751303 +trainer/Policy Loss 19.7255 +trainer/Q1 Predictions Mean -72.313 +trainer/Q1 Predictions Std 20.5476 +trainer/Q1 Predictions Max -0.932745 +trainer/Q1 Predictions Min -87.4245 +trainer/Q2 Predictions Mean -72.3453 +trainer/Q2 Predictions Std 20.591 +trainer/Q2 Predictions Max 0.103312 +trainer/Q2 Predictions Min -87.6449 +trainer/Q Targets Mean -72.6286 +trainer/Q Targets Std 20.4732 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7468 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0105988 +trainer/policy/mean Std 0.712217 +trainer/policy/mean Max 0.998768 +trainer/policy/mean Min -0.998887 +trainer/policy/std Mean 0.415688 +trainer/policy/std Std 0.0187617 +trainer/policy/std Max 0.434207 +trainer/policy/std Min 0.383759 +trainer/Advantage Weights Mean 5.11166 +trainer/Advantage Weights Std 18.5598 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.91137e-15 +trainer/Advantage Score Mean -0.293704 +trainer/Advantage Score Std 0.569306 +trainer/Advantage Score Max 1.37405 +trainer/Advantage Score Min -3.34702 +trainer/V1 Predictions Mean -72.3372 +trainer/V1 Predictions Std 20.5456 +trainer/V1 Predictions Max -0.3761 +trainer/V1 Predictions Min -87.6253 +trainer/VF Loss 0.065121 +expl/num steps total 600000 +expl/num paths total 772 +expl/path length Mean 500 +expl/path length Std 392 +expl/path length Max 892 +expl/path length Min 108 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0136572 +expl/Actions Std 0.811897 +expl/Actions Max 2.40036 +expl/Actions Min -2.41016 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 557502 +eval/num paths total 602 +eval/path length Mean 519 +eval/path length Std 0 +eval/path length Max 519 +eval/path length Min 519 +eval/Rewards Mean 0.00192678 +eval/Rewards Std 0.0438528 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0418086 +eval/Actions Std 0.733411 +eval/Actions Max 0.999681 +eval/Actions Min -0.998706 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.34939e-05 +time/evaluation sampling (s) 4.41534 +time/exploration sampling (s) 6.20102 +time/logging (s) 0.00931966 +time/saving (s) 0.015211 +time/training (s) 19.3508 +time/epoch (s) 29.9917 +time/total (s) 14706.1 +Epoch -401 +------------------------------ ---------------- +2022-05-15 22:08:04.667404 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -400 finished +------------------------------ ---------------- +epoch -400 +replay_buffer/size 999047 +trainer/num train calls 601000 +trainer/QF1 Loss 1.00116 +trainer/QF2 Loss 0.955843 +trainer/Policy Loss 20.9938 +trainer/Q1 Predictions Mean -72.2059 +trainer/Q1 Predictions Std 20.5238 +trainer/Q1 Predictions Max -0.277339 +trainer/Q1 Predictions Min -86.9074 +trainer/Q2 Predictions Mean -72.2284 +trainer/Q2 Predictions Std 20.6192 +trainer/Q2 Predictions Max -0.43637 +trainer/Q2 Predictions Min -87.1939 +trainer/Q Targets Mean -72.3449 +trainer/Q Targets Std 20.2746 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3768 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0233055 +trainer/policy/mean Std 0.723968 +trainer/policy/mean Max 0.999747 +trainer/policy/mean Min -0.999773 +trainer/policy/std Mean 0.416331 +trainer/policy/std Std 0.0206834 +trainer/policy/std Max 0.437016 +trainer/policy/std Min 0.382134 +trainer/Advantage Weights Mean 3.92883 +trainer/Advantage Weights Std 16.9751 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.28904e-19 +trainer/Advantage Score Mean -0.326734 +trainer/Advantage Score Std 0.596489 +trainer/Advantage Score Max 2.17609 +trainer/Advantage Score Min -4.20835 +trainer/V1 Predictions Mean -71.9835 +trainer/V1 Predictions Std 20.5402 +trainer/V1 Predictions Max 0.519885 +trainer/V1 Predictions Min -87.2384 +trainer/VF Loss 0.0925245 +expl/num steps total 601000 +expl/num paths total 774 +expl/path length Mean 500 +expl/path length Std 217 +expl/path length Max 717 +expl/path length Min 283 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0144033 +expl/Actions Std 0.835343 +expl/Actions Max 2.42433 +expl/Actions Min -2.30511 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 558502 +eval/num paths total 603 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00175963 +eval/Actions Std 0.782767 +eval/Actions Max 0.999027 +eval/Actions Min -0.999562 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.08709e-05 +time/evaluation sampling (s) 4.22129 +time/exploration sampling (s) 7.10753 +time/logging (s) 0.0131535 +time/saving (s) 0.0319795 +time/training (s) 18.7716 +time/epoch (s) 30.1455 +time/total (s) 14736.3 +Epoch -400 +------------------------------ ---------------- +2022-05-15 22:08:35.058378 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -399 finished +------------------------------ ---------------- +epoch -399 +replay_buffer/size 999047 +trainer/num train calls 602000 +trainer/QF1 Loss 1.06804 +trainer/QF2 Loss 0.978434 +trainer/Policy Loss 17.862 +trainer/Q1 Predictions Mean -74.8961 +trainer/Q1 Predictions Std 16.6176 +trainer/Q1 Predictions Max -1.34207 +trainer/Q1 Predictions Min -87.7605 +trainer/Q2 Predictions Mean -74.9051 +trainer/Q2 Predictions Std 16.5676 +trainer/Q2 Predictions Max -0.807571 +trainer/Q2 Predictions Min -87.845 +trainer/Q Targets Mean -74.9914 +trainer/Q Targets Std 16.5737 +trainer/Q Targets Max -1.56748 +trainer/Q Targets Min -87.8139 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00237295 +trainer/policy/mean Std 0.735643 +trainer/policy/mean Max 0.999533 +trainer/policy/mean Min -0.998904 +trainer/policy/std Mean 0.416154 +trainer/policy/std Std 0.0191161 +trainer/policy/std Max 0.440357 +trainer/policy/std Min 0.385524 +trainer/Advantage Weights Mean 3.61742 +trainer/Advantage Weights Std 13.6482 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.03613e-21 +trainer/Advantage Score Mean -0.296292 +trainer/Advantage Score Std 0.646172 +trainer/Advantage Score Max 2.68365 +trainer/Advantage Score Min -4.83188 +trainer/V1 Predictions Mean -74.816 +trainer/V1 Predictions Std 16.782 +trainer/V1 Predictions Max -0.529868 +trainer/V1 Predictions Min -87.6911 +trainer/VF Loss 0.083651 +expl/num steps total 602000 +expl/num paths total 776 +expl/path length Mean 500 +expl/path length Std 245 +expl/path length Max 745 +expl/path length Min 255 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.011095 +expl/Actions Std 0.815978 +expl/Actions Max 2.82128 +expl/Actions Min -2.44235 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 559159 +eval/num paths total 604 +eval/path length Mean 657 +eval/path length Std 0 +eval/path length Max 657 +eval/path length Min 657 +eval/Rewards Mean 0.00152207 +eval/Rewards Std 0.038984 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0610162 +eval/Actions Std 0.747902 +eval/Actions Max 0.999723 +eval/Actions Min -0.998972 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.19251e-05 +time/evaluation sampling (s) 4.41414 +time/exploration sampling (s) 6.58692 +time/logging (s) 0.010019 +time/saving (s) 0.0156717 +time/training (s) 19.3422 +time/epoch (s) 30.3689 +time/total (s) 14766.7 +Epoch -399 +------------------------------ ---------------- +2022-05-15 22:09:06.327679 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -398 finished +------------------------------ ---------------- +epoch -398 +replay_buffer/size 999047 +trainer/num train calls 603000 +trainer/QF1 Loss 0.971008 +trainer/QF2 Loss 0.848729 +trainer/Policy Loss 13.0684 +trainer/Q1 Predictions Mean -74.4971 +trainer/Q1 Predictions Std 17.8232 +trainer/Q1 Predictions Max -1.02089 +trainer/Q1 Predictions Min -87.4869 +trainer/Q2 Predictions Mean -74.508 +trainer/Q2 Predictions Std 17.8514 +trainer/Q2 Predictions Max -0.620061 +trainer/Q2 Predictions Min -87.0484 +trainer/Q Targets Mean -74.0364 +trainer/Q Targets Std 17.7313 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5448 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00377366 +trainer/policy/mean Std 0.723372 +trainer/policy/mean Max 0.999915 +trainer/policy/mean Min -0.998728 +trainer/policy/std Mean 0.414548 +trainer/policy/std Std 0.0196244 +trainer/policy/std Max 0.435056 +trainer/policy/std Min 0.385509 +trainer/Advantage Weights Mean 3.26849 +trainer/Advantage Weights Std 16.469 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.16801e-19 +trainer/Advantage Score Mean -0.467108 +trainer/Advantage Score Std 0.55225 +trainer/Advantage Score Max 1.23534 +trainer/Advantage Score Min -4.19297 +trainer/V1 Predictions Mean -73.7563 +trainer/V1 Predictions Std 17.8889 +trainer/V1 Predictions Max -1.84425 +trainer/V1 Predictions Min -86.5445 +trainer/VF Loss 0.0646674 +expl/num steps total 603000 +expl/num paths total 778 +expl/path length Mean 500 +expl/path length Std 69 +expl/path length Max 569 +expl/path length Min 431 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0275779 +expl/Actions Std 0.826664 +expl/Actions Max 2.7447 +expl/Actions Min -2.2823 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 560159 +eval/num paths total 605 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.092153 +eval/Actions Std 0.784795 +eval/Actions Max 0.999622 +eval/Actions Min -0.999611 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.01589e-05 +time/evaluation sampling (s) 4.6568 +time/exploration sampling (s) 7.16549 +time/logging (s) 0.0118402 +time/saving (s) 0.0151772 +time/training (s) 19.4086 +time/epoch (s) 31.258 +time/total (s) 14797.9 +Epoch -398 +------------------------------ ---------------- +2022-05-15 22:09:36.912101 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -397 finished +------------------------------ ---------------- +epoch -397 +replay_buffer/size 999047 +trainer/num train calls 604000 +trainer/QF1 Loss 1.31141 +trainer/QF2 Loss 1.3865 +trainer/Policy Loss 30.5845 +trainer/Q1 Predictions Mean -72.8911 +trainer/Q1 Predictions Std 18.8044 +trainer/Q1 Predictions Max -0.988863 +trainer/Q1 Predictions Min -87.502 +trainer/Q2 Predictions Mean -72.9918 +trainer/Q2 Predictions Std 18.793 +trainer/Q2 Predictions Max -1.56202 +trainer/Q2 Predictions Min -87.801 +trainer/Q Targets Mean -73.0526 +trainer/Q Targets Std 19.1464 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7395 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0156629 +trainer/policy/mean Std 0.736889 +trainer/policy/mean Max 0.999249 +trainer/policy/mean Min -0.999584 +trainer/policy/std Mean 0.413209 +trainer/policy/std Std 0.021182 +trainer/policy/std Max 0.435735 +trainer/policy/std Min 0.380067 +trainer/Advantage Weights Mean 4.47801 +trainer/Advantage Weights Std 16.8512 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.25747e-20 +trainer/Advantage Score Mean -0.44108 +trainer/Advantage Score Std 0.706243 +trainer/Advantage Score Max 1.6115 +trainer/Advantage Score Min -4.58226 +trainer/V1 Predictions Mean -72.751 +trainer/V1 Predictions Std 19.3297 +trainer/V1 Predictions Max -0.305285 +trainer/V1 Predictions Min -87.716 +trainer/VF Loss 0.0913221 +expl/num steps total 604000 +expl/num paths total 779 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0618097 +expl/Actions Std 0.83366 +expl/Actions Max 2.35111 +expl/Actions Min -2.12488 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 560912 +eval/num paths total 606 +eval/path length Mean 753 +eval/path length Std 0 +eval/path length Max 753 +eval/path length Min 753 +eval/Rewards Mean 0.00132802 +eval/Rewards Std 0.0364178 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0341196 +eval/Actions Std 0.747617 +eval/Actions Max 0.999801 +eval/Actions Min -0.999299 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 9.68901e-06 +time/evaluation sampling (s) 4.7132 +time/exploration sampling (s) 6.65366 +time/logging (s) 0.0111991 +time/saving (s) 0.0149934 +time/training (s) 19.1748 +time/epoch (s) 30.5678 +time/total (s) 14828.5 +Epoch -397 +------------------------------ ---------------- +2022-05-15 22:10:07.535694 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -396 finished +------------------------------ ---------------- +epoch -396 +replay_buffer/size 999047 +trainer/num train calls 605000 +trainer/QF1 Loss 0.992499 +trainer/QF2 Loss 0.809969 +trainer/Policy Loss 18.2251 +trainer/Q1 Predictions Mean -74.7345 +trainer/Q1 Predictions Std 17.6607 +trainer/Q1 Predictions Max -1.25385 +trainer/Q1 Predictions Min -88.0391 +trainer/Q2 Predictions Mean -74.6152 +trainer/Q2 Predictions Std 17.5894 +trainer/Q2 Predictions Max -1.32918 +trainer/Q2 Predictions Min -88.2821 +trainer/Q Targets Mean -74.3277 +trainer/Q Targets Std 17.5103 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6036 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0207151 +trainer/policy/mean Std 0.721483 +trainer/policy/mean Max 0.999589 +trainer/policy/mean Min -0.997224 +trainer/policy/std Mean 0.414195 +trainer/policy/std Std 0.0199538 +trainer/policy/std Max 0.433747 +trainer/policy/std Min 0.383037 +trainer/Advantage Weights Mean 5.1024 +trainer/Advantage Weights Std 19.5552 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.22944e-15 +trainer/Advantage Score Mean -0.381866 +trainer/Advantage Score Std 0.556571 +trainer/Advantage Score Max 1.79888 +trainer/Advantage Score Min -3.43322 +trainer/V1 Predictions Mean -74.0848 +trainer/V1 Predictions Std 17.7042 +trainer/V1 Predictions Max -0.27577 +trainer/V1 Predictions Min -87.5144 +trainer/VF Loss 0.0746836 +expl/num steps total 605000 +expl/num paths total 780 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0906672 +expl/Actions Std 0.836433 +expl/Actions Max 2.72536 +expl/Actions Min -2.239 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 561912 +eval/num paths total 607 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0516355 +eval/Actions Std 0.774396 +eval/Actions Max 0.999587 +eval/Actions Min -0.999508 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.19583e-06 +time/evaluation sampling (s) 4.52181 +time/exploration sampling (s) 6.79797 +time/logging (s) 0.00819501 +time/saving (s) 0.0117501 +time/training (s) 19.2687 +time/epoch (s) 30.6084 +time/total (s) 14859.1 +Epoch -396 +------------------------------ ---------------- +2022-05-15 22:10:37.879191 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -395 finished +------------------------------ ---------------- +epoch -395 +replay_buffer/size 999047 +trainer/num train calls 606000 +trainer/QF1 Loss 0.98855 +trainer/QF2 Loss 1.05512 +trainer/Policy Loss 6.87036 +trainer/Q1 Predictions Mean -73.9511 +trainer/Q1 Predictions Std 17.8851 +trainer/Q1 Predictions Max -1.06883 +trainer/Q1 Predictions Min -87.6816 +trainer/Q2 Predictions Mean -73.8846 +trainer/Q2 Predictions Std 17.7883 +trainer/Q2 Predictions Max -1.21488 +trainer/Q2 Predictions Min -87.6262 +trainer/Q Targets Mean -73.8838 +trainer/Q Targets Std 18.1285 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4253 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00943602 +trainer/policy/mean Std 0.732968 +trainer/policy/mean Max 0.999302 +trainer/policy/mean Min -0.999778 +trainer/policy/std Mean 0.413816 +trainer/policy/std Std 0.0203701 +trainer/policy/std Max 0.434125 +trainer/policy/std Min 0.380256 +trainer/Advantage Weights Mean 1.87192 +trainer/Advantage Weights Std 11.2913 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.37484e-13 +trainer/Advantage Score Mean -0.566368 +trainer/Advantage Score Std 0.558139 +trainer/Advantage Score Max 1.23703 +trainer/Advantage Score Min -2.84577 +trainer/V1 Predictions Mean -73.6829 +trainer/V1 Predictions Std 18.0187 +trainer/V1 Predictions Max -0.859892 +trainer/V1 Predictions Min -87.301 +trainer/VF Loss 0.0739694 +expl/num steps total 606000 +expl/num paths total 781 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00219185 +expl/Actions Std 0.81174 +expl/Actions Max 2.33445 +expl/Actions Min -2.5098 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 562796 +eval/num paths total 608 +eval/path length Mean 884 +eval/path length Std 0 +eval/path length Max 884 +eval/path length Min 884 +eval/Rewards Mean 0.00113122 +eval/Rewards Std 0.0336146 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0368209 +eval/Actions Std 0.72536 +eval/Actions Max 0.999988 +eval/Actions Min -0.999236 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.35549e-05 +time/evaluation sampling (s) 4.70324 +time/exploration sampling (s) 6.75066 +time/logging (s) 0.0125548 +time/saving (s) 0.0162628 +time/training (s) 18.8521 +time/epoch (s) 30.3348 +time/total (s) 14889.5 +Epoch -395 +------------------------------ ---------------- +2022-05-15 22:11:08.448136 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -394 finished +------------------------------ ---------------- +epoch -394 +replay_buffer/size 999047 +trainer/num train calls 607000 +trainer/QF1 Loss 0.620838 +trainer/QF2 Loss 0.693619 +trainer/Policy Loss 11.6856 +trainer/Q1 Predictions Mean -74.0812 +trainer/Q1 Predictions Std 17.5797 +trainer/Q1 Predictions Max -1.52202 +trainer/Q1 Predictions Min -87.3317 +trainer/Q2 Predictions Mean -74.1248 +trainer/Q2 Predictions Std 17.5406 +trainer/Q2 Predictions Max -1.75257 +trainer/Q2 Predictions Min -87.6959 +trainer/Q Targets Mean -74.1081 +trainer/Q Targets Std 17.4684 +trainer/Q Targets Max -1.45154 +trainer/Q Targets Min -87.5594 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00157442 +trainer/policy/mean Std 0.728472 +trainer/policy/mean Max 0.998982 +trainer/policy/mean Min -0.999631 +trainer/policy/std Mean 0.413506 +trainer/policy/std Std 0.021043 +trainer/policy/std Max 0.435028 +trainer/policy/std Min 0.378414 +trainer/Advantage Weights Mean 2.66974 +trainer/Advantage Weights Std 13.3991 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.86463e-16 +trainer/Advantage Score Mean -0.436755 +trainer/Advantage Score Std 0.605786 +trainer/Advantage Score Max 1.02225 +trainer/Advantage Score Min -3.62183 +trainer/V1 Predictions Mean -73.7827 +trainer/V1 Predictions Std 17.7139 +trainer/V1 Predictions Max -0.727125 +trainer/V1 Predictions Min -87.5008 +trainer/VF Loss 0.0643352 +expl/num steps total 607000 +expl/num paths total 783 +expl/path length Mean 500 +expl/path length Std 380 +expl/path length Max 880 +expl/path length Min 120 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0102082 +expl/Actions Std 0.841004 +expl/Actions Max 2.41231 +expl/Actions Min -2.34886 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 563356 +eval/num paths total 609 +eval/path length Mean 560 +eval/path length Std 0 +eval/path length Max 560 +eval/path length Min 560 +eval/Rewards Mean 0.00178571 +eval/Rewards Std 0.04222 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0065417 +eval/Actions Std 0.741822 +eval/Actions Max 0.999924 +eval/Actions Min -0.99983 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.38381e-05 +time/evaluation sampling (s) 4.31079 +time/exploration sampling (s) 6.75665 +time/logging (s) 0.00653016 +time/saving (s) 0.0139235 +time/training (s) 19.4551 +time/epoch (s) 30.543 +time/total (s) 14920 +Epoch -394 +------------------------------ ---------------- +2022-05-15 22:11:38.779656 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -393 finished +------------------------------ ---------------- +epoch -393 +replay_buffer/size 999047 +trainer/num train calls 608000 +trainer/QF1 Loss 0.623784 +trainer/QF2 Loss 0.541033 +trainer/Policy Loss 22.8923 +trainer/Q1 Predictions Mean -75.0891 +trainer/Q1 Predictions Std 17.4806 +trainer/Q1 Predictions Max -0.313391 +trainer/Q1 Predictions Min -87.9746 +trainer/Q2 Predictions Mean -75.1296 +trainer/Q2 Predictions Std 17.4753 +trainer/Q2 Predictions Max -0.430771 +trainer/Q2 Predictions Min -87.9026 +trainer/Q Targets Mean -75.3706 +trainer/Q Targets Std 17.5909 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0731 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0146414 +trainer/policy/mean Std 0.723297 +trainer/policy/mean Max 0.999034 +trainer/policy/mean Min -0.998178 +trainer/policy/std Mean 0.415039 +trainer/policy/std Std 0.0203095 +trainer/policy/std Max 0.435759 +trainer/policy/std Min 0.384445 +trainer/Advantage Weights Mean 6.79268 +trainer/Advantage Weights Std 20.9794 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.1582e-17 +trainer/Advantage Score Mean -0.217597 +trainer/Advantage Score Std 0.548036 +trainer/Advantage Score Max 2.35175 +trainer/Advantage Score Min -3.83747 +trainer/V1 Predictions Mean -75.0946 +trainer/V1 Predictions Std 17.7344 +trainer/V1 Predictions Max 0.70927 +trainer/V1 Predictions Min -87.9897 +trainer/VF Loss 0.0698818 +expl/num steps total 608000 +expl/num paths total 784 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0282431 +expl/Actions Std 0.842737 +expl/Actions Max 2.3663 +expl/Actions Min -2.44182 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 564356 +eval/num paths total 610 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.141949 +eval/Actions Std 0.690545 +eval/Actions Max 0.999886 +eval/Actions Min -0.999089 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.05598e-05 +time/evaluation sampling (s) 4.73529 +time/exploration sampling (s) 6.15696 +time/logging (s) 0.0121836 +time/saving (s) 0.0170441 +time/training (s) 19.4056 +time/epoch (s) 30.327 +time/total (s) 14950.3 +Epoch -393 +------------------------------ ---------------- +2022-05-15 22:12:08.961451 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -392 finished +------------------------------ ---------------- +epoch -392 +replay_buffer/size 999047 +trainer/num train calls 609000 +trainer/QF1 Loss 0.73826 +trainer/QF2 Loss 0.768214 +trainer/Policy Loss 8.60542 +trainer/Q1 Predictions Mean -75.4476 +trainer/Q1 Predictions Std 15.6612 +trainer/Q1 Predictions Max -2.15089 +trainer/Q1 Predictions Min -87.2546 +trainer/Q2 Predictions Mean -75.4061 +trainer/Q2 Predictions Std 15.6618 +trainer/Q2 Predictions Max -1.15729 +trainer/Q2 Predictions Min -87.0865 +trainer/Q Targets Mean -75.4705 +trainer/Q Targets Std 15.9178 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4491 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0241327 +trainer/policy/mean Std 0.748329 +trainer/policy/mean Max 0.999278 +trainer/policy/mean Min -0.999257 +trainer/policy/std Mean 0.414978 +trainer/policy/std Std 0.0198036 +trainer/policy/std Max 0.435413 +trainer/policy/std Min 0.382575 +trainer/Advantage Weights Mean 2.48301 +trainer/Advantage Weights Std 12.8401 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.71373e-17 +trainer/Advantage Score Mean -0.441238 +trainer/Advantage Score Std 0.569922 +trainer/Advantage Score Max 1.38787 +trainer/Advantage Score Min -3.71009 +trainer/V1 Predictions Mean -75.3382 +trainer/V1 Predictions Std 15.7422 +trainer/V1 Predictions Max -1.70239 +trainer/V1 Predictions Min -87.3234 +trainer/VF Loss 0.0632206 +expl/num steps total 609000 +expl/num paths total 786 +expl/path length Mean 500 +expl/path length Std 148 +expl/path length Max 648 +expl/path length Min 352 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0202783 +expl/Actions Std 0.829923 +expl/Actions Max 2.44066 +expl/Actions Min -2.41503 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 565356 +eval/num paths total 611 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.263756 +eval/Actions Std 0.660219 +eval/Actions Max 0.999829 +eval/Actions Min -0.999538 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.3466e-05 +time/evaluation sampling (s) 4.23221 +time/exploration sampling (s) 6.44897 +time/logging (s) 0.0078913 +time/saving (s) 0.0149666 +time/training (s) 19.459 +time/epoch (s) 30.1631 +time/total (s) 14980.5 +Epoch -392 +------------------------------ ---------------- +2022-05-15 22:12:39.653134 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -391 finished +------------------------------ ---------------- +epoch -391 +replay_buffer/size 999047 +trainer/num train calls 610000 +trainer/QF1 Loss 0.563904 +trainer/QF2 Loss 0.596031 +trainer/Policy Loss 44.5704 +trainer/Q1 Predictions Mean -75.3768 +trainer/Q1 Predictions Std 14.9549 +trainer/Q1 Predictions Max -1.15676 +trainer/Q1 Predictions Min -86.9903 +trainer/Q2 Predictions Mean -75.358 +trainer/Q2 Predictions Std 14.9199 +trainer/Q2 Predictions Max -1.0012 +trainer/Q2 Predictions Min -86.8451 +trainer/Q Targets Mean -75.5534 +trainer/Q Targets Std 14.9235 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0088 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0136543 +trainer/policy/mean Std 0.715255 +trainer/policy/mean Max 0.99962 +trainer/policy/mean Min -0.997628 +trainer/policy/std Mean 0.416214 +trainer/policy/std Std 0.0201614 +trainer/policy/std Max 0.43715 +trainer/policy/std Min 0.3825 +trainer/Advantage Weights Mean 8.40912 +trainer/Advantage Weights Std 23.6677 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.95678e-10 +trainer/Advantage Score Mean -0.192582 +trainer/Advantage Score Std 0.481517 +trainer/Advantage Score Max 1.2539 +trainer/Advantage Score Min -2.14251 +trainer/V1 Predictions Mean -75.3481 +trainer/V1 Predictions Std 14.987 +trainer/V1 Predictions Max 0.234425 +trainer/V1 Predictions Min -86.8867 +trainer/VF Loss 0.0654716 +expl/num steps total 610000 +expl/num paths total 788 +expl/path length Mean 500 +expl/path length Std 318 +expl/path length Max 818 +expl/path length Min 182 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0404511 +expl/Actions Std 0.827397 +expl/Actions Max 2.48732 +expl/Actions Min -2.28701 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 566042 +eval/num paths total 612 +eval/path length Mean 686 +eval/path length Std 0 +eval/path length Max 686 +eval/path length Min 686 +eval/Rewards Mean 0.00145773 +eval/Rewards Std 0.0381523 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00586885 +eval/Actions Std 0.719828 +eval/Actions Max 0.999623 +eval/Actions Min -0.999212 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.4158e-05 +time/evaluation sampling (s) 4.7641 +time/exploration sampling (s) 6.56062 +time/logging (s) 0.0103914 +time/saving (s) 0.0150568 +time/training (s) 19.3296 +time/epoch (s) 30.6798 +time/total (s) 15011.2 +Epoch -391 +------------------------------ ---------------- +2022-05-15 22:13:11.101443 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -390 finished +------------------------------ ---------------- +epoch -390 +replay_buffer/size 999047 +trainer/num train calls 611000 +trainer/QF1 Loss 1.06961 +trainer/QF2 Loss 0.950492 +trainer/Policy Loss 8.92046 +trainer/Q1 Predictions Mean -73.3369 +trainer/Q1 Predictions Std 19.2745 +trainer/Q1 Predictions Max 0.0484117 +trainer/Q1 Predictions Min -86.6545 +trainer/Q2 Predictions Mean -73.3517 +trainer/Q2 Predictions Std 19.2802 +trainer/Q2 Predictions Max -0.487818 +trainer/Q2 Predictions Min -86.5064 +trainer/Q Targets Mean -72.859 +trainer/Q Targets Std 19.3096 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6541 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0143549 +trainer/policy/mean Std 0.72721 +trainer/policy/mean Max 0.999725 +trainer/policy/mean Min -0.999793 +trainer/policy/std Mean 0.415188 +trainer/policy/std Std 0.021043 +trainer/policy/std Max 0.440526 +trainer/policy/std Min 0.38132 +trainer/Advantage Weights Mean 2.06895 +trainer/Advantage Weights Std 13.8398 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.20282e-18 +trainer/Advantage Score Mean -0.711602 +trainer/Advantage Score Std 0.629324 +trainer/Advantage Score Max 2.3759 +trainer/Advantage Score Min -4.12619 +trainer/V1 Predictions Mean -72.6772 +trainer/V1 Predictions Std 19.1825 +trainer/V1 Predictions Max -0.619085 +trainer/V1 Predictions Min -86.4209 +trainer/VF Loss 0.130788 +expl/num steps total 611000 +expl/num paths total 789 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0359824 +expl/Actions Std 0.813251 +expl/Actions Max 2.33199 +expl/Actions Min -2.37553 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 566484 +eval/num paths total 613 +eval/path length Mean 442 +eval/path length Std 0 +eval/path length Max 442 +eval/path length Min 442 +eval/Rewards Mean 0.00226244 +eval/Rewards Std 0.0475113 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0312413 +eval/Actions Std 0.745046 +eval/Actions Max 0.999572 +eval/Actions Min -0.999853 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.72786e-06 +time/evaluation sampling (s) 5.06895 +time/exploration sampling (s) 7.29216 +time/logging (s) 0.00968116 +time/saving (s) 0.0176788 +time/training (s) 19.0419 +time/epoch (s) 31.4304 +time/total (s) 15042.6 +Epoch -390 +------------------------------ ---------------- +2022-05-15 22:13:41.966702 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -389 finished +------------------------------ ---------------- +epoch -389 +replay_buffer/size 999047 +trainer/num train calls 612000 +trainer/QF1 Loss 0.456712 +trainer/QF2 Loss 0.407548 +trainer/Policy Loss 13.5244 +trainer/Q1 Predictions Mean -73.431 +trainer/Q1 Predictions Std 19.0087 +trainer/Q1 Predictions Max -1.95293 +trainer/Q1 Predictions Min -87.9622 +trainer/Q2 Predictions Mean -73.5006 +trainer/Q2 Predictions Std 18.9411 +trainer/Q2 Predictions Max -2.23954 +trainer/Q2 Predictions Min -87.1577 +trainer/Q Targets Mean -73.3433 +trainer/Q Targets Std 19.1293 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2852 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0252503 +trainer/policy/mean Std 0.717852 +trainer/policy/mean Max 0.999287 +trainer/policy/mean Min -0.99957 +trainer/policy/std Mean 0.415772 +trainer/policy/std Std 0.0210684 +trainer/policy/std Max 0.439959 +trainer/policy/std Min 0.381487 +trainer/Advantage Weights Mean 1.94042 +trainer/Advantage Weights Std 12.2027 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.22979e-17 +trainer/Advantage Score Mean -0.551245 +trainer/Advantage Score Std 0.607354 +trainer/Advantage Score Max 1.16864 +trainer/Advantage Score Min -3.71657 +trainer/V1 Predictions Mean -73.075 +trainer/V1 Predictions Std 19.3303 +trainer/V1 Predictions Max -0.84845 +trainer/V1 Predictions Min -87.4126 +trainer/VF Loss 0.0752994 +expl/num steps total 612000 +expl/num paths total 791 +expl/path length Mean 500 +expl/path length Std 170 +expl/path length Max 670 +expl/path length Min 330 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0384196 +expl/Actions Std 0.813803 +expl/Actions Max 2.78084 +expl/Actions Min -2.36709 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 567484 +eval/num paths total 614 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.174487 +eval/Actions Std 0.780407 +eval/Actions Max 0.999721 +eval/Actions Min -0.997836 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.40586e-06 +time/evaluation sampling (s) 4.48996 +time/exploration sampling (s) 6.79343 +time/logging (s) 0.00840356 +time/saving (s) 0.0108001 +time/training (s) 19.5435 +time/epoch (s) 30.8461 +time/total (s) 15073.5 +Epoch -389 +------------------------------ ---------------- +2022-05-15 22:14:12.351770 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -388 finished +------------------------------ ---------------- +epoch -388 +replay_buffer/size 999047 +trainer/num train calls 613000 +trainer/QF1 Loss 0.7917 +trainer/QF2 Loss 0.891775 +trainer/Policy Loss 48.14 +trainer/Q1 Predictions Mean -74.2308 +trainer/Q1 Predictions Std 17.3566 +trainer/Q1 Predictions Max -1.72067 +trainer/Q1 Predictions Min -87.0219 +trainer/Q2 Predictions Mean -74.3205 +trainer/Q2 Predictions Std 17.2815 +trainer/Q2 Predictions Max -2.04544 +trainer/Q2 Predictions Min -86.9368 +trainer/Q Targets Mean -74.3898 +trainer/Q Targets Std 17.7637 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4946 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0038592 +trainer/policy/mean Std 0.728361 +trainer/policy/mean Max 0.999637 +trainer/policy/mean Min -0.997869 +trainer/policy/std Mean 0.416153 +trainer/policy/std Std 0.0202479 +trainer/policy/std Max 0.436894 +trainer/policy/std Min 0.382479 +trainer/Advantage Weights Mean 12.5369 +trainer/Advantage Weights Std 29.08 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.45957e-13 +trainer/Advantage Score Mean -0.175149 +trainer/Advantage Score Std 0.585464 +trainer/Advantage Score Max 1.4845 +trainer/Advantage Score Min -2.82362 +trainer/V1 Predictions Mean -74.1687 +trainer/V1 Predictions Std 17.7573 +trainer/V1 Predictions Max -0.745398 +trainer/V1 Predictions Min -87.3687 +trainer/VF Loss 0.0764425 +expl/num steps total 613000 +expl/num paths total 793 +expl/path length Mean 500 +expl/path length Std 199 +expl/path length Max 699 +expl/path length Min 301 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0184532 +expl/Actions Std 0.836351 +expl/Actions Max 2.17991 +expl/Actions Min -2.15048 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 568434 +eval/num paths total 615 +eval/path length Mean 950 +eval/path length Std 0 +eval/path length Max 950 +eval/path length Min 950 +eval/Rewards Mean 0.00105263 +eval/Rewards Std 0.0324272 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00866503 +eval/Actions Std 0.754542 +eval/Actions Max 0.999958 +eval/Actions Min -0.999721 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.0618e-05 +time/evaluation sampling (s) 4.91258 +time/exploration sampling (s) 6.46875 +time/logging (s) 0.00854724 +time/saving (s) 0.0145229 +time/training (s) 18.9677 +time/epoch (s) 30.3721 +time/total (s) 15103.9 +Epoch -388 +------------------------------ ---------------- +2022-05-15 22:14:43.695955 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -387 finished +------------------------------ ---------------- +epoch -387 +replay_buffer/size 999047 +trainer/num train calls 614000 +trainer/QF1 Loss 0.673886 +trainer/QF2 Loss 0.667895 +trainer/Policy Loss 29.2303 +trainer/Q1 Predictions Mean -73.016 +trainer/Q1 Predictions Std 18.8372 +trainer/Q1 Predictions Max -1.2317 +trainer/Q1 Predictions Min -86.1336 +trainer/Q2 Predictions Mean -73.0494 +trainer/Q2 Predictions Std 18.9219 +trainer/Q2 Predictions Max -1.17666 +trainer/Q2 Predictions Min -86.0443 +trainer/Q Targets Mean -73.1615 +trainer/Q Targets Std 19.0572 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.235 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0172192 +trainer/policy/mean Std 0.724625 +trainer/policy/mean Max 0.999283 +trainer/policy/mean Min -0.999815 +trainer/policy/std Mean 0.415962 +trainer/policy/std Std 0.0210273 +trainer/policy/std Max 0.43754 +trainer/policy/std Min 0.384038 +trainer/Advantage Weights Mean 5.2147 +trainer/Advantage Weights Std 19.2415 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.70982e-27 +trainer/Advantage Score Mean -0.346156 +trainer/Advantage Score Std 0.635547 +trainer/Advantage Score Max 1.11801 +trainer/Advantage Score Min -5.98967 +trainer/V1 Predictions Mean -72.92 +trainer/V1 Predictions Std 19.1852 +trainer/V1 Predictions Max -0.314425 +trainer/V1 Predictions Min -86.0379 +trainer/VF Loss 0.0713507 +expl/num steps total 614000 +expl/num paths total 795 +expl/path length Mean 500 +expl/path length Std 362 +expl/path length Max 862 +expl/path length Min 138 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0304207 +expl/Actions Std 0.829633 +expl/Actions Max 2.43752 +expl/Actions Min -2.34554 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 569434 +eval/num paths total 616 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.107704 +eval/Actions Std 0.714905 +eval/Actions Max 0.999745 +eval/Actions Min -0.998855 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.33091e-05 +time/evaluation sampling (s) 5.06323 +time/exploration sampling (s) 6.70732 +time/logging (s) 0.0116684 +time/saving (s) 0.0150747 +time/training (s) 19.5382 +time/epoch (s) 31.3355 +time/total (s) 15135.2 +Epoch -387 +------------------------------ ---------------- +2022-05-15 22:15:13.990043 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -386 finished +------------------------------ ---------------- +epoch -386 +replay_buffer/size 999047 +trainer/num train calls 615000 +trainer/QF1 Loss 0.592943 +trainer/QF2 Loss 0.489213 +trainer/Policy Loss 32.8124 +trainer/Q1 Predictions Mean -72.958 +trainer/Q1 Predictions Std 18.7822 +trainer/Q1 Predictions Max -0.299751 +trainer/Q1 Predictions Min -86.7889 +trainer/Q2 Predictions Mean -73.0528 +trainer/Q2 Predictions Std 18.7556 +trainer/Q2 Predictions Max -0.526709 +trainer/Q2 Predictions Min -86.85 +trainer/Q Targets Mean -73.0571 +trainer/Q Targets Std 19.1481 +trainer/Q Targets Max 0.663177 +trainer/Q Targets Min -87.1232 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00171491 +trainer/policy/mean Std 0.72314 +trainer/policy/mean Max 0.999893 +trainer/policy/mean Min -0.998771 +trainer/policy/std Mean 0.415444 +trainer/policy/std Std 0.0199322 +trainer/policy/std Max 0.440756 +trainer/policy/std Min 0.387013 +trainer/Advantage Weights Mean 8.07626 +trainer/Advantage Weights Std 23.115 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.33468e-15 +trainer/Advantage Score Mean -0.261782 +trainer/Advantage Score Std 0.621247 +trainer/Advantage Score Max 1.08112 +trainer/Advantage Score Min -3.42501 +trainer/V1 Predictions Mean -72.8245 +trainer/V1 Predictions Std 19.1718 +trainer/V1 Predictions Max 0.756974 +trainer/V1 Predictions Min -86.9497 +trainer/VF Loss 0.0724256 +expl/num steps total 615000 +expl/num paths total 796 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.170221 +expl/Actions Std 0.822188 +expl/Actions Max 2.5127 +expl/Actions Min -2.51504 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 570065 +eval/num paths total 617 +eval/path length Mean 631 +eval/path length Std 0 +eval/path length Max 631 +eval/path length Min 631 +eval/Rewards Mean 0.00158479 +eval/Rewards Std 0.0397778 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0148147 +eval/Actions Std 0.74169 +eval/Actions Max 0.999768 +eval/Actions Min -0.99973 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.024e-06 +time/evaluation sampling (s) 4.37793 +time/exploration sampling (s) 6.66766 +time/logging (s) 0.0107345 +time/saving (s) 0.0180356 +time/training (s) 19.1998 +time/epoch (s) 30.2742 +time/total (s) 15165.5 +Epoch -386 +------------------------------ ---------------- +2022-05-15 22:15:44.597485 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -385 finished +------------------------------ ---------------- +epoch -385 +replay_buffer/size 999047 +trainer/num train calls 616000 +trainer/QF1 Loss 0.584106 +trainer/QF2 Loss 0.545337 +trainer/Policy Loss 21.6674 +trainer/Q1 Predictions Mean -72.4449 +trainer/Q1 Predictions Std 19.1093 +trainer/Q1 Predictions Max -0.613714 +trainer/Q1 Predictions Min -86.4549 +trainer/Q2 Predictions Mean -72.3709 +trainer/Q2 Predictions Std 19.2081 +trainer/Q2 Predictions Max 0.386575 +trainer/Q2 Predictions Min -86.3175 +trainer/Q Targets Mean -72.272 +trainer/Q Targets Std 19.1163 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.3518 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00829458 +trainer/policy/mean Std 0.717432 +trainer/policy/mean Max 0.999302 +trainer/policy/mean Min -0.998218 +trainer/policy/std Mean 0.414325 +trainer/policy/std Std 0.0192701 +trainer/policy/std Max 0.437791 +trainer/policy/std Min 0.383309 +trainer/Advantage Weights Mean 5.44613 +trainer/Advantage Weights Std 20.4612 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.62145e-13 +trainer/Advantage Score Mean -0.352381 +trainer/Advantage Score Std 0.501433 +trainer/Advantage Score Max 1.20984 +trainer/Advantage Score Min -2.8207 +trainer/V1 Predictions Mean -72.1521 +trainer/V1 Predictions Std 19.1065 +trainer/V1 Predictions Max 0.414383 +trainer/V1 Predictions Min -86.2131 +trainer/VF Loss 0.0661427 +expl/num steps total 616000 +expl/num paths total 798 +expl/path length Mean 500 +expl/path length Std 113 +expl/path length Max 613 +expl/path length Min 387 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.027142 +expl/Actions Std 0.830838 +expl/Actions Max 2.29867 +expl/Actions Min -2.35041 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 571065 +eval/num paths total 618 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.21351 +eval/Actions Std 0.690273 +eval/Actions Max 0.999527 +eval/Actions Min -0.99929 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.36371e-06 +time/evaluation sampling (s) 4.86537 +time/exploration sampling (s) 6.86005 +time/logging (s) 0.0096668 +time/saving (s) 0.0173152 +time/training (s) 18.8358 +time/epoch (s) 30.5882 +time/total (s) 15196.1 +Epoch -385 +------------------------------ ---------------- +2022-05-15 22:16:15.649580 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -384 finished +------------------------------ ---------------- +epoch -384 +replay_buffer/size 999047 +trainer/num train calls 617000 +trainer/QF1 Loss 0.652981 +trainer/QF2 Loss 0.44789 +trainer/Policy Loss 12.1472 +trainer/Q1 Predictions Mean -74.1033 +trainer/Q1 Predictions Std 17.5376 +trainer/Q1 Predictions Max -0.443055 +trainer/Q1 Predictions Min -86.777 +trainer/Q2 Predictions Mean -73.9655 +trainer/Q2 Predictions Std 17.6733 +trainer/Q2 Predictions Max -0.114659 +trainer/Q2 Predictions Min -86.6701 +trainer/Q Targets Mean -73.9128 +trainer/Q Targets Std 17.6651 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1598 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00481654 +trainer/policy/mean Std 0.733497 +trainer/policy/mean Max 0.999383 +trainer/policy/mean Min -0.999587 +trainer/policy/std Mean 0.41692 +trainer/policy/std Std 0.0199687 +trainer/policy/std Max 0.441403 +trainer/policy/std Min 0.387554 +trainer/Advantage Weights Mean 2.92016 +trainer/Advantage Weights Std 13.4997 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.01583e-17 +trainer/Advantage Score Mean -0.406374 +trainer/Advantage Score Std 0.67162 +trainer/Advantage Score Max 1.24266 +trainer/Advantage Score Min -3.84429 +trainer/V1 Predictions Mean -73.6619 +trainer/V1 Predictions Std 17.7961 +trainer/V1 Predictions Max -0.688119 +trainer/V1 Predictions Min -86.8433 +trainer/VF Loss 0.0721996 +expl/num steps total 617000 +expl/num paths total 799 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0378343 +expl/Actions Std 0.832668 +expl/Actions Max 2.38041 +expl/Actions Min -2.29876 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 571798 +eval/num paths total 619 +eval/path length Mean 733 +eval/path length Std 0 +eval/path length Max 733 +eval/path length Min 733 +eval/Rewards Mean 0.00136426 +eval/Rewards Std 0.0369106 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00617341 +eval/Actions Std 0.729693 +eval/Actions Max 0.999756 +eval/Actions Min -0.999431 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.0957e-05 +time/evaluation sampling (s) 4.8459 +time/exploration sampling (s) 7.1448 +time/logging (s) 0.00757793 +time/saving (s) 0.0137501 +time/training (s) 19.0217 +time/epoch (s) 31.0338 +time/total (s) 15227.1 +Epoch -384 +------------------------------ ---------------- +2022-05-15 22:16:46.877143 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -383 finished +------------------------------ ---------------- +epoch -383 +replay_buffer/size 999047 +trainer/num train calls 618000 +trainer/QF1 Loss 0.89902 +trainer/QF2 Loss 0.850441 +trainer/Policy Loss 14.7729 +trainer/Q1 Predictions Mean -72.0397 +trainer/Q1 Predictions Std 19.7165 +trainer/Q1 Predictions Max -1.19849 +trainer/Q1 Predictions Min -86.6355 +trainer/Q2 Predictions Mean -72.0899 +trainer/Q2 Predictions Std 19.6875 +trainer/Q2 Predictions Max -2.57552 +trainer/Q2 Predictions Min -86.605 +trainer/Q Targets Mean -71.9486 +trainer/Q Targets Std 20.1056 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7022 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00988505 +trainer/policy/mean Std 0.724873 +trainer/policy/mean Max 0.999813 +trainer/policy/mean Min -0.999186 +trainer/policy/std Mean 0.417452 +trainer/policy/std Std 0.0203568 +trainer/policy/std Max 0.441094 +trainer/policy/std Min 0.383767 +trainer/Advantage Weights Mean 2.22211 +trainer/Advantage Weights Std 12.7527 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.63646e-18 +trainer/Advantage Score Mean -0.568434 +trainer/Advantage Score Std 0.689831 +trainer/Advantage Score Max 1.10624 +trainer/Advantage Score Min -4.0954 +trainer/V1 Predictions Mean -71.747 +trainer/V1 Predictions Std 20.0776 +trainer/V1 Predictions Max -0.728759 +trainer/V1 Predictions Min -86.7504 +trainer/VF Loss 0.088157 +expl/num steps total 618000 +expl/num paths total 801 +expl/path length Mean 500 +expl/path length Std 253 +expl/path length Max 753 +expl/path length Min 247 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0174487 +expl/Actions Std 0.83877 +expl/Actions Max 2.15914 +expl/Actions Min -2.39348 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 572253 +eval/num paths total 620 +eval/path length Mean 455 +eval/path length Std 0 +eval/path length Max 455 +eval/path length Min 455 +eval/Rewards Mean 0.0021978 +eval/Rewards Std 0.0468292 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0402139 +eval/Actions Std 0.745869 +eval/Actions Max 0.999935 +eval/Actions Min -0.999928 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.25458e-05 +time/evaluation sampling (s) 5.13507 +time/exploration sampling (s) 7.07112 +time/logging (s) 0.0102841 +time/saving (s) 0.0188534 +time/training (s) 18.9828 +time/epoch (s) 31.2181 +time/total (s) 15258.3 +Epoch -383 +------------------------------ ---------------- +2022-05-15 22:17:18.086579 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -382 finished +------------------------------ ---------------- +epoch -382 +replay_buffer/size 999047 +trainer/num train calls 619000 +trainer/QF1 Loss 0.471466 +trainer/QF2 Loss 0.421618 +trainer/Policy Loss 4.71044 +trainer/Q1 Predictions Mean -75.855 +trainer/Q1 Predictions Std 15.0205 +trainer/Q1 Predictions Max -0.491055 +trainer/Q1 Predictions Min -86.9372 +trainer/Q2 Predictions Mean -75.8518 +trainer/Q2 Predictions Std 14.938 +trainer/Q2 Predictions Max -1.00891 +trainer/Q2 Predictions Min -86.6429 +trainer/Q Targets Mean -75.9723 +trainer/Q Targets Std 15.0247 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7731 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.018003 +trainer/policy/mean Std 0.737372 +trainer/policy/mean Max 0.998646 +trainer/policy/mean Min -0.999167 +trainer/policy/std Mean 0.416172 +trainer/policy/std Std 0.0208114 +trainer/policy/std Max 0.440151 +trainer/policy/std Min 0.384187 +trainer/Advantage Weights Mean 2.03661 +trainer/Advantage Weights Std 12.0901 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.04194e-14 +trainer/Advantage Score Mean -0.416748 +trainer/Advantage Score Std 0.509803 +trainer/Advantage Score Max 0.693375 +trainer/Advantage Score Min -3.15223 +trainer/V1 Predictions Mean -75.6466 +trainer/V1 Predictions Std 15.2169 +trainer/V1 Predictions Max -0.693454 +trainer/V1 Predictions Min -86.744 +trainer/VF Loss 0.048659 +expl/num steps total 619000 +expl/num paths total 803 +expl/path length Mean 500 +expl/path length Std 488 +expl/path length Max 988 +expl/path length Min 12 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.032758 +expl/Actions Std 0.820711 +expl/Actions Max 2.29095 +expl/Actions Min -2.22463 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 573168 +eval/num paths total 622 +eval/path length Mean 457.5 +eval/path length Std 84.5 +eval/path length Max 542 +eval/path length Min 373 +eval/Rewards Mean 0.00218579 +eval/Rewards Std 0.0467013 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00618467 +eval/Actions Std 0.740232 +eval/Actions Max 0.999933 +eval/Actions Min -0.999837 +eval/Num Paths 2 +eval/Average Returns 1 +time/data storing (s) 1.27531e-05 +time/evaluation sampling (s) 5.31382 +time/exploration sampling (s) 6.5915 +time/logging (s) 0.0123939 +time/saving (s) 0.0156441 +time/training (s) 19.2599 +time/epoch (s) 31.1933 +time/total (s) 15289.5 +Epoch -382 +------------------------------ ---------------- +2022-05-15 22:17:48.879519 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -381 finished +------------------------------ ---------------- +epoch -381 +replay_buffer/size 999047 +trainer/num train calls 620000 +trainer/QF1 Loss 1.22815 +trainer/QF2 Loss 1.25677 +trainer/Policy Loss 20.5651 +trainer/Q1 Predictions Mean -74.4143 +trainer/Q1 Predictions Std 16.7053 +trainer/Q1 Predictions Max -1.4302 +trainer/Q1 Predictions Min -87.2001 +trainer/Q2 Predictions Mean -74.4642 +trainer/Q2 Predictions Std 16.7473 +trainer/Q2 Predictions Max -2.07749 +trainer/Q2 Predictions Min -87.0997 +trainer/Q Targets Mean -74.4049 +trainer/Q Targets Std 17.3148 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0243 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00945089 +trainer/policy/mean Std 0.71486 +trainer/policy/mean Max 0.999978 +trainer/policy/mean Min -0.999468 +trainer/policy/std Mean 0.41438 +trainer/policy/std Std 0.0228154 +trainer/policy/std Max 0.440739 +trainer/policy/std Min 0.380555 +trainer/Advantage Weights Mean 4.09729 +trainer/Advantage Weights Std 15.324 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.42218e-26 +trainer/Advantage Score Mean -0.358843 +trainer/Advantage Score Std 0.724507 +trainer/Advantage Score Max 0.694901 +trainer/Advantage Score Min -5.9515 +trainer/V1 Predictions Mean -74.2393 +trainer/V1 Predictions Std 17.0963 +trainer/V1 Predictions Max -1.96211 +trainer/V1 Predictions Min -87.0287 +trainer/VF Loss 0.0766566 +expl/num steps total 620000 +expl/num paths total 805 +expl/path length Mean 500 +expl/path length Std 83 +expl/path length Max 583 +expl/path length Min 417 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0404947 +expl/Actions Std 0.82647 +expl/Actions Max 2.43708 +expl/Actions Min -2.28291 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 574168 +eval/num paths total 623 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.363485 +eval/Actions Std 0.665103 +eval/Actions Max 0.999827 +eval/Actions Min -0.999733 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.25552e-05 +time/evaluation sampling (s) 4.8253 +time/exploration sampling (s) 6.73336 +time/logging (s) 0.0126045 +time/saving (s) 0.0184982 +time/training (s) 19.189 +time/epoch (s) 30.7787 +time/total (s) 15320.3 +Epoch -381 +------------------------------ ---------------- +2022-05-15 22:18:19.588651 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -380 finished +------------------------------ ---------------- +epoch -380 +replay_buffer/size 999047 +trainer/num train calls 621000 +trainer/QF1 Loss 0.888582 +trainer/QF2 Loss 0.831446 +trainer/Policy Loss 27.9843 +trainer/Q1 Predictions Mean -74.7331 +trainer/Q1 Predictions Std 17.2409 +trainer/Q1 Predictions Max -1.59825 +trainer/Q1 Predictions Min -87.2601 +trainer/Q2 Predictions Mean -74.6259 +trainer/Q2 Predictions Std 17.3458 +trainer/Q2 Predictions Max -1.66899 +trainer/Q2 Predictions Min -87.2877 +trainer/Q Targets Mean -74.8291 +trainer/Q Targets Std 17.1901 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2304 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00935238 +trainer/policy/mean Std 0.726829 +trainer/policy/mean Max 0.999631 +trainer/policy/mean Min -0.999449 +trainer/policy/std Mean 0.414511 +trainer/policy/std Std 0.0228689 +trainer/policy/std Max 0.444901 +trainer/policy/std Min 0.375017 +trainer/Advantage Weights Mean 5.76974 +trainer/Advantage Weights Std 21.3425 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.1795e-11 +trainer/Advantage Score Mean -0.285829 +trainer/Advantage Score Std 0.464331 +trainer/Advantage Score Max 1.56639 +trainer/Advantage Score Min -2.51633 +trainer/V1 Predictions Mean -74.6396 +trainer/V1 Predictions Std 17.2707 +trainer/V1 Predictions Max 0.46605 +trainer/V1 Predictions Min -86.5563 +trainer/VF Loss 0.0569414 +expl/num steps total 621000 +expl/num paths total 806 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.259577 +expl/Actions Std 0.863702 +expl/Actions Max 2.35031 +expl/Actions Min -2.31334 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 575168 +eval/num paths total 624 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.140818 +eval/Actions Std 0.757512 +eval/Actions Max 0.999864 +eval/Actions Min -0.999672 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.17533e-05 +time/evaluation sampling (s) 4.43813 +time/exploration sampling (s) 6.63639 +time/logging (s) 0.00785255 +time/saving (s) 0.0205367 +time/training (s) 19.5832 +time/epoch (s) 30.6861 +time/total (s) 15351 +Epoch -380 +------------------------------ ---------------- +2022-05-15 22:18:50.077524 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -379 finished +------------------------------ ---------------- +epoch -379 +replay_buffer/size 999047 +trainer/num train calls 622000 +trainer/QF1 Loss 0.934146 +trainer/QF2 Loss 0.816603 +trainer/Policy Loss 11.7875 +trainer/Q1 Predictions Mean -73.1918 +trainer/Q1 Predictions Std 18.6783 +trainer/Q1 Predictions Max -0.536797 +trainer/Q1 Predictions Min -87.33 +trainer/Q2 Predictions Mean -73.2289 +trainer/Q2 Predictions Std 18.6243 +trainer/Q2 Predictions Max -0.703722 +trainer/Q2 Predictions Min -87.4684 +trainer/Q Targets Mean -73.337 +trainer/Q Targets Std 18.6902 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8373 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00448405 +trainer/policy/mean Std 0.719401 +trainer/policy/mean Max 0.99978 +trainer/policy/mean Min -0.999325 +trainer/policy/std Mean 0.412137 +trainer/policy/std Std 0.0216513 +trainer/policy/std Max 0.437569 +trainer/policy/std Min 0.375207 +trainer/Advantage Weights Mean 3.52085 +trainer/Advantage Weights Std 15.6484 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.30097e-20 +trainer/Advantage Score Mean -0.359023 +trainer/Advantage Score Std 0.607476 +trainer/Advantage Score Max 1.66859 +trainer/Advantage Score Min -4.4211 +trainer/V1 Predictions Mean -73.0716 +trainer/V1 Predictions Std 18.9871 +trainer/V1 Predictions Max -1.13189 +trainer/V1 Predictions Min -87.2572 +trainer/VF Loss 0.0719384 +expl/num steps total 622000 +expl/num paths total 808 +expl/path length Mean 500 +expl/path length Std 244 +expl/path length Max 744 +expl/path length Min 256 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0134931 +expl/Actions Std 0.812399 +expl/Actions Max 2.27704 +expl/Actions Min -2.36037 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 575704 +eval/num paths total 625 +eval/path length Mean 536 +eval/path length Std 0 +eval/path length Max 536 +eval/path length Min 536 +eval/Rewards Mean 0.00186567 +eval/Rewards Std 0.0431531 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0172386 +eval/Actions Std 0.727692 +eval/Actions Max 0.999957 +eval/Actions Min -0.999604 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.32103e-05 +time/evaluation sampling (s) 4.70751 +time/exploration sampling (s) 6.75809 +time/logging (s) 0.0107011 +time/saving (s) 0.0171802 +time/training (s) 18.9862 +time/epoch (s) 30.4797 +time/total (s) 15381.5 +Epoch -379 +------------------------------ ---------------- +2022-05-15 22:19:20.727209 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -378 finished +------------------------------ ---------------- +epoch -378 +replay_buffer/size 999047 +trainer/num train calls 623000 +trainer/QF1 Loss 0.725564 +trainer/QF2 Loss 0.610695 +trainer/Policy Loss 52.4813 +trainer/Q1 Predictions Mean -74.1771 +trainer/Q1 Predictions Std 19.2781 +trainer/Q1 Predictions Max -0.238855 +trainer/Q1 Predictions Min -87.0866 +trainer/Q2 Predictions Mean -74.2501 +trainer/Q2 Predictions Std 19.2468 +trainer/Q2 Predictions Max -0.42555 +trainer/Q2 Predictions Min -87.2406 +trainer/Q Targets Mean -74.302 +trainer/Q Targets Std 19.5409 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5108 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0213712 +trainer/policy/mean Std 0.728544 +trainer/policy/mean Max 0.999449 +trainer/policy/mean Min -0.997449 +trainer/policy/std Mean 0.413258 +trainer/policy/std Std 0.0210932 +trainer/policy/std Max 0.438244 +trainer/policy/std Min 0.378977 +trainer/Advantage Weights Mean 9.7263 +trainer/Advantage Weights Std 24.0411 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.56857e-21 +trainer/Advantage Score Mean -0.231223 +trainer/Advantage Score Std 0.75683 +trainer/Advantage Score Max 1.27486 +trainer/Advantage Score Min -4.79041 +trainer/V1 Predictions Mean -74.0632 +trainer/V1 Predictions Std 19.6794 +trainer/V1 Predictions Max 0.0934562 +trainer/V1 Predictions Min -88.0002 +trainer/VF Loss 0.0926126 +expl/num steps total 623000 +expl/num paths total 809 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.107031 +expl/Actions Std 0.833988 +expl/Actions Max 2.24125 +expl/Actions Min -2.5697 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 576704 +eval/num paths total 626 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.211107 +eval/Actions Std 0.65493 +eval/Actions Max 0.999768 +eval/Actions Min -0.9997 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.2811e-06 +time/evaluation sampling (s) 4.85845 +time/exploration sampling (s) 6.51079 +time/logging (s) 0.010667 +time/saving (s) 0.0134816 +time/training (s) 19.2381 +time/epoch (s) 30.6315 +time/total (s) 15412.1 +Epoch -378 +------------------------------ ---------------- +2022-05-15 22:19:51.902494 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -377 finished +------------------------------ ---------------- +epoch -377 +replay_buffer/size 999047 +trainer/num train calls 624000 +trainer/QF1 Loss 0.785442 +trainer/QF2 Loss 0.665661 +trainer/Policy Loss 10.316 +trainer/Q1 Predictions Mean -73.914 +trainer/Q1 Predictions Std 15.3211 +trainer/Q1 Predictions Max -5.97337 +trainer/Q1 Predictions Min -88.1907 +trainer/Q2 Predictions Mean -73.8432 +trainer/Q2 Predictions Std 15.3201 +trainer/Q2 Predictions Max -5.51809 +trainer/Q2 Predictions Min -87.8882 +trainer/Q Targets Mean -73.8914 +trainer/Q Targets Std 15.102 +trainer/Q Targets Max -4.37111 +trainer/Q Targets Min -87.063 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0204556 +trainer/policy/mean Std 0.740535 +trainer/policy/mean Max 0.998747 +trainer/policy/mean Min -0.998398 +trainer/policy/std Mean 0.411865 +trainer/policy/std Std 0.0204581 +trainer/policy/std Max 0.435234 +trainer/policy/std Min 0.378989 +trainer/Advantage Weights Mean 3.66293 +trainer/Advantage Weights Std 16.9642 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.12365e-19 +trainer/Advantage Score Mean -0.555098 +trainer/Advantage Score Std 0.642988 +trainer/Advantage Score Max 1.36324 +trainer/Advantage Score Min -4.21152 +trainer/V1 Predictions Mean -73.5486 +trainer/V1 Predictions Std 15.3375 +trainer/V1 Predictions Max -3.88837 +trainer/V1 Predictions Min -86.8179 +trainer/VF Loss 0.0870968 +expl/num steps total 624000 +expl/num paths total 811 +expl/path length Mean 500 +expl/path length Std 435 +expl/path length Max 935 +expl/path length Min 65 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0363632 +expl/Actions Std 0.819901 +expl/Actions Max 2.41882 +expl/Actions Min -2.38495 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 577269 +eval/num paths total 627 +eval/path length Mean 565 +eval/path length Std 0 +eval/path length Max 565 +eval/path length Min 565 +eval/Rewards Mean 0.00176991 +eval/Rewards Std 0.0420331 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0177434 +eval/Actions Std 0.743017 +eval/Actions Max 0.999655 +eval/Actions Min -0.999753 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.13926e-06 +time/evaluation sampling (s) 5.25791 +time/exploration sampling (s) 7.06487 +time/logging (s) 0.00994514 +time/saving (s) 0.017187 +time/training (s) 18.8096 +time/epoch (s) 31.1595 +time/total (s) 15443.3 +Epoch -377 +------------------------------ ---------------- +2022-05-15 22:20:22.951693 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -376 finished +------------------------------ ---------------- +epoch -376 +replay_buffer/size 999047 +trainer/num train calls 625000 +trainer/QF1 Loss 0.627154 +trainer/QF2 Loss 0.480618 +trainer/Policy Loss 15.0316 +trainer/Q1 Predictions Mean -74.4937 +trainer/Q1 Predictions Std 17.7344 +trainer/Q1 Predictions Max -1.57689 +trainer/Q1 Predictions Min -87.3927 +trainer/Q2 Predictions Mean -74.4991 +trainer/Q2 Predictions Std 17.7859 +trainer/Q2 Predictions Max -1.34848 +trainer/Q2 Predictions Min -87.2283 +trainer/Q Targets Mean -74.3594 +trainer/Q Targets Std 17.9799 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5288 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00902566 +trainer/policy/mean Std 0.716131 +trainer/policy/mean Max 0.9988 +trainer/policy/mean Min -0.997092 +trainer/policy/std Mean 0.415092 +trainer/policy/std Std 0.0201651 +trainer/policy/std Max 0.437765 +trainer/policy/std Min 0.383785 +trainer/Advantage Weights Mean 2.61861 +trainer/Advantage Weights Std 11.0528 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.55238e-20 +trainer/Advantage Score Mean -0.384244 +trainer/Advantage Score Std 0.642753 +trainer/Advantage Score Max 0.711155 +trainer/Advantage Score Min -4.56119 +trainer/V1 Predictions Mean -74.1136 +trainer/V1 Predictions Std 18.0173 +trainer/V1 Predictions Max -0.957281 +trainer/V1 Predictions Min -87.3464 +trainer/VF Loss 0.0631638 +expl/num steps total 625000 +expl/num paths total 812 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.346351 +expl/Actions Std 0.876324 +expl/Actions Max 2.12626 +expl/Actions Min -2.49672 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 578269 +eval/num paths total 628 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.150319 +eval/Actions Std 0.758481 +eval/Actions Max 0.999544 +eval/Actions Min -0.999889 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.88878e-06 +time/evaluation sampling (s) 4.85207 +time/exploration sampling (s) 7.54437 +time/logging (s) 0.0105547 +time/saving (s) 0.0161233 +time/training (s) 18.6089 +time/epoch (s) 31.0321 +time/total (s) 15474.3 +Epoch -376 +------------------------------ ---------------- +2022-05-15 22:20:53.596072 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -375 finished +------------------------------ ---------------- +epoch -375 +replay_buffer/size 999047 +trainer/num train calls 626000 +trainer/QF1 Loss 0.737104 +trainer/QF2 Loss 0.747579 +trainer/Policy Loss 19.0801 +trainer/Q1 Predictions Mean -73.9245 +trainer/Q1 Predictions Std 19.2594 +trainer/Q1 Predictions Max -0.853991 +trainer/Q1 Predictions Min -87.7113 +trainer/Q2 Predictions Mean -73.9666 +trainer/Q2 Predictions Std 19.3018 +trainer/Q2 Predictions Max -1.04297 +trainer/Q2 Predictions Min -87.159 +trainer/Q Targets Mean -73.6032 +trainer/Q Targets Std 19.2089 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0598 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00273012 +trainer/policy/mean Std 0.734185 +trainer/policy/mean Max 0.999401 +trainer/policy/mean Min -0.999144 +trainer/policy/std Mean 0.414502 +trainer/policy/std Std 0.0204822 +trainer/policy/std Max 0.439291 +trainer/policy/std Min 0.382388 +trainer/Advantage Weights Mean 1.8551 +trainer/Advantage Weights Std 12.4418 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.81581e-16 +trainer/Advantage Score Mean -0.573848 +trainer/Advantage Score Std 0.632935 +trainer/Advantage Score Max 0.795532 +trainer/Advantage Score Min -3.62448 +trainer/V1 Predictions Mean -73.2715 +trainer/V1 Predictions Std 19.4513 +trainer/V1 Predictions Max -0.850214 +trainer/V1 Predictions Min -87.0049 +trainer/VF Loss 0.0794676 +expl/num steps total 626000 +expl/num paths total 814 +expl/path length Mean 500 +expl/path length Std 115 +expl/path length Max 615 +expl/path length Min 385 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0114307 +expl/Actions Std 0.836421 +expl/Actions Max 2.2421 +expl/Actions Min -2.20853 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 579056 +eval/num paths total 629 +eval/path length Mean 787 +eval/path length Std 0 +eval/path length Max 787 +eval/path length Min 787 +eval/Rewards Mean 0.00127065 +eval/Rewards Std 0.0356235 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0109576 +eval/Actions Std 0.737342 +eval/Actions Max 0.999035 +eval/Actions Min -0.999734 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.4268e-06 +time/evaluation sampling (s) 5.07532 +time/exploration sampling (s) 6.41347 +time/logging (s) 0.0115659 +time/saving (s) 0.0169779 +time/training (s) 19.1168 +time/epoch (s) 30.6342 +time/total (s) 15505 +Epoch -375 +------------------------------ ---------------- +2022-05-15 22:21:23.596451 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -374 finished +------------------------------ ---------------- +epoch -374 +replay_buffer/size 999047 +trainer/num train calls 627000 +trainer/QF1 Loss 0.533557 +trainer/QF2 Loss 0.537888 +trainer/Policy Loss 26.7291 +trainer/Q1 Predictions Mean -75.4469 +trainer/Q1 Predictions Std 16.3476 +trainer/Q1 Predictions Max -1.4636 +trainer/Q1 Predictions Min -87.2807 +trainer/Q2 Predictions Mean -75.4317 +trainer/Q2 Predictions Std 16.3099 +trainer/Q2 Predictions Max -0.874419 +trainer/Q2 Predictions Min -88.1264 +trainer/Q Targets Mean -75.5811 +trainer/Q Targets Std 16.1818 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5536 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.015614 +trainer/policy/mean Std 0.726739 +trainer/policy/mean Max 0.999508 +trainer/policy/mean Min -0.998803 +trainer/policy/std Mean 0.413623 +trainer/policy/std Std 0.0193936 +trainer/policy/std Max 0.438144 +trainer/policy/std Min 0.380766 +trainer/Advantage Weights Mean 7.29261 +trainer/Advantage Weights Std 22.1204 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.81161e-17 +trainer/Advantage Score Mean -0.272168 +trainer/Advantage Score Std 0.552533 +trainer/Advantage Score Max 1.04177 +trainer/Advantage Score Min -3.85497 +trainer/V1 Predictions Mean -75.331 +trainer/V1 Predictions Std 16.2845 +trainer/V1 Predictions Max -1.80526 +trainer/V1 Predictions Min -86.8853 +trainer/VF Loss 0.064246 +expl/num steps total 627000 +expl/num paths total 816 +expl/path length Mean 500 +expl/path length Std 201 +expl/path length Max 701 +expl/path length Min 299 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0251832 +expl/Actions Std 0.834256 +expl/Actions Max 2.16023 +expl/Actions Min -2.20787 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 579629 +eval/num paths total 630 +eval/path length Mean 573 +eval/path length Std 0 +eval/path length Max 573 +eval/path length Min 573 +eval/Rewards Mean 0.0017452 +eval/Rewards Std 0.0417391 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.001161 +eval/Actions Std 0.726003 +eval/Actions Max 0.99993 +eval/Actions Min -0.999748 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.0282e-06 +time/evaluation sampling (s) 4.35798 +time/exploration sampling (s) 6.5137 +time/logging (s) 0.0106646 +time/saving (s) 0.0224008 +time/training (s) 19.0805 +time/epoch (s) 29.9852 +time/total (s) 15535 +Epoch -374 +------------------------------ ---------------- +2022-05-15 22:21:54.483452 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -373 finished +------------------------------ ---------------- +epoch -373 +replay_buffer/size 999047 +trainer/num train calls 628000 +trainer/QF1 Loss 0.690082 +trainer/QF2 Loss 0.675186 +trainer/Policy Loss 24.1002 +trainer/Q1 Predictions Mean -73.987 +trainer/Q1 Predictions Std 17.4004 +trainer/Q1 Predictions Max -0.654683 +trainer/Q1 Predictions Min -86.985 +trainer/Q2 Predictions Mean -73.9653 +trainer/Q2 Predictions Std 17.4725 +trainer/Q2 Predictions Max -0.669838 +trainer/Q2 Predictions Min -87.4011 +trainer/Q Targets Mean -74.2148 +trainer/Q Targets Std 17.6444 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7523 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0137568 +trainer/policy/mean Std 0.723336 +trainer/policy/mean Max 0.998379 +trainer/policy/mean Min -0.999323 +trainer/policy/std Mean 0.413457 +trainer/policy/std Std 0.0200457 +trainer/policy/std Max 0.438406 +trainer/policy/std Min 0.380433 +trainer/Advantage Weights Mean 8.17397 +trainer/Advantage Weights Std 23.0166 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.702e-16 +trainer/Advantage Score Mean -0.228133 +trainer/Advantage Score Std 0.614439 +trainer/Advantage Score Max 1.58098 +trainer/Advantage Score Min -3.47999 +trainer/V1 Predictions Mean -73.9909 +trainer/V1 Predictions Std 17.7087 +trainer/V1 Predictions Max -0.0899854 +trainer/V1 Predictions Min -87.5717 +trainer/VF Loss 0.0837202 +expl/num steps total 628000 +expl/num paths total 817 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.183634 +expl/Actions Std 0.819636 +expl/Actions Max 2.20691 +expl/Actions Min -2.61249 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 580629 +eval/num paths total 631 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0230991 +eval/Actions Std 0.686233 +eval/Actions Max 0.999237 +eval/Actions Min -0.999835 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.40003e-06 +time/evaluation sampling (s) 4.78843 +time/exploration sampling (s) 6.94019 +time/logging (s) 0.0119111 +time/saving (s) 0.0150677 +time/training (s) 19.1148 +time/epoch (s) 30.8704 +time/total (s) 15565.9 +Epoch -373 +------------------------------ ---------------- +2022-05-15 22:22:25.561515 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -372 finished +------------------------------ ---------------- +epoch -372 +replay_buffer/size 999047 +trainer/num train calls 629000 +trainer/QF1 Loss 0.863354 +trainer/QF2 Loss 1.08642 +trainer/Policy Loss 17.47 +trainer/Q1 Predictions Mean -72.1888 +trainer/Q1 Predictions Std 19.2903 +trainer/Q1 Predictions Max -0.415746 +trainer/Q1 Predictions Min -87.042 +trainer/Q2 Predictions Mean -72.1719 +trainer/Q2 Predictions Std 19.2991 +trainer/Q2 Predictions Max -0.479417 +trainer/Q2 Predictions Min -87.049 +trainer/Q Targets Mean -72.23 +trainer/Q Targets Std 19.3658 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2258 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0234828 +trainer/policy/mean Std 0.70664 +trainer/policy/mean Max 0.996941 +trainer/policy/mean Min -0.996004 +trainer/policy/std Mean 0.41473 +trainer/policy/std Std 0.0203093 +trainer/policy/std Max 0.43763 +trainer/policy/std Min 0.382984 +trainer/Advantage Weights Mean 3.91371 +trainer/Advantage Weights Std 15.9641 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.70824e-19 +trainer/Advantage Score Mean -0.346045 +trainer/Advantage Score Std 0.592328 +trainer/Advantage Score Max 0.868581 +trainer/Advantage Score Min -4.32137 +trainer/V1 Predictions Mean -72.0209 +trainer/V1 Predictions Std 19.4412 +trainer/V1 Predictions Max -0.110405 +trainer/V1 Predictions Min -87.0999 +trainer/VF Loss 0.0589083 +expl/num steps total 629000 +expl/num paths total 818 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.24173 +expl/Actions Std 0.869895 +expl/Actions Max 2.13206 +expl/Actions Min -2.35093 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 581444 +eval/num paths total 632 +eval/path length Mean 815 +eval/path length Std 0 +eval/path length Max 815 +eval/path length Min 815 +eval/Rewards Mean 0.00122699 +eval/Rewards Std 0.035007 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0279158 +eval/Actions Std 0.71628 +eval/Actions Max 0.999864 +eval/Actions Min -0.998858 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.13519e-05 +time/evaluation sampling (s) 4.64875 +time/exploration sampling (s) 7.21785 +time/logging (s) 0.0113601 +time/saving (s) 0.0133701 +time/training (s) 19.1685 +time/epoch (s) 31.0598 +time/total (s) 15596.9 +Epoch -372 +------------------------------ ---------------- +2022-05-15 22:22:55.855371 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -371 finished +------------------------------ ---------------- +epoch -371 +replay_buffer/size 999047 +trainer/num train calls 630000 +trainer/QF1 Loss 0.641961 +trainer/QF2 Loss 0.782635 +trainer/Policy Loss 5.61582 +trainer/Q1 Predictions Mean -71.8606 +trainer/Q1 Predictions Std 19.3478 +trainer/Q1 Predictions Max -0.680425 +trainer/Q1 Predictions Min -87.1303 +trainer/Q2 Predictions Mean -71.8497 +trainer/Q2 Predictions Std 19.5149 +trainer/Q2 Predictions Max -0.183344 +trainer/Q2 Predictions Min -87.4312 +trainer/Q Targets Mean -71.7927 +trainer/Q Targets Std 19.0955 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1743 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0013533 +trainer/policy/mean Std 0.729156 +trainer/policy/mean Max 0.998434 +trainer/policy/mean Min -0.997846 +trainer/policy/std Mean 0.412746 +trainer/policy/std Std 0.0210465 +trainer/policy/std Max 0.439119 +trainer/policy/std Min 0.379087 +trainer/Advantage Weights Mean 1.87075 +trainer/Advantage Weights Std 11.8686 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.23906e-20 +trainer/Advantage Score Mean -0.613939 +trainer/Advantage Score Std 0.5613 +trainer/Advantage Score Max 0.709108 +trainer/Advantage Score Min -4.52456 +trainer/V1 Predictions Mean -71.5242 +trainer/V1 Predictions Std 19.2336 +trainer/V1 Predictions Max -0.471094 +trainer/V1 Predictions Min -87.1125 +trainer/VF Loss 0.074472 +expl/num steps total 630000 +expl/num paths total 819 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0167307 +expl/Actions Std 0.827468 +expl/Actions Max 2.25877 +expl/Actions Min -2.11359 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 582444 +eval/num paths total 633 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.338329 +eval/Actions Std 0.741048 +eval/Actions Max 0.999937 +eval/Actions Min -0.999945 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.32048e-05 +time/evaluation sampling (s) 4.73364 +time/exploration sampling (s) 6.77525 +time/logging (s) 0.00807859 +time/saving (s) 0.0138319 +time/training (s) 18.7426 +time/epoch (s) 30.2734 +time/total (s) 15627.2 +Epoch -371 +------------------------------ ---------------- +2022-05-15 22:23:27.379669 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -370 finished +------------------------------ ---------------- +epoch -370 +replay_buffer/size 999047 +trainer/num train calls 631000 +trainer/QF1 Loss 1.0856 +trainer/QF2 Loss 1.13519 +trainer/Policy Loss 17.9368 +trainer/Q1 Predictions Mean -73.2696 +trainer/Q1 Predictions Std 16.6828 +trainer/Q1 Predictions Max -0.626783 +trainer/Q1 Predictions Min -86.947 +trainer/Q2 Predictions Mean -73.1987 +trainer/Q2 Predictions Std 16.737 +trainer/Q2 Predictions Max -0.473235 +trainer/Q2 Predictions Min -86.9517 +trainer/Q Targets Mean -73.7242 +trainer/Q Targets Std 16.2271 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9225 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00133287 +trainer/policy/mean Std 0.717829 +trainer/policy/mean Max 0.999344 +trainer/policy/mean Min -0.998739 +trainer/policy/std Mean 0.414569 +trainer/policy/std Std 0.0208292 +trainer/policy/std Max 0.436411 +trainer/policy/std Min 0.381251 +trainer/Advantage Weights Mean 4.79776 +trainer/Advantage Weights Std 19.6772 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.50444e-22 +trainer/Advantage Score Mean -0.358384 +trainer/Advantage Score Std 0.5791 +trainer/Advantage Score Max 1.89317 +trainer/Advantage Score Min -4.86414 +trainer/V1 Predictions Mean -73.3825 +trainer/V1 Predictions Std 16.4979 +trainer/V1 Predictions Max 0.213274 +trainer/V1 Predictions Min -86.7516 +trainer/VF Loss 0.0834692 +expl/num steps total 631000 +expl/num paths total 821 +expl/path length Mean 500 +expl/path length Std 242 +expl/path length Max 742 +expl/path length Min 258 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0250212 +expl/Actions Std 0.84794 +expl/Actions Max 2.22206 +expl/Actions Min -2.47574 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 583028 +eval/num paths total 634 +eval/path length Mean 584 +eval/path length Std 0 +eval/path length Max 584 +eval/path length Min 584 +eval/Rewards Mean 0.00171233 +eval/Rewards Std 0.0413449 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0112844 +eval/Actions Std 0.737264 +eval/Actions Max 0.999845 +eval/Actions Min -0.998719 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.2289e-06 +time/evaluation sampling (s) 5.21962 +time/exploration sampling (s) 7.19469 +time/logging (s) 0.0106778 +time/saving (s) 0.017923 +time/training (s) 19.0713 +time/epoch (s) 31.5142 +time/total (s) 15658.7 +Epoch -370 +------------------------------ ---------------- +2022-05-15 22:23:57.750336 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -369 finished +------------------------------ ---------------- +epoch -369 +replay_buffer/size 999047 +trainer/num train calls 632000 +trainer/QF1 Loss 0.72085 +trainer/QF2 Loss 0.639998 +trainer/Policy Loss 38.835 +trainer/Q1 Predictions Mean -74.3759 +trainer/Q1 Predictions Std 17.3239 +trainer/Q1 Predictions Max -0.795939 +trainer/Q1 Predictions Min -87.2845 +trainer/Q2 Predictions Mean -74.3676 +trainer/Q2 Predictions Std 17.3246 +trainer/Q2 Predictions Max -0.474676 +trainer/Q2 Predictions Min -87.2695 +trainer/Q Targets Mean -74.6043 +trainer/Q Targets Std 17.2727 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2748 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0289581 +trainer/policy/mean Std 0.725722 +trainer/policy/mean Max 0.999653 +trainer/policy/mean Min -0.998605 +trainer/policy/std Mean 0.415956 +trainer/policy/std Std 0.0210725 +trainer/policy/std Max 0.438068 +trainer/policy/std Min 0.382622 +trainer/Advantage Weights Mean 7.83078 +trainer/Advantage Weights Std 23.7259 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.08827e-17 +trainer/Advantage Score Mean -0.276703 +trainer/Advantage Score Std 0.64666 +trainer/Advantage Score Max 2.48834 +trainer/Advantage Score Min -3.77358 +trainer/V1 Predictions Mean -74.3437 +trainer/V1 Predictions Std 17.3634 +trainer/V1 Predictions Max -0.25025 +trainer/V1 Predictions Min -87.1088 +trainer/VF Loss 0.108305 +expl/num steps total 632000 +expl/num paths total 823 +expl/path length Mean 500 +expl/path length Std 243 +expl/path length Max 743 +expl/path length Min 257 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0324703 +expl/Actions Std 0.8187 +expl/Actions Max 2.5089 +expl/Actions Min -2.38947 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 583582 +eval/num paths total 635 +eval/path length Mean 554 +eval/path length Std 0 +eval/path length Max 554 +eval/path length Min 554 +eval/Rewards Mean 0.00180505 +eval/Rewards Std 0.0424476 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0370122 +eval/Actions Std 0.733316 +eval/Actions Max 0.999207 +eval/Actions Min -0.999171 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.11763e-05 +time/evaluation sampling (s) 5.42419 +time/exploration sampling (s) 6.14711 +time/logging (s) 0.0101429 +time/saving (s) 0.0144125 +time/training (s) 18.7557 +time/epoch (s) 30.3515 +time/total (s) 15689.1 +Epoch -369 +------------------------------ ---------------- +2022-05-15 22:24:27.742370 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -368 finished +------------------------------ ---------------- +epoch -368 +replay_buffer/size 999047 +trainer/num train calls 633000 +trainer/QF1 Loss 6.3375 +trainer/QF2 Loss 6.65678 +trainer/Policy Loss 19.3539 +trainer/Q1 Predictions Mean -72.299 +trainer/Q1 Predictions Std 19.42 +trainer/Q1 Predictions Max -0.269068 +trainer/Q1 Predictions Min -87.3989 +trainer/Q2 Predictions Mean -72.2594 +trainer/Q2 Predictions Std 19.3846 +trainer/Q2 Predictions Max -0.362824 +trainer/Q2 Predictions Min -87.2074 +trainer/Q Targets Mean -72.4942 +trainer/Q Targets Std 19.1787 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2621 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0265841 +trainer/policy/mean Std 0.70928 +trainer/policy/mean Max 0.999514 +trainer/policy/mean Min -0.997478 +trainer/policy/std Mean 0.414695 +trainer/policy/std Std 0.0196064 +trainer/policy/std Max 0.440783 +trainer/policy/std Min 0.383684 +trainer/Advantage Weights Mean 3.03346 +trainer/Advantage Weights Std 14.4964 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.05826e-22 +trainer/Advantage Score Mean -0.480534 +trainer/Advantage Score Std 0.651926 +trainer/Advantage Score Max 1.17696 +trainer/Advantage Score Min -5.06002 +trainer/V1 Predictions Mean -72.0942 +trainer/V1 Predictions Std 19.3825 +trainer/V1 Predictions Max 0.408295 +trainer/V1 Predictions Min -87.1139 +trainer/VF Loss 0.0806357 +expl/num steps total 633000 +expl/num paths total 825 +expl/path length Mean 500 +expl/path length Std 487 +expl/path length Max 987 +expl/path length Min 13 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0431993 +expl/Actions Std 0.828616 +expl/Actions Max 2.44125 +expl/Actions Min -2.39237 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 584471 +eval/num paths total 637 +eval/path length Mean 444.5 +eval/path length Std 9.5 +eval/path length Max 454 +eval/path length Min 435 +eval/Rewards Mean 0.00224972 +eval/Rewards Std 0.0473778 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0198088 +eval/Actions Std 0.734057 +eval/Actions Max 0.999869 +eval/Actions Min -0.999055 +eval/Num Paths 2 +eval/Average Returns 1 +time/data storing (s) 4.37954e-06 +time/evaluation sampling (s) 4.57919 +time/exploration sampling (s) 6.02327 +time/logging (s) 0.0148099 +time/saving (s) 0.0149885 +time/training (s) 19.3481 +time/epoch (s) 29.9804 +time/total (s) 15719.1 +Epoch -368 +------------------------------ ---------------- +2022-05-15 22:24:58.187044 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -367 finished +------------------------------ ---------------- +epoch -367 +replay_buffer/size 999047 +trainer/num train calls 634000 +trainer/QF1 Loss 5.24148 +trainer/QF2 Loss 5.11305 +trainer/Policy Loss 22.5746 +trainer/Q1 Predictions Mean -72.3734 +trainer/Q1 Predictions Std 18.965 +trainer/Q1 Predictions Max -0.72754 +trainer/Q1 Predictions Min -87.5981 +trainer/Q2 Predictions Mean -72.465 +trainer/Q2 Predictions Std 18.9781 +trainer/Q2 Predictions Max 0.0502765 +trainer/Q2 Predictions Min -87.5121 +trainer/Q Targets Mean -72.3449 +trainer/Q Targets Std 19.1612 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3872 +trainer/rewards Mean -0.976562 +trainer/rewards Std 0.151288 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0234375 +trainer/terminals Std 0.151288 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00123419 +trainer/policy/mean Std 0.725721 +trainer/policy/mean Max 0.999593 +trainer/policy/mean Min -0.999615 +trainer/policy/std Mean 0.415983 +trainer/policy/std Std 0.0194121 +trainer/policy/std Max 0.438041 +trainer/policy/std Min 0.383446 +trainer/Advantage Weights Mean 4.01151 +trainer/Advantage Weights Std 16.9466 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.70018e-14 +trainer/Advantage Score Mean -0.394836 +trainer/Advantage Score Std 0.5951 +trainer/Advantage Score Max 1.40751 +trainer/Advantage Score Min -3.09278 +trainer/V1 Predictions Mean -72.2328 +trainer/V1 Predictions Std 19.0289 +trainer/V1 Predictions Max -0.179275 +trainer/V1 Predictions Min -87.1618 +trainer/VF Loss 0.0703726 +expl/num steps total 634000 +expl/num paths total 827 +expl/path length Mean 500 +expl/path length Std 228 +expl/path length Max 728 +expl/path length Min 272 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0237003 +expl/Actions Std 0.82488 +expl/Actions Max 2.69725 +expl/Actions Min -2.23385 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 585471 +eval/num paths total 638 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.195125 +eval/Actions Std 0.730549 +eval/Actions Max 0.9987 +eval/Actions Min -0.998867 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0414e-05 +time/evaluation sampling (s) 4.61834 +time/exploration sampling (s) 7.04723 +time/logging (s) 0.010897 +time/saving (s) 0.0136742 +time/training (s) 18.7348 +time/epoch (s) 30.4249 +time/total (s) 15749.5 +Epoch -367 +------------------------------ ---------------- +2022-05-15 22:25:28.174922 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -366 finished +------------------------------ ---------------- +epoch -366 +replay_buffer/size 999047 +trainer/num train calls 635000 +trainer/QF1 Loss 0.624859 +trainer/QF2 Loss 0.515134 +trainer/Policy Loss 9.29689 +trainer/Q1 Predictions Mean -72.8485 +trainer/Q1 Predictions Std 18.7929 +trainer/Q1 Predictions Max -1.42793 +trainer/Q1 Predictions Min -86.7857 +trainer/Q2 Predictions Mean -72.8603 +trainer/Q2 Predictions Std 18.7967 +trainer/Q2 Predictions Max -1.04861 +trainer/Q2 Predictions Min -86.7402 +trainer/Q Targets Mean -73.0161 +trainer/Q Targets Std 18.7824 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2199 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0116284 +trainer/policy/mean Std 0.728587 +trainer/policy/mean Max 0.998896 +trainer/policy/mean Min -0.999407 +trainer/policy/std Mean 0.416563 +trainer/policy/std Std 0.0198719 +trainer/policy/std Max 0.436593 +trainer/policy/std Min 0.386564 +trainer/Advantage Weights Mean 3.88872 +trainer/Advantage Weights Std 16.884 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.50794e-16 +trainer/Advantage Score Mean -0.386432 +trainer/Advantage Score Std 0.53503 +trainer/Advantage Score Max 1.08381 +trainer/Advantage Score Min -3.49683 +trainer/V1 Predictions Mean -72.7961 +trainer/V1 Predictions Std 18.7724 +trainer/V1 Predictions Max -1.6175 +trainer/V1 Predictions Min -87.0759 +trainer/VF Loss 0.0588678 +expl/num steps total 635000 +expl/num paths total 829 +expl/path length Mean 500 +expl/path length Std 200 +expl/path length Max 700 +expl/path length Min 300 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.010613 +expl/Actions Std 0.841746 +expl/Actions Max 2.35045 +expl/Actions Min -2.26709 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 586471 +eval/num paths total 639 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0174472 +eval/Actions Std 0.728674 +eval/Actions Max 0.999777 +eval/Actions Min -0.999614 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.50084e-06 +time/evaluation sampling (s) 4.63625 +time/exploration sampling (s) 6.32984 +time/logging (s) 0.0110617 +time/saving (s) 0.0164629 +time/training (s) 18.9777 +time/epoch (s) 29.9713 +time/total (s) 15779.5 +Epoch -366 +------------------------------ ---------------- +2022-05-15 22:25:59.007320 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -365 finished +------------------------------ ---------------- +epoch -365 +replay_buffer/size 999047 +trainer/num train calls 636000 +trainer/QF1 Loss 0.968417 +trainer/QF2 Loss 0.780116 +trainer/Policy Loss 24.5797 +trainer/Q1 Predictions Mean -74.3732 +trainer/Q1 Predictions Std 18.0245 +trainer/Q1 Predictions Max -1.32009 +trainer/Q1 Predictions Min -87.4814 +trainer/Q2 Predictions Mean -74.492 +trainer/Q2 Predictions Std 18.0905 +trainer/Q2 Predictions Max -0.484642 +trainer/Q2 Predictions Min -87.6654 +trainer/Q Targets Mean -74.7249 +trainer/Q Targets Std 18.265 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.353 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0208531 +trainer/policy/mean Std 0.733395 +trainer/policy/mean Max 0.999669 +trainer/policy/mean Min -0.999117 +trainer/policy/std Mean 0.415996 +trainer/policy/std Std 0.0204664 +trainer/policy/std Max 0.441022 +trainer/policy/std Min 0.383865 +trainer/Advantage Weights Mean 5.32063 +trainer/Advantage Weights Std 18.3307 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.4331e-18 +trainer/Advantage Score Mean -0.233837 +trainer/Advantage Score Std 0.587915 +trainer/Advantage Score Max 2.12227 +trainer/Advantage Score Min -4.10867 +trainer/V1 Predictions Mean -74.5035 +trainer/V1 Predictions Std 18.2402 +trainer/V1 Predictions Max -0.368296 +trainer/V1 Predictions Min -87.436 +trainer/VF Loss 0.070993 +expl/num steps total 636000 +expl/num paths total 831 +expl/path length Mean 500 +expl/path length Std 376 +expl/path length Max 876 +expl/path length Min 124 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0302553 +expl/Actions Std 0.842875 +expl/Actions Max 2.31099 +expl/Actions Min -2.55943 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 587471 +eval/num paths total 640 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0156669 +eval/Actions Std 0.747963 +eval/Actions Max 0.999755 +eval/Actions Min -0.999536 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.05631e-05 +time/evaluation sampling (s) 4.56354 +time/exploration sampling (s) 7.07985 +time/logging (s) 0.0117629 +time/saving (s) 0.0136081 +time/training (s) 19.1469 +time/epoch (s) 30.8157 +time/total (s) 15810.3 +Epoch -365 +------------------------------ ---------------- +2022-05-15 22:26:30.097920 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -364 finished +------------------------------ ---------------- +epoch -364 +replay_buffer/size 999047 +trainer/num train calls 637000 +trainer/QF1 Loss 1.29142 +trainer/QF2 Loss 1.45836 +trainer/Policy Loss 7.20887 +trainer/Q1 Predictions Mean -72.5536 +trainer/Q1 Predictions Std 19.1589 +trainer/Q1 Predictions Max -0.309112 +trainer/Q1 Predictions Min -86.9875 +trainer/Q2 Predictions Mean -72.5126 +trainer/Q2 Predictions Std 19.2639 +trainer/Q2 Predictions Max -0.261828 +trainer/Q2 Predictions Min -87.193 +trainer/Q Targets Mean -72.3216 +trainer/Q Targets Std 19.002 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5083 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00649398 +trainer/policy/mean Std 0.72852 +trainer/policy/mean Max 0.999012 +trainer/policy/mean Min -0.999423 +trainer/policy/std Mean 0.416601 +trainer/policy/std Std 0.0213224 +trainer/policy/std Max 0.440957 +trainer/policy/std Min 0.382372 +trainer/Advantage Weights Mean 2.14017 +trainer/Advantage Weights Std 12.705 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.31e-13 +trainer/Advantage Score Mean -0.525924 +trainer/Advantage Score Std 0.458288 +trainer/Advantage Score Max 0.583067 +trainer/Advantage Score Min -2.96636 +trainer/V1 Predictions Mean -72.1558 +trainer/V1 Predictions Std 19.0963 +trainer/V1 Predictions Max 0.0401661 +trainer/V1 Predictions Min -87.147 +trainer/VF Loss 0.054013 +expl/num steps total 637000 +expl/num paths total 832 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.209419 +expl/Actions Std 0.859956 +expl/Actions Max 2.38477 +expl/Actions Min -2.54827 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 587950 +eval/num paths total 641 +eval/path length Mean 479 +eval/path length Std 0 +eval/path length Max 479 +eval/path length Min 479 +eval/Rewards Mean 0.00208768 +eval/Rewards Std 0.0456434 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0409504 +eval/Actions Std 0.738197 +eval/Actions Max 0.999636 +eval/Actions Min -0.998916 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.23642e-05 +time/evaluation sampling (s) 4.8848 +time/exploration sampling (s) 7.09071 +time/logging (s) 0.00855484 +time/saving (s) 0.0199359 +time/training (s) 19.0678 +time/epoch (s) 31.0718 +time/total (s) 15841.4 +Epoch -364 +------------------------------ ---------------- +2022-05-15 22:27:00.437182 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -363 finished +------------------------------ ---------------- +epoch -363 +replay_buffer/size 999047 +trainer/num train calls 638000 +trainer/QF1 Loss 0.584586 +trainer/QF2 Loss 0.553439 +trainer/Policy Loss 12.8106 +trainer/Q1 Predictions Mean -76.6036 +trainer/Q1 Predictions Std 14.1411 +trainer/Q1 Predictions Max -0.727463 +trainer/Q1 Predictions Min -87.6642 +trainer/Q2 Predictions Mean -76.6322 +trainer/Q2 Predictions Std 14.0749 +trainer/Q2 Predictions Max -1.0175 +trainer/Q2 Predictions Min -87.764 +trainer/Q Targets Mean -76.5148 +trainer/Q Targets Std 14.0088 +trainer/Q Targets Max -2.4602 +trainer/Q Targets Min -87.3858 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.012299 +trainer/policy/mean Std 0.732411 +trainer/policy/mean Max 0.999311 +trainer/policy/mean Min -0.999431 +trainer/policy/std Mean 0.415395 +trainer/policy/std Std 0.0191137 +trainer/policy/std Max 0.435994 +trainer/policy/std Min 0.385509 +trainer/Advantage Weights Mean 2.52001 +trainer/Advantage Weights Std 13.0873 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.07791e-12 +trainer/Advantage Score Mean -0.345356 +trainer/Advantage Score Std 0.491293 +trainer/Advantage Score Max 1.79663 +trainer/Advantage Score Min -2.7556 +trainer/V1 Predictions Mean -76.2847 +trainer/V1 Predictions Std 14.069 +trainer/V1 Predictions Max -1.20723 +trainer/V1 Predictions Min -87.3179 +trainer/VF Loss 0.0568172 +expl/num steps total 638000 +expl/num paths total 833 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0232861 +expl/Actions Std 0.824344 +expl/Actions Max 2.32858 +expl/Actions Min -2.31017 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 588950 +eval/num paths total 642 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.224947 +eval/Actions Std 0.736963 +eval/Actions Max 0.999284 +eval/Actions Min -0.999632 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0048e-05 +time/evaluation sampling (s) 4.69851 +time/exploration sampling (s) 6.49572 +time/logging (s) 0.0089898 +time/saving (s) 0.0143031 +time/training (s) 19.1051 +time/epoch (s) 30.3226 +time/total (s) 15871.7 +Epoch -363 +------------------------------ ---------------- +2022-05-15 22:27:31.513857 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -362 finished +------------------------------ ---------------- +epoch -362 +replay_buffer/size 999047 +trainer/num train calls 639000 +trainer/QF1 Loss 0.79119 +trainer/QF2 Loss 0.873043 +trainer/Policy Loss 24.6038 +trainer/Q1 Predictions Mean -74.1191 +trainer/Q1 Predictions Std 17.0322 +trainer/Q1 Predictions Max -2.2286 +trainer/Q1 Predictions Min -86.8641 +trainer/Q2 Predictions Mean -74.1685 +trainer/Q2 Predictions Std 16.9596 +trainer/Q2 Predictions Max -2.4585 +trainer/Q2 Predictions Min -86.7458 +trainer/Q Targets Mean -74.2889 +trainer/Q Targets Std 17.4563 +trainer/Q Targets Max -2.24629 +trainer/Q Targets Min -87.4473 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00805055 +trainer/policy/mean Std 0.727275 +trainer/policy/mean Max 0.999918 +trainer/policy/mean Min -0.99813 +trainer/policy/std Mean 0.41557 +trainer/policy/std Std 0.018389 +trainer/policy/std Max 0.435682 +trainer/policy/std Min 0.389183 +trainer/Advantage Weights Mean 6.42247 +trainer/Advantage Weights Std 20.7027 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.51323e-20 +trainer/Advantage Score Mean -0.310001 +trainer/Advantage Score Std 0.7465 +trainer/Advantage Score Max 1.29388 +trainer/Advantage Score Min -4.56375 +trainer/V1 Predictions Mean -74.059 +trainer/V1 Predictions Std 17.5266 +trainer/V1 Predictions Max -1.17843 +trainer/V1 Predictions Min -87.2127 +trainer/VF Loss 0.0937493 +expl/num steps total 639000 +expl/num paths total 834 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0362319 +expl/Actions Std 0.832021 +expl/Actions Max 2.35207 +expl/Actions Min -2.2826 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 589950 +eval/num paths total 643 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.100294 +eval/Actions Std 0.768532 +eval/Actions Max 0.999948 +eval/Actions Min -0.999453 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.55304e-06 +time/evaluation sampling (s) 4.98322 +time/exploration sampling (s) 6.81757 +time/logging (s) 0.0118086 +time/saving (s) 0.0157993 +time/training (s) 19.2344 +time/epoch (s) 31.0628 +time/total (s) 15902.8 +Epoch -362 +------------------------------ ---------------- +2022-05-15 22:28:01.049238 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -361 finished +------------------------------ ---------------- +epoch -361 +replay_buffer/size 999047 +trainer/num train calls 640000 +trainer/QF1 Loss 1.71003 +trainer/QF2 Loss 1.58814 +trainer/Policy Loss 29.1567 +trainer/Q1 Predictions Mean -71.6372 +trainer/Q1 Predictions Std 19.5777 +trainer/Q1 Predictions Max -1.45053 +trainer/Q1 Predictions Min -87.8753 +trainer/Q2 Predictions Mean -71.6099 +trainer/Q2 Predictions Std 19.6047 +trainer/Q2 Predictions Max 0.238539 +trainer/Q2 Predictions Min -87.9972 +trainer/Q Targets Mean -71.9644 +trainer/Q Targets Std 19.256 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1262 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0142809 +trainer/policy/mean Std 0.719959 +trainer/policy/mean Max 0.999138 +trainer/policy/mean Min -0.997379 +trainer/policy/std Mean 0.415072 +trainer/policy/std Std 0.019979 +trainer/policy/std Max 0.437814 +trainer/policy/std Min 0.385722 +trainer/Advantage Weights Mean 8.30988 +trainer/Advantage Weights Std 24.5289 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.32401e-14 +trainer/Advantage Score Mean -0.191827 +trainer/Advantage Score Std 0.538144 +trainer/Advantage Score Max 2.53418 +trainer/Advantage Score Min -3.0564 +trainer/V1 Predictions Mean -71.7768 +trainer/V1 Predictions Std 19.2524 +trainer/V1 Predictions Max -1.30048 +trainer/V1 Predictions Min -87.9704 +trainer/VF Loss 0.0942318 +expl/num steps total 640000 +expl/num paths total 836 +expl/path length Mean 500 +expl/path length Std 460 +expl/path length Max 960 +expl/path length Min 40 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0216272 +expl/Actions Std 0.824471 +expl/Actions Max 2.26058 +expl/Actions Min -2.357 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 590702 +eval/num paths total 644 +eval/path length Mean 752 +eval/path length Std 0 +eval/path length Max 752 +eval/path length Min 752 +eval/Rewards Mean 0.00132979 +eval/Rewards Std 0.036442 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0171825 +eval/Actions Std 0.747629 +eval/Actions Max 0.99998 +eval/Actions Min -0.999679 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 9.14186e-06 +time/evaluation sampling (s) 4.55732 +time/exploration sampling (s) 6.43228 +time/logging (s) 0.0120874 +time/saving (s) 0.0125639 +time/training (s) 18.4949 +time/epoch (s) 29.5091 +time/total (s) 15932.3 +Epoch -361 +------------------------------ ---------------- +2022-05-15 22:28:31.415782 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -360 finished +------------------------------ ---------------- +epoch -360 +replay_buffer/size 999047 +trainer/num train calls 641000 +trainer/QF1 Loss 0.911385 +trainer/QF2 Loss 0.901382 +trainer/Policy Loss 17.0998 +trainer/Q1 Predictions Mean -73.0833 +trainer/Q1 Predictions Std 18.4449 +trainer/Q1 Predictions Max -0.614348 +trainer/Q1 Predictions Min -86.7819 +trainer/Q2 Predictions Mean -73.1002 +trainer/Q2 Predictions Std 18.498 +trainer/Q2 Predictions Max -0.412135 +trainer/Q2 Predictions Min -86.7953 +trainer/Q Targets Mean -73.0906 +trainer/Q Targets Std 18.4716 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.237 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00260422 +trainer/policy/mean Std 0.734593 +trainer/policy/mean Max 0.999898 +trainer/policy/mean Min -0.997765 +trainer/policy/std Mean 0.415285 +trainer/policy/std Std 0.0200988 +trainer/policy/std Max 0.436569 +trainer/policy/std Min 0.385793 +trainer/Advantage Weights Mean 4.72816 +trainer/Advantage Weights Std 16.5633 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.35731e-19 +trainer/Advantage Score Mean -0.359243 +trainer/Advantage Score Std 0.629305 +trainer/Advantage Score Max 2.10905 +trainer/Advantage Score Min -4.28916 +trainer/V1 Predictions Mean -72.8849 +trainer/V1 Predictions Std 18.5144 +trainer/V1 Predictions Max -0.97914 +trainer/V1 Predictions Min -87.1012 +trainer/VF Loss 0.0824043 +expl/num steps total 641000 +expl/num paths total 838 +expl/path length Mean 500 +expl/path length Std 479 +expl/path length Max 979 +expl/path length Min 21 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0348967 +expl/Actions Std 0.812617 +expl/Actions Max 2.40323 +expl/Actions Min -2.14248 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 591702 +eval/num paths total 645 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.125808 +eval/Actions Std 0.634901 +eval/Actions Max 0.99822 +eval/Actions Min -0.998957 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.93204e-06 +time/evaluation sampling (s) 5.10702 +time/exploration sampling (s) 6.72552 +time/logging (s) 0.0114821 +time/saving (s) 0.0126341 +time/training (s) 18.4823 +time/epoch (s) 30.3389 +time/total (s) 15962.6 +Epoch -360 +------------------------------ ---------------- +2022-05-15 22:29:01.690782 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -359 finished +------------------------------ ---------------- +epoch -359 +replay_buffer/size 999047 +trainer/num train calls 642000 +trainer/QF1 Loss 0.986564 +trainer/QF2 Loss 0.751884 +trainer/Policy Loss 22.0954 +trainer/Q1 Predictions Mean -72.4375 +trainer/Q1 Predictions Std 19.6259 +trainer/Q1 Predictions Max -1.49863 +trainer/Q1 Predictions Min -87.6379 +trainer/Q2 Predictions Mean -72.4834 +trainer/Q2 Predictions Std 19.6734 +trainer/Q2 Predictions Max -1.40538 +trainer/Q2 Predictions Min -87.8373 +trainer/Q Targets Mean -72.8858 +trainer/Q Targets Std 19.5006 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.124 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0120505 +trainer/policy/mean Std 0.729434 +trainer/policy/mean Max 0.999305 +trainer/policy/mean Min -0.99929 +trainer/policy/std Mean 0.415348 +trainer/policy/std Std 0.0195486 +trainer/policy/std Max 0.434644 +trainer/policy/std Min 0.385145 +trainer/Advantage Weights Mean 8.52038 +trainer/Advantage Weights Std 25.9552 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.86708e-14 +trainer/Advantage Score Mean -0.270229 +trainer/Advantage Score Std 0.570362 +trainer/Advantage Score Max 1.98972 +trainer/Advantage Score Min -2.9947 +trainer/V1 Predictions Mean -72.6155 +trainer/V1 Predictions Std 19.7068 +trainer/V1 Predictions Max -1.39827 +trainer/V1 Predictions Min -87.875 +trainer/VF Loss 0.0948744 +expl/num steps total 642000 +expl/num paths total 840 +expl/path length Mean 500 +expl/path length Std 372 +expl/path length Max 872 +expl/path length Min 128 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0287921 +expl/Actions Std 0.816148 +expl/Actions Max 2.21022 +expl/Actions Min -2.38037 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 592702 +eval/num paths total 646 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0251619 +eval/Actions Std 0.742301 +eval/Actions Max 0.999544 +eval/Actions Min -0.999407 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.33789e-05 +time/evaluation sampling (s) 4.88188 +time/exploration sampling (s) 6.48496 +time/logging (s) 0.0123306 +time/saving (s) 0.0182737 +time/training (s) 18.8631 +time/epoch (s) 30.2606 +time/total (s) 15992.9 +Epoch -359 +------------------------------ ---------------- +2022-05-15 22:29:32.406466 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -358 finished +------------------------------ ---------------- +epoch -358 +replay_buffer/size 999047 +trainer/num train calls 643000 +trainer/QF1 Loss 0.576069 +trainer/QF2 Loss 0.553793 +trainer/Policy Loss 16.795 +trainer/Q1 Predictions Mean -73.4183 +trainer/Q1 Predictions Std 19.2884 +trainer/Q1 Predictions Max -0.74855 +trainer/Q1 Predictions Min -87.575 +trainer/Q2 Predictions Mean -73.4055 +trainer/Q2 Predictions Std 19.3459 +trainer/Q2 Predictions Max -0.317194 +trainer/Q2 Predictions Min -87.4186 +trainer/Q Targets Mean -73.3535 +trainer/Q Targets Std 19.2211 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.219 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0152244 +trainer/policy/mean Std 0.728555 +trainer/policy/mean Max 0.999744 +trainer/policy/mean Min -0.999366 +trainer/policy/std Mean 0.416321 +trainer/policy/std Std 0.0198438 +trainer/policy/std Max 0.438756 +trainer/policy/std Min 0.383104 +trainer/Advantage Weights Mean 3.52271 +trainer/Advantage Weights Std 14.9069 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.12489e-09 +trainer/Advantage Score Mean -0.276351 +trainer/Advantage Score Std 0.437093 +trainer/Advantage Score Max 1.24984 +trainer/Advantage Score Min -2.06056 +trainer/V1 Predictions Mean -73.105 +trainer/V1 Predictions Std 19.4193 +trainer/V1 Predictions Max 0.767107 +trainer/V1 Predictions Min -87.0899 +trainer/VF Loss 0.0419911 +expl/num steps total 643000 +expl/num paths total 841 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.037724 +expl/Actions Std 0.833577 +expl/Actions Max 2.33649 +expl/Actions Min -2.48658 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 593702 +eval/num paths total 647 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0534209 +eval/Actions Std 0.734941 +eval/Actions Max 0.999892 +eval/Actions Min -0.999557 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.16068e-06 +time/evaluation sampling (s) 4.74645 +time/exploration sampling (s) 6.41917 +time/logging (s) 0.0119245 +time/saving (s) 0.0154541 +time/training (s) 19.5038 +time/epoch (s) 30.6969 +time/total (s) 16023.6 +Epoch -358 +------------------------------ ---------------- +2022-05-15 22:30:02.256326 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -357 finished +------------------------------ ---------------- +epoch -357 +replay_buffer/size 999047 +trainer/num train calls 644000 +trainer/QF1 Loss 0.675748 +trainer/QF2 Loss 0.696277 +trainer/Policy Loss 18.153 +trainer/Q1 Predictions Mean -75.3361 +trainer/Q1 Predictions Std 15.6936 +trainer/Q1 Predictions Max -0.960386 +trainer/Q1 Predictions Min -87.7854 +trainer/Q2 Predictions Mean -75.424 +trainer/Q2 Predictions Std 15.6581 +trainer/Q2 Predictions Max -1.4526 +trainer/Q2 Predictions Min -87.9453 +trainer/Q Targets Mean -75.0013 +trainer/Q Targets Std 15.8214 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6544 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.01254 +trainer/policy/mean Std 0.744123 +trainer/policy/mean Max 0.999917 +trainer/policy/mean Min -0.999555 +trainer/policy/std Mean 0.414789 +trainer/policy/std Std 0.0187367 +trainer/policy/std Max 0.435586 +trainer/policy/std Min 0.385444 +trainer/Advantage Weights Mean 4.03744 +trainer/Advantage Weights Std 16.5105 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.97483e-12 +trainer/Advantage Score Mean -0.370475 +trainer/Advantage Score Std 0.601616 +trainer/Advantage Score Max 2.70102 +trainer/Advantage Score Min -2.65408 +trainer/V1 Predictions Mean -74.7903 +trainer/V1 Predictions Std 15.7792 +trainer/V1 Predictions Max -0.0876841 +trainer/V1 Predictions Min -87.427 +trainer/VF Loss 0.087905 +expl/num steps total 644000 +expl/num paths total 842 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0658064 +expl/Actions Std 0.831078 +expl/Actions Max 2.49518 +expl/Actions Min -2.41901 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 594702 +eval/num paths total 648 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0279367 +eval/Actions Std 0.821033 +eval/Actions Max 0.999385 +eval/Actions Min -0.998155 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.53255e-06 +time/evaluation sampling (s) 4.57254 +time/exploration sampling (s) 6.32362 +time/logging (s) 0.00767196 +time/saving (s) 0.0119033 +time/training (s) 18.914 +time/epoch (s) 29.8297 +time/total (s) 16053.5 +Epoch -357 +------------------------------ ---------------- +2022-05-15 22:30:32.807584 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -356 finished +------------------------------ ---------------- +epoch -356 +replay_buffer/size 999047 +trainer/num train calls 645000 +trainer/QF1 Loss 1.13089 +trainer/QF2 Loss 0.944007 +trainer/Policy Loss 41.8672 +trainer/Q1 Predictions Mean -73.1104 +trainer/Q1 Predictions Std 17.8387 +trainer/Q1 Predictions Max -2.32947 +trainer/Q1 Predictions Min -87.2463 +trainer/Q2 Predictions Mean -73.1704 +trainer/Q2 Predictions Std 17.8389 +trainer/Q2 Predictions Max -1.13297 +trainer/Q2 Predictions Min -87.3495 +trainer/Q Targets Mean -73.5824 +trainer/Q Targets Std 18.0468 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8361 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00458203 +trainer/policy/mean Std 0.718839 +trainer/policy/mean Max 0.998305 +trainer/policy/mean Min -0.998678 +trainer/policy/std Mean 0.413279 +trainer/policy/std Std 0.0189671 +trainer/policy/std Max 0.434548 +trainer/policy/std Min 0.383682 +trainer/Advantage Weights Mean 10.0434 +trainer/Advantage Weights Std 25.7735 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.82288e-19 +trainer/Advantage Score Mean -0.141854 +trainer/Advantage Score Std 0.522756 +trainer/Advantage Score Max 1.10868 +trainer/Advantage Score Min -4.18288 +trainer/V1 Predictions Mean -73.394 +trainer/V1 Predictions Std 17.9491 +trainer/V1 Predictions Max -3.16104 +trainer/V1 Predictions Min -87.703 +trainer/VF Loss 0.0660108 +expl/num steps total 645000 +expl/num paths total 843 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0525794 +expl/Actions Std 0.805586 +expl/Actions Max 2.29925 +expl/Actions Min -2.27939 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 595702 +eval/num paths total 649 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.194596 +eval/Actions Std 0.76956 +eval/Actions Max 0.999939 +eval/Actions Min -0.99969 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.36392e-05 +time/evaluation sampling (s) 5.19061 +time/exploration sampling (s) 6.45471 +time/logging (s) 0.0117657 +time/saving (s) 0.0148645 +time/training (s) 18.871 +time/epoch (s) 30.5429 +time/total (s) 16084 +Epoch -356 +------------------------------ ---------------- +2022-05-15 22:31:03.319314 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -355 finished +------------------------------ ---------------- +epoch -355 +replay_buffer/size 999047 +trainer/num train calls 646000 +trainer/QF1 Loss 0.606315 +trainer/QF2 Loss 0.626907 +trainer/Policy Loss 30.6939 +trainer/Q1 Predictions Mean -74.6786 +trainer/Q1 Predictions Std 15.7629 +trainer/Q1 Predictions Max -4.10673 +trainer/Q1 Predictions Min -88.2397 +trainer/Q2 Predictions Mean -74.7013 +trainer/Q2 Predictions Std 15.8188 +trainer/Q2 Predictions Max -4.0562 +trainer/Q2 Predictions Min -87.841 +trainer/Q Targets Mean -74.8611 +trainer/Q Targets Std 15.914 +trainer/Q Targets Max -3.09571 +trainer/Q Targets Min -88.3993 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00238645 +trainer/policy/mean Std 0.721858 +trainer/policy/mean Max 0.999769 +trainer/policy/mean Min -0.99965 +trainer/policy/std Mean 0.413015 +trainer/policy/std Std 0.0192432 +trainer/policy/std Max 0.435084 +trainer/policy/std Min 0.383708 +trainer/Advantage Weights Mean 6.57258 +trainer/Advantage Weights Std 21.5247 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.88792e-14 +trainer/Advantage Score Mean -0.28826 +trainer/Advantage Score Std 0.608654 +trainer/Advantage Score Max 1.67595 +trainer/Advantage Score Min -3.00515 +trainer/V1 Predictions Mean -74.6384 +trainer/V1 Predictions Std 15.9992 +trainer/V1 Predictions Max -3.54266 +trainer/V1 Predictions Min -88.3378 +trainer/VF Loss 0.0761469 +expl/num steps total 646000 +expl/num paths total 844 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.210326 +expl/Actions Std 0.845871 +expl/Actions Max 2.283 +expl/Actions Min -2.36658 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 596044 +eval/num paths total 650 +eval/path length Mean 342 +eval/path length Std 0 +eval/path length Max 342 +eval/path length Min 342 +eval/Rewards Mean 0.00292398 +eval/Rewards Std 0.0539947 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0143721 +eval/Actions Std 0.755139 +eval/Actions Max 0.99961 +eval/Actions Min -0.997767 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.32942e-05 +time/evaluation sampling (s) 4.82737 +time/exploration sampling (s) 6.73135 +time/logging (s) 0.00940209 +time/saving (s) 0.0188118 +time/training (s) 18.9067 +time/epoch (s) 30.4937 +time/total (s) 16114.5 +Epoch -355 +------------------------------ ---------------- +2022-05-15 22:31:35.406968 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -354 finished +------------------------------ ---------------- +epoch -354 +replay_buffer/size 999047 +trainer/num train calls 647000 +trainer/QF1 Loss 1.25026 +trainer/QF2 Loss 1.1986 +trainer/Policy Loss 16.1425 +trainer/Q1 Predictions Mean -73.1649 +trainer/Q1 Predictions Std 18.3102 +trainer/Q1 Predictions Max -0.218052 +trainer/Q1 Predictions Min -86.8703 +trainer/Q2 Predictions Mean -73.1039 +trainer/Q2 Predictions Std 18.2514 +trainer/Q2 Predictions Max 0.830953 +trainer/Q2 Predictions Min -86.6371 +trainer/Q Targets Mean -73.3877 +trainer/Q Targets Std 18.2828 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7977 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0239424 +trainer/policy/mean Std 0.733338 +trainer/policy/mean Max 0.998348 +trainer/policy/mean Min -0.999439 +trainer/policy/std Mean 0.413634 +trainer/policy/std Std 0.019924 +trainer/policy/std Max 0.439815 +trainer/policy/std Min 0.382984 +trainer/Advantage Weights Mean 3.28569 +trainer/Advantage Weights Std 14.5797 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.18671e-13 +trainer/Advantage Score Mean -0.392648 +trainer/Advantage Score Std 0.531161 +trainer/Advantage Score Max 1.08774 +trainer/Advantage Score Min -2.97624 +trainer/V1 Predictions Mean -73.1019 +trainer/V1 Predictions Std 18.3229 +trainer/V1 Predictions Max -0.377659 +trainer/V1 Predictions Min -86.734 +trainer/VF Loss 0.056768 +expl/num steps total 647000 +expl/num paths total 845 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0258762 +expl/Actions Std 0.824377 +expl/Actions Max 2.16174 +expl/Actions Min -2.16384 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 597044 +eval/num paths total 651 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0224829 +eval/Actions Std 0.609372 +eval/Actions Max 0.998866 +eval/Actions Min -0.999135 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0299e-05 +time/evaluation sampling (s) 5.28494 +time/exploration sampling (s) 7.31786 +time/logging (s) 0.00963908 +time/saving (s) 0.0149356 +time/training (s) 19.4421 +time/epoch (s) 32.0695 +time/total (s) 16146.6 +Epoch -354 +------------------------------ ---------------- +2022-05-15 22:32:06.287817 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -353 finished +------------------------------ ---------------- +epoch -353 +replay_buffer/size 999047 +trainer/num train calls 648000 +trainer/QF1 Loss 1.17857 +trainer/QF2 Loss 1.03491 +trainer/Policy Loss 24.9697 +trainer/Q1 Predictions Mean -74.1108 +trainer/Q1 Predictions Std 17.0855 +trainer/Q1 Predictions Max -1.26449 +trainer/Q1 Predictions Min -87.0924 +trainer/Q2 Predictions Mean -74.0961 +trainer/Q2 Predictions Std 17.1043 +trainer/Q2 Predictions Max -0.701028 +trainer/Q2 Predictions Min -86.8866 +trainer/Q Targets Mean -74.0774 +trainer/Q Targets Std 17.3858 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3014 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0207603 +trainer/policy/mean Std 0.715615 +trainer/policy/mean Max 0.998768 +trainer/policy/mean Min -0.998529 +trainer/policy/std Mean 0.414366 +trainer/policy/std Std 0.0190679 +trainer/policy/std Max 0.438101 +trainer/policy/std Min 0.384286 +trainer/Advantage Weights Mean 3.49935 +trainer/Advantage Weights Std 15.723 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.88332e-21 +trainer/Advantage Score Mean -0.377162 +trainer/Advantage Score Std 0.577528 +trainer/Advantage Score Max 0.84671 +trainer/Advantage Score Min -4.77213 +trainer/V1 Predictions Mean -73.8545 +trainer/V1 Predictions Std 17.4928 +trainer/V1 Predictions Max 0.136007 +trainer/V1 Predictions Min -87.0176 +trainer/VF Loss 0.059776 +expl/num steps total 648000 +expl/num paths total 846 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.372082 +expl/Actions Std 0.808298 +expl/Actions Max 2.24285 +expl/Actions Min -2.43309 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 598044 +eval/num paths total 652 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0445859 +eval/Actions Std 0.751511 +eval/Actions Max 0.999884 +eval/Actions Min -0.999046 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.23829e-05 +time/evaluation sampling (s) 5.15681 +time/exploration sampling (s) 6.8074 +time/logging (s) 0.0114243 +time/saving (s) 0.0146582 +time/training (s) 18.8779 +time/epoch (s) 30.8682 +time/total (s) 16177.5 +Epoch -353 +------------------------------ ---------------- +2022-05-15 22:32:37.578565 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -352 finished +------------------------------ ---------------- +epoch -352 +replay_buffer/size 999047 +trainer/num train calls 649000 +trainer/QF1 Loss 0.998565 +trainer/QF2 Loss 0.898962 +trainer/Policy Loss 7.31809 +trainer/Q1 Predictions Mean -72.5926 +trainer/Q1 Predictions Std 20.2294 +trainer/Q1 Predictions Max -0.274605 +trainer/Q1 Predictions Min -87.8044 +trainer/Q2 Predictions Mean -72.6116 +trainer/Q2 Predictions Std 20.2974 +trainer/Q2 Predictions Max -0.0429588 +trainer/Q2 Predictions Min -88.0795 +trainer/Q Targets Mean -72.3458 +trainer/Q Targets Std 20.1429 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2949 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000243211 +trainer/policy/mean Std 0.726445 +trainer/policy/mean Max 0.999365 +trainer/policy/mean Min -0.998709 +trainer/policy/std Mean 0.414174 +trainer/policy/std Std 0.020627 +trainer/policy/std Max 0.439886 +trainer/policy/std Min 0.383188 +trainer/Advantage Weights Mean 2.30211 +trainer/Advantage Weights Std 11.6792 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.40546e-14 +trainer/Advantage Score Mean -0.475461 +trainer/Advantage Score Std 0.518865 +trainer/Advantage Score Max 1.28429 +trainer/Advantage Score Min -3.05488 +trainer/V1 Predictions Mean -72.0878 +trainer/V1 Predictions Std 20.2811 +trainer/V1 Predictions Max 0.119948 +trainer/V1 Predictions Min -87.1666 +trainer/VF Loss 0.0595924 +expl/num steps total 649000 +expl/num paths total 847 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0367497 +expl/Actions Std 0.808172 +expl/Actions Max 2.42266 +expl/Actions Min -2.44913 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 599044 +eval/num paths total 653 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.143239 +eval/Actions Std 0.703946 +eval/Actions Max 0.999727 +eval/Actions Min -0.999565 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.65688e-06 +time/evaluation sampling (s) 5.17365 +time/exploration sampling (s) 6.92223 +time/logging (s) 0.0114072 +time/saving (s) 0.017288 +time/training (s) 19.1504 +time/epoch (s) 31.275 +time/total (s) 16208.7 +Epoch -352 +------------------------------ ---------------- +2022-05-15 22:33:08.053299 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -351 finished +------------------------------ ---------------- +epoch -351 +replay_buffer/size 999047 +trainer/num train calls 650000 +trainer/QF1 Loss 1.32723 +trainer/QF2 Loss 1.16386 +trainer/Policy Loss 35.5689 +trainer/Q1 Predictions Mean -73.8836 +trainer/Q1 Predictions Std 17.5163 +trainer/Q1 Predictions Max -1.20045 +trainer/Q1 Predictions Min -87.4381 +trainer/Q2 Predictions Mean -73.8272 +trainer/Q2 Predictions Std 17.5985 +trainer/Q2 Predictions Max -1.25542 +trainer/Q2 Predictions Min -87.5989 +trainer/Q Targets Mean -73.9187 +trainer/Q Targets Std 18.0051 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7338 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0020734 +trainer/policy/mean Std 0.726193 +trainer/policy/mean Max 0.998643 +trainer/policy/mean Min -0.99903 +trainer/policy/std Mean 0.415144 +trainer/policy/std Std 0.0210566 +trainer/policy/std Max 0.440358 +trainer/policy/std Min 0.381994 +trainer/Advantage Weights Mean 7.41361 +trainer/Advantage Weights Std 21.5676 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.26624e-14 +trainer/Advantage Score Mean -0.237067 +trainer/Advantage Score Std 0.580822 +trainer/Advantage Score Max 1.74375 +trainer/Advantage Score Min -3.05749 +trainer/V1 Predictions Mean -73.8125 +trainer/V1 Predictions Std 17.7565 +trainer/V1 Predictions Max -0.336957 +trainer/V1 Predictions Min -87.7232 +trainer/VF Loss 0.0825159 +expl/num steps total 650000 +expl/num paths total 848 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0827717 +expl/Actions Std 0.86917 +expl/Actions Max 2.10554 +expl/Actions Min -2.4897 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 600044 +eval/num paths total 654 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.197731 +eval/Actions Std 0.715412 +eval/Actions Max 0.998155 +eval/Actions Min -0.997408 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.40681e-05 +time/evaluation sampling (s) 4.76981 +time/exploration sampling (s) 6.36711 +time/logging (s) 0.0110846 +time/saving (s) 0.0150557 +time/training (s) 19.2939 +time/epoch (s) 30.4569 +time/total (s) 16239.2 +Epoch -351 +------------------------------ ---------------- +2022-05-15 22:33:38.171776 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -350 finished +------------------------------ ---------------- +epoch -350 +replay_buffer/size 999047 +trainer/num train calls 651000 +trainer/QF1 Loss 0.527978 +trainer/QF2 Loss 0.70364 +trainer/Policy Loss 21.949 +trainer/Q1 Predictions Mean -74.8884 +trainer/Q1 Predictions Std 15.5013 +trainer/Q1 Predictions Max -1.43903 +trainer/Q1 Predictions Min -87.1599 +trainer/Q2 Predictions Mean -74.9562 +trainer/Q2 Predictions Std 15.5587 +trainer/Q2 Predictions Max -1.50796 +trainer/Q2 Predictions Min -87.383 +trainer/Q Targets Mean -74.789 +trainer/Q Targets Std 15.5461 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9988 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0134826 +trainer/policy/mean Std 0.724661 +trainer/policy/mean Max 0.999763 +trainer/policy/mean Min -0.999151 +trainer/policy/std Mean 0.412538 +trainer/policy/std Std 0.0202741 +trainer/policy/std Max 0.434055 +trainer/policy/std Min 0.380421 +trainer/Advantage Weights Mean 5.38155 +trainer/Advantage Weights Std 20.7607 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.03021e-16 +trainer/Advantage Score Mean -0.471249 +trainer/Advantage Score Std 0.567864 +trainer/Advantage Score Max 1.41332 +trainer/Advantage Score Min -3.52259 +trainer/V1 Predictions Mean -74.5625 +trainer/V1 Predictions Std 15.5291 +trainer/V1 Predictions Max -0.92798 +trainer/V1 Predictions Min -86.8504 +trainer/VF Loss 0.0736627 +expl/num steps total 651000 +expl/num paths total 850 +expl/path length Mean 500 +expl/path length Std 355 +expl/path length Max 855 +expl/path length Min 145 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0328814 +expl/Actions Std 0.814616 +expl/Actions Max 2.28313 +expl/Actions Min -2.16492 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 600924 +eval/num paths total 655 +eval/path length Mean 880 +eval/path length Std 0 +eval/path length Max 880 +eval/path length Min 880 +eval/Rewards Mean 0.00113636 +eval/Rewards Std 0.0336908 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0786459 +eval/Actions Std 0.692124 +eval/Actions Max 0.999793 +eval/Actions Min -0.999205 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.45265e-06 +time/evaluation sampling (s) 4.88804 +time/exploration sampling (s) 5.94005 +time/logging (s) 0.0121815 +time/saving (s) 0.0316775 +time/training (s) 19.2296 +time/epoch (s) 30.1016 +time/total (s) 16269.3 +Epoch -350 +------------------------------ ---------------- +2022-05-15 22:34:08.266407 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -349 finished +------------------------------ ---------------- +epoch -349 +replay_buffer/size 999047 +trainer/num train calls 652000 +trainer/QF1 Loss 1.61695 +trainer/QF2 Loss 1.5536 +trainer/Policy Loss 10.5659 +trainer/Q1 Predictions Mean -72.9465 +trainer/Q1 Predictions Std 19.4597 +trainer/Q1 Predictions Max -6.61549 +trainer/Q1 Predictions Min -87.0654 +trainer/Q2 Predictions Mean -72.8639 +trainer/Q2 Predictions Std 19.4857 +trainer/Q2 Predictions Max -4.97986 +trainer/Q2 Predictions Min -86.9755 +trainer/Q Targets Mean -72.7486 +trainer/Q Targets Std 19.3707 +trainer/Q Targets Max -3.08977 +trainer/Q Targets Min -87.0833 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0254109 +trainer/policy/mean Std 0.740312 +trainer/policy/mean Max 0.999109 +trainer/policy/mean Min -0.999663 +trainer/policy/std Mean 0.413193 +trainer/policy/std Std 0.0194474 +trainer/policy/std Max 0.436383 +trainer/policy/std Min 0.382215 +trainer/Advantage Weights Mean 3.12744 +trainer/Advantage Weights Std 15.0538 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.48928e-23 +trainer/Advantage Score Mean -0.50493 +trainer/Advantage Score Std 0.636271 +trainer/Advantage Score Max 1.13315 +trainer/Advantage Score Min -5.10893 +trainer/V1 Predictions Mean -72.6067 +trainer/V1 Predictions Std 19.3775 +trainer/V1 Predictions Max -4.71248 +trainer/V1 Predictions Min -86.6467 +trainer/VF Loss 0.0801958 +expl/num steps total 652000 +expl/num paths total 852 +expl/path length Mean 500 +expl/path length Std 255 +expl/path length Max 755 +expl/path length Min 245 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0194223 +expl/Actions Std 0.851236 +expl/Actions Max 2.33824 +expl/Actions Min -2.31297 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 601924 +eval/num paths total 656 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.116545 +eval/Actions Std 0.783649 +eval/Actions Max 0.999991 +eval/Actions Min -0.999598 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.38981e-05 +time/evaluation sampling (s) 4.15476 +time/exploration sampling (s) 6.8143 +time/logging (s) 0.00948713 +time/saving (s) 0.0201523 +time/training (s) 19.0682 +time/epoch (s) 30.067 +time/total (s) 16299.4 +Epoch -349 +------------------------------ ---------------- +2022-05-15 22:34:38.533385 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -348 finished +------------------------------ ---------------- +epoch -348 +replay_buffer/size 999047 +trainer/num train calls 653000 +trainer/QF1 Loss 0.921552 +trainer/QF2 Loss 0.924781 +trainer/Policy Loss 4.74647 +trainer/Q1 Predictions Mean -74.1849 +trainer/Q1 Predictions Std 17.277 +trainer/Q1 Predictions Max -4.86465 +trainer/Q1 Predictions Min -88.3144 +trainer/Q2 Predictions Mean -74.184 +trainer/Q2 Predictions Std 17.3408 +trainer/Q2 Predictions Max -5.23927 +trainer/Q2 Predictions Min -88.3419 +trainer/Q Targets Mean -73.8688 +trainer/Q Targets Std 17.521 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0627 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0208265 +trainer/policy/mean Std 0.725884 +trainer/policy/mean Max 0.999421 +trainer/policy/mean Min -0.998979 +trainer/policy/std Mean 0.413442 +trainer/policy/std Std 0.0206673 +trainer/policy/std Max 0.436162 +trainer/policy/std Min 0.376879 +trainer/Advantage Weights Mean 1.13477 +trainer/Advantage Weights Std 8.91769 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.21206e-15 +trainer/Advantage Score Mean -0.523302 +trainer/Advantage Score Std 0.556288 +trainer/Advantage Score Max 0.774132 +trainer/Advantage Score Min -3.28878 +trainer/V1 Predictions Mean -73.6283 +trainer/V1 Predictions Std 17.5585 +trainer/V1 Predictions Max -2.24745 +trainer/V1 Predictions Min -87.899 +trainer/VF Loss 0.0621324 +expl/num steps total 653000 +expl/num paths total 854 +expl/path length Mean 500 +expl/path length Std 183 +expl/path length Max 683 +expl/path length Min 317 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0131319 +expl/Actions Std 0.824954 +expl/Actions Max 2.32757 +expl/Actions Min -2.32402 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 602924 +eval/num paths total 657 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0112153 +eval/Actions Std 0.755578 +eval/Actions Max 0.999823 +eval/Actions Min -0.999611 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.25128e-05 +time/evaluation sampling (s) 4.52493 +time/exploration sampling (s) 6.88448 +time/logging (s) 0.00906296 +time/saving (s) 0.0165389 +time/training (s) 18.814 +time/epoch (s) 30.249 +time/total (s) 16329.6 +Epoch -348 +------------------------------ ---------------- +2022-05-15 22:35:08.268607 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -347 finished +------------------------------ ---------------- +epoch -347 +replay_buffer/size 999047 +trainer/num train calls 654000 +trainer/QF1 Loss 0.973696 +trainer/QF2 Loss 1.06008 +trainer/Policy Loss 9.59952 +trainer/Q1 Predictions Mean -72.7467 +trainer/Q1 Predictions Std 19.6439 +trainer/Q1 Predictions Max -1.71031 +trainer/Q1 Predictions Min -87.6238 +trainer/Q2 Predictions Mean -72.8062 +trainer/Q2 Predictions Std 19.7111 +trainer/Q2 Predictions Max -1.25579 +trainer/Q2 Predictions Min -87.8216 +trainer/Q Targets Mean -72.3908 +trainer/Q Targets Std 19.8734 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4355 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0139253 +trainer/policy/mean Std 0.724627 +trainer/policy/mean Max 0.998123 +trainer/policy/mean Min -0.999736 +trainer/policy/std Mean 0.414922 +trainer/policy/std Std 0.0195839 +trainer/policy/std Max 0.436023 +trainer/policy/std Min 0.38223 +trainer/Advantage Weights Mean 1.88956 +trainer/Advantage Weights Std 12.2702 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.03133e-27 +trainer/Advantage Score Mean -0.614676 +trainer/Advantage Score Std 0.711536 +trainer/Advantage Score Max 1.22646 +trainer/Advantage Score Min -6.02194 +trainer/V1 Predictions Mean -72.0904 +trainer/V1 Predictions Std 19.9645 +trainer/V1 Predictions Max -0.977626 +trainer/V1 Predictions Min -87.2428 +trainer/VF Loss 0.0983892 +expl/num steps total 654000 +expl/num paths total 856 +expl/path length Mean 500 +expl/path length Std 80 +expl/path length Max 580 +expl/path length Min 420 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00851508 +expl/Actions Std 0.841151 +expl/Actions Max 2.29621 +expl/Actions Min -2.24751 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 603452 +eval/num paths total 658 +eval/path length Mean 528 +eval/path length Std 0 +eval/path length Max 528 +eval/path length Min 528 +eval/Rewards Mean 0.00189394 +eval/Rewards Std 0.0434782 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0195655 +eval/Actions Std 0.743707 +eval/Actions Max 0.999778 +eval/Actions Min -0.999303 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.953e-06 +time/evaluation sampling (s) 4.67791 +time/exploration sampling (s) 6.73434 +time/logging (s) 0.0105283 +time/saving (s) 0.0183099 +time/training (s) 18.2815 +time/epoch (s) 29.7226 +time/total (s) 16359.4 +Epoch -347 +------------------------------ ---------------- +2022-05-15 22:35:39.541950 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -346 finished +------------------------------ ---------------- +epoch -346 +replay_buffer/size 999047 +trainer/num train calls 655000 +trainer/QF1 Loss 1.10397 +trainer/QF2 Loss 0.994993 +trainer/Policy Loss 13.8648 +trainer/Q1 Predictions Mean -73.0721 +trainer/Q1 Predictions Std 20.2624 +trainer/Q1 Predictions Max -0.317448 +trainer/Q1 Predictions Min -87.7703 +trainer/Q2 Predictions Mean -73.0709 +trainer/Q2 Predictions Std 20.198 +trainer/Q2 Predictions Max 0.0321601 +trainer/Q2 Predictions Min -87.6386 +trainer/Q Targets Mean -72.9044 +trainer/Q Targets Std 20.2986 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2621 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00667143 +trainer/policy/mean Std 0.719052 +trainer/policy/mean Max 0.99789 +trainer/policy/mean Min -0.999361 +trainer/policy/std Mean 0.412855 +trainer/policy/std Std 0.0201031 +trainer/policy/std Max 0.436772 +trainer/policy/std Min 0.377382 +trainer/Advantage Weights Mean 2.93269 +trainer/Advantage Weights Std 13.6466 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.41031e-24 +trainer/Advantage Score Mean -0.51967 +trainer/Advantage Score Std 0.797812 +trainer/Advantage Score Max 0.863331 +trainer/Advantage Score Min -5.32592 +trainer/V1 Predictions Mean -72.6684 +trainer/V1 Predictions Std 20.3977 +trainer/V1 Predictions Max 0.336452 +trainer/V1 Predictions Min -87.1723 +trainer/VF Loss 0.0997666 +expl/num steps total 655000 +expl/num paths total 857 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0583297 +expl/Actions Std 0.807832 +expl/Actions Max 2.31566 +expl/Actions Min -2.33606 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 604389 +eval/num paths total 659 +eval/path length Mean 937 +eval/path length Std 0 +eval/path length Max 937 +eval/path length Min 937 +eval/Rewards Mean 0.00106724 +eval/Rewards Std 0.0326511 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0525871 +eval/Actions Std 0.753984 +eval/Actions Max 0.999922 +eval/Actions Min -0.999079 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.06059e-05 +time/evaluation sampling (s) 5.04114 +time/exploration sampling (s) 6.92048 +time/logging (s) 0.0111142 +time/saving (s) 0.0173392 +time/training (s) 19.265 +time/epoch (s) 31.255 +time/total (s) 16390.6 +Epoch -346 +------------------------------ ---------------- +2022-05-15 22:36:09.896813 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -345 finished +------------------------------ ---------------- +epoch -345 +replay_buffer/size 999047 +trainer/num train calls 656000 +trainer/QF1 Loss 1.24947 +trainer/QF2 Loss 1.01495 +trainer/Policy Loss 60.0894 +trainer/Q1 Predictions Mean -73.9384 +trainer/Q1 Predictions Std 17.3971 +trainer/Q1 Predictions Max -8.10983 +trainer/Q1 Predictions Min -87.3555 +trainer/Q2 Predictions Mean -74.0806 +trainer/Q2 Predictions Std 17.2592 +trainer/Q2 Predictions Max -7.58829 +trainer/Q2 Predictions Min -87.7554 +trainer/Q Targets Mean -74.4705 +trainer/Q Targets Std 17.4705 +trainer/Q Targets Max -6.96294 +trainer/Q Targets Min -87.9447 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0154377 +trainer/policy/mean Std 0.730988 +trainer/policy/mean Max 0.998634 +trainer/policy/mean Min -0.99917 +trainer/policy/std Mean 0.413058 +trainer/policy/std Std 0.02043 +trainer/policy/std Max 0.436411 +trainer/policy/std Min 0.37993 +trainer/Advantage Weights Mean 13.0604 +trainer/Advantage Weights Std 27.3188 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.02737e-14 +trainer/Advantage Score Mean -0.134373 +trainer/Advantage Score Std 0.660853 +trainer/Advantage Score Max 0.866959 +trainer/Advantage Score Min -3.08431 +trainer/V1 Predictions Mean -74.1331 +trainer/V1 Predictions Std 17.6544 +trainer/V1 Predictions Max -8.60637 +trainer/V1 Predictions Min -87.8573 +trainer/VF Loss 0.0898719 +expl/num steps total 656000 +expl/num paths total 858 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0421786 +expl/Actions Std 0.830916 +expl/Actions Max 2.40916 +expl/Actions Min -2.25525 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 604880 +eval/num paths total 660 +eval/path length Mean 491 +eval/path length Std 0 +eval/path length Max 491 +eval/path length Min 491 +eval/Rewards Mean 0.00203666 +eval/Rewards Std 0.0450834 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0156756 +eval/Actions Std 0.741506 +eval/Actions Max 0.999504 +eval/Actions Min -0.999163 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.013e-05 +time/evaluation sampling (s) 4.42908 +time/exploration sampling (s) 6.90928 +time/logging (s) 0.0105935 +time/saving (s) 0.0171693 +time/training (s) 18.9698 +time/epoch (s) 30.3359 +time/total (s) 16421 +Epoch -345 +------------------------------ ---------------- +2022-05-15 22:36:40.829671 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -344 finished +------------------------------ ---------------- +epoch -344 +replay_buffer/size 999047 +trainer/num train calls 657000 +trainer/QF1 Loss 1.15007 +trainer/QF2 Loss 1.09344 +trainer/Policy Loss 38.5239 +trainer/Q1 Predictions Mean -73.1811 +trainer/Q1 Predictions Std 18.9155 +trainer/Q1 Predictions Max -1.73428 +trainer/Q1 Predictions Min -87.069 +trainer/Q2 Predictions Mean -73.1763 +trainer/Q2 Predictions Std 18.8916 +trainer/Q2 Predictions Max -1.91423 +trainer/Q2 Predictions Min -87.2946 +trainer/Q Targets Mean -73.3445 +trainer/Q Targets Std 19.4241 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9116 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00468488 +trainer/policy/mean Std 0.727221 +trainer/policy/mean Max 0.999895 +trainer/policy/mean Min -0.997712 +trainer/policy/std Mean 0.410937 +trainer/policy/std Std 0.0207869 +trainer/policy/std Max 0.430415 +trainer/policy/std Min 0.376603 +trainer/Advantage Weights Mean 8.84811 +trainer/Advantage Weights Std 22.0894 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.33678e-25 +trainer/Advantage Score Mean -0.295573 +trainer/Advantage Score Std 0.813408 +trainer/Advantage Score Max 1.02763 +trainer/Advantage Score Min -5.55717 +trainer/V1 Predictions Mean -73.0262 +trainer/V1 Predictions Std 19.4755 +trainer/V1 Predictions Max -1.16305 +trainer/V1 Predictions Min -87.6594 +trainer/VF Loss 0.105508 +expl/num steps total 657000 +expl/num paths total 859 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0207748 +expl/Actions Std 0.825877 +expl/Actions Max 2.35735 +expl/Actions Min -2.68798 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 605269 +eval/num paths total 661 +eval/path length Mean 389 +eval/path length Std 0 +eval/path length Max 389 +eval/path length Min 389 +eval/Rewards Mean 0.00257069 +eval/Rewards Std 0.0506368 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0336962 +eval/Actions Std 0.738471 +eval/Actions Max 0.99978 +eval/Actions Min -0.999081 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.50224e-06 +time/evaluation sampling (s) 5.0477 +time/exploration sampling (s) 6.76162 +time/logging (s) 0.00795667 +time/saving (s) 0.0165323 +time/training (s) 19.0786 +time/epoch (s) 30.9124 +time/total (s) 16451.9 +Epoch -344 +------------------------------ ---------------- +2022-05-15 22:37:10.986232 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -343 finished +------------------------------ ---------------- +epoch -343 +replay_buffer/size 999047 +trainer/num train calls 658000 +trainer/QF1 Loss 0.703864 +trainer/QF2 Loss 0.664998 +trainer/Policy Loss 44.872 +trainer/Q1 Predictions Mean -74.5584 +trainer/Q1 Predictions Std 18.1295 +trainer/Q1 Predictions Max -0.493782 +trainer/Q1 Predictions Min -88.4693 +trainer/Q2 Predictions Mean -74.578 +trainer/Q2 Predictions Std 18.159 +trainer/Q2 Predictions Max -0.243464 +trainer/Q2 Predictions Min -88.2422 +trainer/Q Targets Mean -74.6512 +trainer/Q Targets Std 18.4668 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.5952 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00806072 +trainer/policy/mean Std 0.723588 +trainer/policy/mean Max 0.998109 +trainer/policy/mean Min -0.999056 +trainer/policy/std Mean 0.413569 +trainer/policy/std Std 0.0205325 +trainer/policy/std Max 0.433627 +trainer/policy/std Min 0.381836 +trainer/Advantage Weights Mean 8.63519 +trainer/Advantage Weights Std 24.5152 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.02504e-18 +trainer/Advantage Score Mean -0.271489 +trainer/Advantage Score Std 0.673465 +trainer/Advantage Score Max 0.858627 +trainer/Advantage Score Min -3.94971 +trainer/V1 Predictions Mean -74.388 +trainer/V1 Predictions Std 18.5879 +trainer/V1 Predictions Max 0.78156 +trainer/V1 Predictions Min -88.4429 +trainer/VF Loss 0.078419 +expl/num steps total 658000 +expl/num paths total 860 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.129193 +expl/Actions Std 0.839659 +expl/Actions Max 2.38686 +expl/Actions Min -2.14128 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 606269 +eval/num paths total 662 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.313265 +eval/Actions Std 0.697302 +eval/Actions Max 0.999944 +eval/Actions Min -0.999109 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.11388e-06 +time/evaluation sampling (s) 4.71872 +time/exploration sampling (s) 6.60983 +time/logging (s) 0.0124778 +time/saving (s) 0.0181768 +time/training (s) 18.785 +time/epoch (s) 30.1442 +time/total (s) 16482 +Epoch -343 +------------------------------ ---------------- +2022-05-15 22:37:41.212055 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -342 finished +------------------------------ ---------------- +epoch -342 +replay_buffer/size 999047 +trainer/num train calls 659000 +trainer/QF1 Loss 4.80079 +trainer/QF2 Loss 4.80696 +trainer/Policy Loss 9.86787 +trainer/Q1 Predictions Mean -73.2976 +trainer/Q1 Predictions Std 18.4817 +trainer/Q1 Predictions Max -2.56082 +trainer/Q1 Predictions Min -87.1422 +trainer/Q2 Predictions Mean -73.2842 +trainer/Q2 Predictions Std 18.455 +trainer/Q2 Predictions Max -1.88891 +trainer/Q2 Predictions Min -87.1229 +trainer/Q Targets Mean -73.2753 +trainer/Q Targets Std 18.76 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1576 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0194329 +trainer/policy/mean Std 0.743233 +trainer/policy/mean Max 0.999055 +trainer/policy/mean Min -0.999321 +trainer/policy/std Mean 0.415758 +trainer/policy/std Std 0.0214331 +trainer/policy/std Max 0.436652 +trainer/policy/std Min 0.37937 +trainer/Advantage Weights Mean 2.24128 +trainer/Advantage Weights Std 13.0396 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.62542e-13 +trainer/Advantage Score Mean -0.521331 +trainer/Advantage Score Std 0.546877 +trainer/Advantage Score Max 1.43394 +trainer/Advantage Score Min -2.94478 +trainer/V1 Predictions Mean -73.172 +trainer/V1 Predictions Std 18.7916 +trainer/V1 Predictions Max -0.663518 +trainer/V1 Predictions Min -87.0519 +trainer/VF Loss 0.0718069 +expl/num steps total 659000 +expl/num paths total 862 +expl/path length Mean 500 +expl/path length Std 418 +expl/path length Max 918 +expl/path length Min 82 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0296272 +expl/Actions Std 0.837122 +expl/Actions Max 2.41305 +expl/Actions Min -2.22653 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 606829 +eval/num paths total 663 +eval/path length Mean 560 +eval/path length Std 0 +eval/path length Max 560 +eval/path length Min 560 +eval/Rewards Mean 0.00178571 +eval/Rewards Std 0.04222 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0173718 +eval/Actions Std 0.730704 +eval/Actions Max 0.99977 +eval/Actions Min -0.999607 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.29109e-05 +time/evaluation sampling (s) 4.64495 +time/exploration sampling (s) 6.46099 +time/logging (s) 0.0106357 +time/saving (s) 0.018192 +time/training (s) 19.067 +time/epoch (s) 30.2017 +time/total (s) 16512.2 +Epoch -342 +------------------------------ ---------------- +2022-05-15 22:38:12.022402 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -341 finished +------------------------------ ---------------- +epoch -341 +replay_buffer/size 999047 +trainer/num train calls 660000 +trainer/QF1 Loss 0.72254 +trainer/QF2 Loss 0.741799 +trainer/Policy Loss 31.5041 +trainer/Q1 Predictions Mean -73.5558 +trainer/Q1 Predictions Std 19.2832 +trainer/Q1 Predictions Max -0.820485 +trainer/Q1 Predictions Min -87.5765 +trainer/Q2 Predictions Mean -73.5015 +trainer/Q2 Predictions Std 19.2859 +trainer/Q2 Predictions Max 0.0947084 +trainer/Q2 Predictions Min -87.4957 +trainer/Q Targets Mean -73.6558 +trainer/Q Targets Std 19.5609 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5788 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00826353 +trainer/policy/mean Std 0.724929 +trainer/policy/mean Max 0.999709 +trainer/policy/mean Min -0.999557 +trainer/policy/std Mean 0.414222 +trainer/policy/std Std 0.0200411 +trainer/policy/std Max 0.436013 +trainer/policy/std Min 0.382484 +trainer/Advantage Weights Mean 6.83982 +trainer/Advantage Weights Std 19.4906 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.9215e-19 +trainer/Advantage Score Mean -0.270588 +trainer/Advantage Score Std 0.701349 +trainer/Advantage Score Max 0.950382 +trainer/Advantage Score Min -4.2677 +trainer/V1 Predictions Mean -73.3371 +trainer/V1 Predictions Std 19.8567 +trainer/V1 Predictions Max 1.4608 +trainer/V1 Predictions Min -87.3921 +trainer/VF Loss 0.0779417 +expl/num steps total 660000 +expl/num paths total 863 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.070432 +expl/Actions Std 0.914411 +expl/Actions Max 2.548 +expl/Actions Min -2.41876 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 607829 +eval/num paths total 664 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0142104 +eval/Actions Std 0.692085 +eval/Actions Max 0.999948 +eval/Actions Min -0.999911 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.75092e-06 +time/evaluation sampling (s) 4.95453 +time/exploration sampling (s) 6.76268 +time/logging (s) 0.0127327 +time/saving (s) 0.0187538 +time/training (s) 19.0497 +time/epoch (s) 30.7984 +time/total (s) 16543.1 +Epoch -341 +------------------------------ ---------------- +2022-05-15 22:38:42.657205 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -340 finished +------------------------------ ---------------- +epoch -340 +replay_buffer/size 999047 +trainer/num train calls 661000 +trainer/QF1 Loss 0.61919 +trainer/QF2 Loss 0.513877 +trainer/Policy Loss 22.8619 +trainer/Q1 Predictions Mean -73.9608 +trainer/Q1 Predictions Std 18.0483 +trainer/Q1 Predictions Max -0.901047 +trainer/Q1 Predictions Min -88.129 +trainer/Q2 Predictions Mean -74.0256 +trainer/Q2 Predictions Std 18.0101 +trainer/Q2 Predictions Max -1.17608 +trainer/Q2 Predictions Min -88.0695 +trainer/Q Targets Mean -74.0597 +trainer/Q Targets Std 17.9317 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1269 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00988824 +trainer/policy/mean Std 0.723784 +trainer/policy/mean Max 0.999772 +trainer/policy/mean Min -0.999034 +trainer/policy/std Mean 0.414296 +trainer/policy/std Std 0.0208844 +trainer/policy/std Max 0.438994 +trainer/policy/std Min 0.381093 +trainer/Advantage Weights Mean 5.76151 +trainer/Advantage Weights Std 20.9795 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.30377e-12 +trainer/Advantage Score Mean -0.311894 +trainer/Advantage Score Std 0.572915 +trainer/Advantage Score Max 1.6342 +trainer/Advantage Score Min -2.73658 +trainer/V1 Predictions Mean -73.8018 +trainer/V1 Predictions Std 18.0424 +trainer/V1 Predictions Max 0.189106 +trainer/V1 Predictions Min -88.1503 +trainer/VF Loss 0.0728995 +expl/num steps total 661000 +expl/num paths total 865 +expl/path length Mean 500 +expl/path length Std 119 +expl/path length Max 619 +expl/path length Min 381 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0329251 +expl/Actions Std 0.828324 +expl/Actions Max 2.40352 +expl/Actions Min -2.404 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 608621 +eval/num paths total 665 +eval/path length Mean 792 +eval/path length Std 0 +eval/path length Max 792 +eval/path length Min 792 +eval/Rewards Mean 0.00126263 +eval/Rewards Std 0.035511 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0398375 +eval/Actions Std 0.734551 +eval/Actions Max 0.99995 +eval/Actions Min -0.999653 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.31531e-05 +time/evaluation sampling (s) 4.74041 +time/exploration sampling (s) 6.31867 +time/logging (s) 0.00846524 +time/saving (s) 0.0118482 +time/training (s) 19.5327 +time/epoch (s) 30.6121 +time/total (s) 16573.7 +Epoch -340 +------------------------------ ---------------- +2022-05-15 22:39:12.960222 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -339 finished +------------------------------ ---------------- +epoch -339 +replay_buffer/size 999047 +trainer/num train calls 662000 +trainer/QF1 Loss 0.6156 +trainer/QF2 Loss 0.622107 +trainer/Policy Loss 18.0434 +trainer/Q1 Predictions Mean -73.8738 +trainer/Q1 Predictions Std 18.7684 +trainer/Q1 Predictions Max -1.03596 +trainer/Q1 Predictions Min -88.8994 +trainer/Q2 Predictions Mean -73.7988 +trainer/Q2 Predictions Std 18.7685 +trainer/Q2 Predictions Max -0.384181 +trainer/Q2 Predictions Min -88.2896 +trainer/Q Targets Mean -73.6945 +trainer/Q Targets Std 18.6505 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.3598 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0203621 +trainer/policy/mean Std 0.73238 +trainer/policy/mean Max 0.999073 +trainer/policy/mean Min -0.997889 +trainer/policy/std Mean 0.414111 +trainer/policy/std Std 0.0200197 +trainer/policy/std Max 0.435416 +trainer/policy/std Min 0.384727 +trainer/Advantage Weights Mean 4.10607 +trainer/Advantage Weights Std 18.2953 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.77535e-13 +trainer/Advantage Score Mean -0.45737 +trainer/Advantage Score Std 0.540986 +trainer/Advantage Score Max 2.67024 +trainer/Advantage Score Min -2.818 +trainer/V1 Predictions Mean -73.457 +trainer/V1 Predictions Std 18.6694 +trainer/V1 Predictions Max -0.759733 +trainer/V1 Predictions Min -88.2564 +trainer/VF Loss 0.0918899 +expl/num steps total 662000 +expl/num paths total 866 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0278732 +expl/Actions Std 0.844085 +expl/Actions Max 2.27857 +expl/Actions Min -2.1677 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 609621 +eval/num paths total 666 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.145473 +eval/Actions Std 0.666297 +eval/Actions Max 0.999896 +eval/Actions Min -0.999852 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 6.38608e-06 +time/evaluation sampling (s) 4.37405 +time/exploration sampling (s) 6.46458 +time/logging (s) 0.0146355 +time/saving (s) 0.0198192 +time/training (s) 19.4236 +time/epoch (s) 30.2967 +time/total (s) 16604 +Epoch -339 +------------------------------ ---------------- +2022-05-15 22:39:45.026751 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -338 finished +------------------------------ ---------------- +epoch -338 +replay_buffer/size 999047 +trainer/num train calls 663000 +trainer/QF1 Loss 1.57746 +trainer/QF2 Loss 1.56524 +trainer/Policy Loss 11.0789 +trainer/Q1 Predictions Mean -73.5638 +trainer/Q1 Predictions Std 18.6987 +trainer/Q1 Predictions Max -0.999016 +trainer/Q1 Predictions Min -88.1744 +trainer/Q2 Predictions Mean -73.5067 +trainer/Q2 Predictions Std 18.6868 +trainer/Q2 Predictions Max -1.56197 +trainer/Q2 Predictions Min -88.1119 +trainer/Q Targets Mean -73.1648 +trainer/Q Targets Std 18.9765 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8282 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0116284 +trainer/policy/mean Std 0.733272 +trainer/policy/mean Max 0.999717 +trainer/policy/mean Min -0.996887 +trainer/policy/std Mean 0.415362 +trainer/policy/std Std 0.0198769 +trainer/policy/std Max 0.440298 +trainer/policy/std Min 0.385715 +trainer/Advantage Weights Mean 2.86727 +trainer/Advantage Weights Std 12.2823 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.27366e-15 +trainer/Advantage Score Mean -0.459954 +trainer/Advantage Score Std 0.68045 +trainer/Advantage Score Max 1.04607 +trainer/Advantage Score Min -3.37174 +trainer/V1 Predictions Mean -72.9906 +trainer/V1 Predictions Std 19.1831 +trainer/V1 Predictions Max 0.106179 +trainer/V1 Predictions Min -87.7373 +trainer/VF Loss 0.0772168 +expl/num steps total 663000 +expl/num paths total 868 +expl/path length Mean 500 +expl/path length Std 203 +expl/path length Max 703 +expl/path length Min 297 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00334668 +expl/Actions Std 0.831102 +expl/Actions Max 2.48689 +expl/Actions Min -2.40862 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 610621 +eval/num paths total 667 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0295361 +eval/Actions Std 0.781101 +eval/Actions Max 0.999963 +eval/Actions Min -0.999628 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.12453e-05 +time/evaluation sampling (s) 4.86962 +time/exploration sampling (s) 7.51616 +time/logging (s) 0.00893914 +time/saving (s) 0.0157745 +time/training (s) 19.6299 +time/epoch (s) 32.0404 +time/total (s) 16636 +Epoch -338 +------------------------------ ---------------- +2022-05-15 22:40:16.065729 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -337 finished +------------------------------ ---------------- +epoch -337 +replay_buffer/size 999047 +trainer/num train calls 664000 +trainer/QF1 Loss 0.680938 +trainer/QF2 Loss 0.642604 +trainer/Policy Loss 19.1108 +trainer/Q1 Predictions Mean -74.0668 +trainer/Q1 Predictions Std 18.6552 +trainer/Q1 Predictions Max -0.927251 +trainer/Q1 Predictions Min -88.4389 +trainer/Q2 Predictions Mean -74.0963 +trainer/Q2 Predictions Std 18.5124 +trainer/Q2 Predictions Max -0.624699 +trainer/Q2 Predictions Min -89.1216 +trainer/Q Targets Mean -73.9845 +trainer/Q Targets Std 18.366 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5839 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00330103 +trainer/policy/mean Std 0.727886 +trainer/policy/mean Max 0.999401 +trainer/policy/mean Min -0.998953 +trainer/policy/std Mean 0.412792 +trainer/policy/std Std 0.0194495 +trainer/policy/std Max 0.438948 +trainer/policy/std Min 0.383305 +trainer/Advantage Weights Mean 5.14525 +trainer/Advantage Weights Std 20.3645 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.85553e-16 +trainer/Advantage Score Mean -0.483901 +trainer/Advantage Score Std 0.610481 +trainer/Advantage Score Max 2.44613 +trainer/Advantage Score Min -3.49163 +trainer/V1 Predictions Mean -73.7395 +trainer/V1 Predictions Std 18.4297 +trainer/V1 Predictions Max -1.7947 +trainer/V1 Predictions Min -87.2503 +trainer/VF Loss 0.106312 +expl/num steps total 664000 +expl/num paths total 869 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.057332 +expl/Actions Std 0.829488 +expl/Actions Max 2.38704 +expl/Actions Min -2.2563 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 611621 +eval/num paths total 668 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.249792 +eval/Actions Std 0.708227 +eval/Actions Max 0.997758 +eval/Actions Min -0.999609 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0822e-05 +time/evaluation sampling (s) 4.70009 +time/exploration sampling (s) 6.90518 +time/logging (s) 0.0124194 +time/saving (s) 0.0205684 +time/training (s) 19.3895 +time/epoch (s) 31.0278 +time/total (s) 16667.1 +Epoch -337 +------------------------------ ---------------- +2022-05-15 22:40:47.130828 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -336 finished +------------------------------ ---------------- +epoch -336 +replay_buffer/size 999047 +trainer/num train calls 665000 +trainer/QF1 Loss 0.682624 +trainer/QF2 Loss 0.525376 +trainer/Policy Loss 11.535 +trainer/Q1 Predictions Mean -73.9815 +trainer/Q1 Predictions Std 17.2858 +trainer/Q1 Predictions Max -0.656772 +trainer/Q1 Predictions Min -87.3337 +trainer/Q2 Predictions Mean -73.9391 +trainer/Q2 Predictions Std 17.2709 +trainer/Q2 Predictions Max -1.60069 +trainer/Q2 Predictions Min -87.4326 +trainer/Q Targets Mean -73.8712 +trainer/Q Targets Std 17.4764 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9936 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0203457 +trainer/policy/mean Std 0.730267 +trainer/policy/mean Max 0.999528 +trainer/policy/mean Min -0.999031 +trainer/policy/std Mean 0.412664 +trainer/policy/std Std 0.0198357 +trainer/policy/std Max 0.436103 +trainer/policy/std Min 0.381725 +trainer/Advantage Weights Mean 3.13055 +trainer/Advantage Weights Std 13.9474 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.02947e-15 +trainer/Advantage Score Mean -0.464524 +trainer/Advantage Score Std 0.5638 +trainer/Advantage Score Max 0.881592 +trainer/Advantage Score Min -3.45097 +trainer/V1 Predictions Mean -73.6253 +trainer/V1 Predictions Std 17.5826 +trainer/V1 Predictions Max -0.435963 +trainer/V1 Predictions Min -86.8759 +trainer/VF Loss 0.0632598 +expl/num steps total 665000 +expl/num paths total 871 +expl/path length Mean 500 +expl/path length Std 10 +expl/path length Max 510 +expl/path length Min 490 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.0708857 +expl/Actions Std 0.848874 +expl/Actions Max 2.30118 +expl/Actions Min -2.43419 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 612621 +eval/num paths total 669 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.435363 +eval/Actions Std 0.54185 +eval/Actions Max 0.999833 +eval/Actions Min -0.99959 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.2558e-05 +time/evaluation sampling (s) 4.74229 +time/exploration sampling (s) 6.67563 +time/logging (s) 0.00832762 +time/saving (s) 0.0129874 +time/training (s) 19.6075 +time/epoch (s) 31.0468 +time/total (s) 16698.1 +Epoch -336 +------------------------------ ---------------- +2022-05-15 22:41:18.082003 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -335 finished +------------------------------ ---------------- +epoch -335 +replay_buffer/size 999047 +trainer/num train calls 666000 +trainer/QF1 Loss 0.761111 +trainer/QF2 Loss 0.698024 +trainer/Policy Loss 12.8856 +trainer/Q1 Predictions Mean -72.0098 +trainer/Q1 Predictions Std 19.7734 +trainer/Q1 Predictions Max -0.476424 +trainer/Q1 Predictions Min -87.4259 +trainer/Q2 Predictions Mean -72.0452 +trainer/Q2 Predictions Std 19.7937 +trainer/Q2 Predictions Max -0.421449 +trainer/Q2 Predictions Min -87.628 +trainer/Q Targets Mean -72.0195 +trainer/Q Targets Std 19.4044 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5073 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00430901 +trainer/policy/mean Std 0.734588 +trainer/policy/mean Max 0.999262 +trainer/policy/mean Min -0.998387 +trainer/policy/std Mean 0.413342 +trainer/policy/std Std 0.0194696 +trainer/policy/std Max 0.433158 +trainer/policy/std Min 0.383299 +trainer/Advantage Weights Mean 4.78405 +trainer/Advantage Weights Std 18.4869 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.09131e-10 +trainer/Advantage Score Mean -0.336612 +trainer/Advantage Score Std 0.510146 +trainer/Advantage Score Max 1.49895 +trainer/Advantage Score Min -2.13983 +trainer/V1 Predictions Mean -71.6987 +trainer/V1 Predictions Std 19.6934 +trainer/V1 Predictions Max 0.393168 +trainer/V1 Predictions Min -87.1963 +trainer/VF Loss 0.0712573 +expl/num steps total 666000 +expl/num paths total 873 +expl/path length Mean 500 +expl/path length Std 331 +expl/path length Max 831 +expl/path length Min 169 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0158748 +expl/Actions Std 0.833535 +expl/Actions Max 2.38781 +expl/Actions Min -2.49596 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 613621 +eval/num paths total 670 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.131721 +eval/Actions Std 0.795661 +eval/Actions Max 0.99998 +eval/Actions Min -0.99937 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.65036e-06 +time/evaluation sampling (s) 4.92135 +time/exploration sampling (s) 6.47471 +time/logging (s) 0.0115214 +time/saving (s) 0.0156468 +time/training (s) 19.5191 +time/epoch (s) 30.9423 +time/total (s) 16729.1 +Epoch -335 +------------------------------ ---------------- +2022-05-15 22:41:48.226312 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -334 finished +------------------------------ ---------------- +epoch -334 +replay_buffer/size 999047 +trainer/num train calls 667000 +trainer/QF1 Loss 0.555302 +trainer/QF2 Loss 0.49639 +trainer/Policy Loss 8.97495 +trainer/Q1 Predictions Mean -73.3475 +trainer/Q1 Predictions Std 19.8069 +trainer/Q1 Predictions Max -0.504337 +trainer/Q1 Predictions Min -87.7487 +trainer/Q2 Predictions Mean -73.4453 +trainer/Q2 Predictions Std 19.7533 +trainer/Q2 Predictions Max -0.266512 +trainer/Q2 Predictions Min -87.7304 +trainer/Q Targets Mean -73.3075 +trainer/Q Targets Std 19.821 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7492 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0106511 +trainer/policy/mean Std 0.721116 +trainer/policy/mean Max 0.99876 +trainer/policy/mean Min -0.999059 +trainer/policy/std Mean 0.414163 +trainer/policy/std Std 0.0192238 +trainer/policy/std Max 0.43263 +trainer/policy/std Min 0.384287 +trainer/Advantage Weights Mean 2.70142 +trainer/Advantage Weights Std 14.0685 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.88472e-21 +trainer/Advantage Score Mean -0.532441 +trainer/Advantage Score Std 0.719813 +trainer/Advantage Score Max 0.71687 +trainer/Advantage Score Min -4.77205 +trainer/V1 Predictions Mean -73.0261 +trainer/V1 Predictions Std 19.9618 +trainer/V1 Predictions Max 0.430548 +trainer/V1 Predictions Min -87.5978 +trainer/VF Loss 0.088686 +expl/num steps total 667000 +expl/num paths total 874 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.140544 +expl/Actions Std 0.853418 +expl/Actions Max 2.35456 +expl/Actions Min -2.63143 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 614264 +eval/num paths total 671 +eval/path length Mean 643 +eval/path length Std 0 +eval/path length Max 643 +eval/path length Min 643 +eval/Rewards Mean 0.00155521 +eval/Rewards Std 0.0394055 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0286751 +eval/Actions Std 0.727854 +eval/Actions Max 0.999926 +eval/Actions Min -0.99929 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.28332e-05 +time/evaluation sampling (s) 4.41829 +time/exploration sampling (s) 6.68314 +time/logging (s) 0.010171 +time/saving (s) 0.0150839 +time/training (s) 18.9992 +time/epoch (s) 30.1259 +time/total (s) 16759.2 +Epoch -334 +------------------------------ ---------------- +2022-05-15 22:42:18.915200 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -333 finished +------------------------------ ---------------- +epoch -333 +replay_buffer/size 999047 +trainer/num train calls 668000 +trainer/QF1 Loss 0.695041 +trainer/QF2 Loss 0.739992 +trainer/Policy Loss 25.8395 +trainer/Q1 Predictions Mean -74.4793 +trainer/Q1 Predictions Std 17.1829 +trainer/Q1 Predictions Max -1.43907 +trainer/Q1 Predictions Min -86.9466 +trainer/Q2 Predictions Mean -74.3889 +trainer/Q2 Predictions Std 17.1572 +trainer/Q2 Predictions Max -2.01837 +trainer/Q2 Predictions Min -86.833 +trainer/Q Targets Mean -74.7248 +trainer/Q Targets Std 17.1029 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.016 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00400986 +trainer/policy/mean Std 0.710526 +trainer/policy/mean Max 0.998727 +trainer/policy/mean Min -0.997922 +trainer/policy/std Mean 0.412955 +trainer/policy/std Std 0.0189206 +trainer/policy/std Max 0.433511 +trainer/policy/std Min 0.383622 +trainer/Advantage Weights Mean 6.30916 +trainer/Advantage Weights Std 19.3185 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.26302e-13 +trainer/Advantage Score Mean -0.222003 +trainer/Advantage Score Std 0.500057 +trainer/Advantage Score Max 1.37405 +trainer/Advantage Score Min -2.8751 +trainer/V1 Predictions Mean -74.4128 +trainer/V1 Predictions Std 17.2424 +trainer/V1 Predictions Max -1.38616 +trainer/V1 Predictions Min -87.0049 +trainer/VF Loss 0.054359 +expl/num steps total 668000 +expl/num paths total 876 +expl/path length Mean 500 +expl/path length Std 352 +expl/path length Max 852 +expl/path length Min 148 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0165297 +expl/Actions Std 0.826576 +expl/Actions Max 2.47525 +expl/Actions Min -2.52897 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 614892 +eval/num paths total 672 +eval/path length Mean 628 +eval/path length Std 0 +eval/path length Max 628 +eval/path length Min 628 +eval/Rewards Mean 0.00159236 +eval/Rewards Std 0.0398726 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0210215 +eval/Actions Std 0.704723 +eval/Actions Max 0.999558 +eval/Actions Min -0.998869 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.28299e-05 +time/evaluation sampling (s) 4.68398 +time/exploration sampling (s) 6.95732 +time/logging (s) 0.013967 +time/saving (s) 0.016162 +time/training (s) 18.9951 +time/epoch (s) 30.6666 +time/total (s) 16789.9 +Epoch -333 +------------------------------ ---------------- +2022-05-15 22:42:49.075149 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -332 finished +------------------------------ ---------------- +epoch -332 +replay_buffer/size 999047 +trainer/num train calls 669000 +trainer/QF1 Loss 0.892243 +trainer/QF2 Loss 0.877887 +trainer/Policy Loss 15.0837 +trainer/Q1 Predictions Mean -74.851 +trainer/Q1 Predictions Std 17.7506 +trainer/Q1 Predictions Max -2.68806 +trainer/Q1 Predictions Min -88.2643 +trainer/Q2 Predictions Mean -74.7643 +trainer/Q2 Predictions Std 17.8472 +trainer/Q2 Predictions Max -2.39053 +trainer/Q2 Predictions Min -88.5273 +trainer/Q Targets Mean -74.7106 +trainer/Q Targets Std 17.9464 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2501 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0181513 +trainer/policy/mean Std 0.723276 +trainer/policy/mean Max 0.999054 +trainer/policy/mean Min -0.999362 +trainer/policy/std Mean 0.414136 +trainer/policy/std Std 0.0186285 +trainer/policy/std Max 0.433843 +trainer/policy/std Min 0.38429 +trainer/Advantage Weights Mean 3.55845 +trainer/Advantage Weights Std 15.3327 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.8397e-16 +trainer/Advantage Score Mean -0.319411 +trainer/Advantage Score Std 0.539661 +trainer/Advantage Score Max 0.789932 +trainer/Advantage Score Min -3.5496 +trainer/V1 Predictions Mean -74.516 +trainer/V1 Predictions Std 17.9521 +trainer/V1 Predictions Max -0.796392 +trainer/V1 Predictions Min -88.155 +trainer/VF Loss 0.0496838 +expl/num steps total 669000 +expl/num paths total 878 +expl/path length Mean 500 +expl/path length Std 426 +expl/path length Max 926 +expl/path length Min 74 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.032713 +expl/Actions Std 0.810575 +expl/Actions Max 2.43753 +expl/Actions Min -2.28369 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 615892 +eval/num paths total 673 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.229824 +eval/Actions Std 0.702592 +eval/Actions Max 0.999331 +eval/Actions Min -0.99938 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.32718e-05 +time/evaluation sampling (s) 4.60292 +time/exploration sampling (s) 6.40674 +time/logging (s) 0.0124083 +time/saving (s) 0.0150117 +time/training (s) 19.1034 +time/epoch (s) 30.1405 +time/total (s) 16820 +Epoch -332 +------------------------------ ---------------- +2022-05-15 22:43:20.188730 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -331 finished +------------------------------ ---------------- +epoch -331 +replay_buffer/size 999047 +trainer/num train calls 670000 +trainer/QF1 Loss 0.793014 +trainer/QF2 Loss 0.779905 +trainer/Policy Loss 14.8842 +trainer/Q1 Predictions Mean -73.7168 +trainer/Q1 Predictions Std 18.1534 +trainer/Q1 Predictions Max -0.555904 +trainer/Q1 Predictions Min -87.9455 +trainer/Q2 Predictions Mean -73.7511 +trainer/Q2 Predictions Std 18.0963 +trainer/Q2 Predictions Max -1.20523 +trainer/Q2 Predictions Min -87.656 +trainer/Q Targets Mean -73.8534 +trainer/Q Targets Std 18.3235 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4473 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0163772 +trainer/policy/mean Std 0.717911 +trainer/policy/mean Max 0.999083 +trainer/policy/mean Min -0.999416 +trainer/policy/std Mean 0.412373 +trainer/policy/std Std 0.0193967 +trainer/policy/std Max 0.43309 +trainer/policy/std Min 0.380648 +trainer/Advantage Weights Mean 3.55407 +trainer/Advantage Weights Std 16.0312 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.8931e-21 +trainer/Advantage Score Mean -0.434834 +trainer/Advantage Score Std 0.601307 +trainer/Advantage Score Max 0.719807 +trainer/Advantage Score Min -4.65805 +trainer/V1 Predictions Mean -73.577 +trainer/V1 Predictions Std 18.4638 +trainer/V1 Predictions Max -0.381693 +trainer/V1 Predictions Min -87.2848 +trainer/VF Loss 0.0640569 +expl/num steps total 670000 +expl/num paths total 879 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.060339 +expl/Actions Std 0.926127 +expl/Actions Max 2.38304 +expl/Actions Min -2.35664 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 616892 +eval/num paths total 674 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0244831 +eval/Actions Std 0.695203 +eval/Actions Max 0.999433 +eval/Actions Min -0.998647 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.27732e-06 +time/evaluation sampling (s) 5.07938 +time/exploration sampling (s) 6.3404 +time/logging (s) 0.00856478 +time/saving (s) 0.0125352 +time/training (s) 19.6521 +time/epoch (s) 31.093 +time/total (s) 16851.1 +Epoch -331 +------------------------------ ---------------- +2022-05-15 22:43:50.361389 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -330 finished +------------------------------ ---------------- +epoch -330 +replay_buffer/size 999047 +trainer/num train calls 671000 +trainer/QF1 Loss 1.47421 +trainer/QF2 Loss 1.18963 +trainer/Policy Loss 11.061 +trainer/Q1 Predictions Mean -71.267 +trainer/Q1 Predictions Std 20.5056 +trainer/Q1 Predictions Max -0.557277 +trainer/Q1 Predictions Min -87.4531 +trainer/Q2 Predictions Mean -71.3584 +trainer/Q2 Predictions Std 20.4844 +trainer/Q2 Predictions Max -1.09731 +trainer/Q2 Predictions Min -87.5243 +trainer/Q Targets Mean -71.2361 +trainer/Q Targets Std 20.02 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3092 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00579815 +trainer/policy/mean Std 0.725669 +trainer/policy/mean Max 0.997997 +trainer/policy/mean Min -0.999179 +trainer/policy/std Mean 0.412848 +trainer/policy/std Std 0.0191333 +trainer/policy/std Max 0.433833 +trainer/policy/std Min 0.376577 +trainer/Advantage Weights Mean 3.1929 +trainer/Advantage Weights Std 15.6633 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.25609e-17 +trainer/Advantage Score Mean -0.46834 +trainer/Advantage Score Std 0.610242 +trainer/Advantage Score Max 2.02861 +trainer/Advantage Score Min -3.89159 +trainer/V1 Predictions Mean -70.8583 +trainer/V1 Predictions Std 20.3015 +trainer/V1 Predictions Max -0.00212276 +trainer/V1 Predictions Min -87.109 +trainer/VF Loss 0.0835256 +expl/num steps total 671000 +expl/num paths total 881 +expl/path length Mean 500 +expl/path length Std 457 +expl/path length Max 957 +expl/path length Min 43 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0472101 +expl/Actions Std 0.804535 +expl/Actions Max 2.39201 +expl/Actions Min -2.14962 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 617813 +eval/num paths total 675 +eval/path length Mean 921 +eval/path length Std 0 +eval/path length Max 921 +eval/path length Min 921 +eval/Rewards Mean 0.00108578 +eval/Rewards Std 0.0329332 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0275829 +eval/Actions Std 0.719189 +eval/Actions Max 0.999756 +eval/Actions Min -0.999668 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.0259e-05 +time/evaluation sampling (s) 4.15559 +time/exploration sampling (s) 6.84383 +time/logging (s) 0.0118098 +time/saving (s) 0.0180196 +time/training (s) 19.129 +time/epoch (s) 30.1583 +time/total (s) 16881.3 +Epoch -330 +------------------------------ ---------------- +2022-05-15 22:44:20.484716 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -329 finished +------------------------------ ---------------- +epoch -329 +replay_buffer/size 999047 +trainer/num train calls 672000 +trainer/QF1 Loss 1.15062 +trainer/QF2 Loss 1.11833 +trainer/Policy Loss 7.66825 +trainer/Q1 Predictions Mean -73.6784 +trainer/Q1 Predictions Std 18.0787 +trainer/Q1 Predictions Max -0.34224 +trainer/Q1 Predictions Min -87.9076 +trainer/Q2 Predictions Mean -73.6174 +trainer/Q2 Predictions Std 18.0398 +trainer/Q2 Predictions Max -0.685746 +trainer/Q2 Predictions Min -87.9513 +trainer/Q Targets Mean -73.7008 +trainer/Q Targets Std 18.1034 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5873 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0069763 +trainer/policy/mean Std 0.73683 +trainer/policy/mean Max 0.999538 +trainer/policy/mean Min -0.999147 +trainer/policy/std Mean 0.413615 +trainer/policy/std Std 0.0189385 +trainer/policy/std Max 0.434592 +trainer/policy/std Min 0.382579 +trainer/Advantage Weights Mean 2.49993 +trainer/Advantage Weights Std 13.6718 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.68442e-17 +trainer/Advantage Score Mean -0.507333 +trainer/Advantage Score Std 0.60019 +trainer/Advantage Score Max 0.859917 +trainer/Advantage Score Min -3.78398 +trainer/V1 Predictions Mean -73.432 +trainer/V1 Predictions Std 18.2043 +trainer/V1 Predictions Max 0.127486 +trainer/V1 Predictions Min -87.6239 +trainer/VF Loss 0.0701606 +expl/num steps total 672000 +expl/num paths total 882 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0207338 +expl/Actions Std 0.825815 +expl/Actions Max 2.33504 +expl/Actions Min -2.41512 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 618813 +eval/num paths total 676 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.115375 +eval/Actions Std 0.620344 +eval/Actions Max 0.998826 +eval/Actions Min -0.999342 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.28867e-05 +time/evaluation sampling (s) 5.1247 +time/exploration sampling (s) 5.96268 +time/logging (s) 0.0104742 +time/saving (s) 0.0150251 +time/training (s) 18.9957 +time/epoch (s) 30.1086 +time/total (s) 16911.4 +Epoch -329 +------------------------------ ---------------- +2022-05-15 22:44:50.954728 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -328 finished +------------------------------ ---------------- +epoch -328 +replay_buffer/size 999047 +trainer/num train calls 673000 +trainer/QF1 Loss 0.447395 +trainer/QF2 Loss 0.452616 +trainer/Policy Loss 7.27432 +trainer/Q1 Predictions Mean -75.1177 +trainer/Q1 Predictions Std 15.7967 +trainer/Q1 Predictions Max 0.0142147 +trainer/Q1 Predictions Min -87.2963 +trainer/Q2 Predictions Mean -75.2053 +trainer/Q2 Predictions Std 15.8604 +trainer/Q2 Predictions Max -0.271411 +trainer/Q2 Predictions Min -87.383 +trainer/Q Targets Mean -75.3289 +trainer/Q Targets Std 15.8287 +trainer/Q Targets Max -3.19919 +trainer/Q Targets Min -87.4235 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00904426 +trainer/policy/mean Std 0.729757 +trainer/policy/mean Max 0.99909 +trainer/policy/mean Min -0.998691 +trainer/policy/std Mean 0.414432 +trainer/policy/std Std 0.0195413 +trainer/policy/std Max 0.437323 +trainer/policy/std Min 0.382696 +trainer/Advantage Weights Mean 1.73531 +trainer/Advantage Weights Std 9.96707 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.16574e-17 +trainer/Advantage Score Mean -0.443523 +trainer/Advantage Score Std 0.487175 +trainer/Advantage Score Max 0.525837 +trainer/Advantage Score Min -3.89906 +trainer/V1 Predictions Mean -75.049 +trainer/V1 Predictions Std 16.0015 +trainer/V1 Predictions Max 0.396447 +trainer/V1 Predictions Min -87.2967 +trainer/VF Loss 0.0475441 +expl/num steps total 673000 +expl/num paths total 883 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.168146 +expl/Actions Std 0.84031 +expl/Actions Max 2.34515 +expl/Actions Min -2.60068 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 619790 +eval/num paths total 677 +eval/path length Mean 977 +eval/path length Std 0 +eval/path length Max 977 +eval/path length Min 977 +eval/Rewards Mean 0.00102354 +eval/Rewards Std 0.0319765 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00224447 +eval/Actions Std 0.73107 +eval/Actions Max 0.99975 +eval/Actions Min -0.999741 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.26073e-05 +time/evaluation sampling (s) 4.48568 +time/exploration sampling (s) 6.56019 +time/logging (s) 0.0124054 +time/saving (s) 0.0160539 +time/training (s) 19.3787 +time/epoch (s) 30.453 +time/total (s) 16941.9 +Epoch -328 +------------------------------ ---------------- +2022-05-15 22:45:21.386458 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -327 finished +------------------------------ ---------------- +epoch -327 +replay_buffer/size 999047 +trainer/num train calls 674000 +trainer/QF1 Loss 0.58393 +trainer/QF2 Loss 0.507138 +trainer/Policy Loss 11.8866 +trainer/Q1 Predictions Mean -74.4168 +trainer/Q1 Predictions Std 17.3531 +trainer/Q1 Predictions Max -4.03355 +trainer/Q1 Predictions Min -87.2455 +trainer/Q2 Predictions Mean -74.4141 +trainer/Q2 Predictions Std 17.3291 +trainer/Q2 Predictions Max -4.31302 +trainer/Q2 Predictions Min -87.4148 +trainer/Q Targets Mean -74.3201 +trainer/Q Targets Std 17.2508 +trainer/Q Targets Max -3.64591 +trainer/Q Targets Min -87.334 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00467553 +trainer/policy/mean Std 0.722639 +trainer/policy/mean Max 0.998492 +trainer/policy/mean Min -0.997373 +trainer/policy/std Mean 0.413944 +trainer/policy/std Std 0.0193142 +trainer/policy/std Max 0.435345 +trainer/policy/std Min 0.383658 +trainer/Advantage Weights Mean 3.29971 +trainer/Advantage Weights Std 16.4752 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.48206e-20 +trainer/Advantage Score Mean -0.494757 +trainer/Advantage Score Std 0.635953 +trainer/Advantage Score Max 2.12232 +trainer/Advantage Score Min -4.48041 +trainer/V1 Predictions Mean -74.0768 +trainer/V1 Predictions Std 17.412 +trainer/V1 Predictions Max -2.94754 +trainer/V1 Predictions Min -87.2041 +trainer/VF Loss 0.0935309 +expl/num steps total 674000 +expl/num paths total 885 +expl/path length Mean 500 +expl/path length Std 284 +expl/path length Max 784 +expl/path length Min 216 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0160741 +expl/Actions Std 0.816732 +expl/Actions Max 2.35529 +expl/Actions Min -2.1043 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 620452 +eval/num paths total 678 +eval/path length Mean 662 +eval/path length Std 0 +eval/path length Max 662 +eval/path length Min 662 +eval/Rewards Mean 0.00151057 +eval/Rewards Std 0.0388367 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0402068 +eval/Actions Std 0.745196 +eval/Actions Max 0.999816 +eval/Actions Min -0.999719 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.31768e-05 +time/evaluation sampling (s) 4.74586 +time/exploration sampling (s) 6.60276 +time/logging (s) 0.0104187 +time/saving (s) 0.0166233 +time/training (s) 19.0408 +time/epoch (s) 30.4164 +time/total (s) 16972.3 +Epoch -327 +------------------------------ ---------------- +2022-05-15 22:45:51.937255 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -326 finished +------------------------------ ---------------- +epoch -326 +replay_buffer/size 999047 +trainer/num train calls 675000 +trainer/QF1 Loss 0.988546 +trainer/QF2 Loss 0.896341 +trainer/Policy Loss 5.33147 +trainer/Q1 Predictions Mean -73.0175 +trainer/Q1 Predictions Std 19.4046 +trainer/Q1 Predictions Max -2.67052 +trainer/Q1 Predictions Min -87.3594 +trainer/Q2 Predictions Mean -73.073 +trainer/Q2 Predictions Std 19.4494 +trainer/Q2 Predictions Max -1.94372 +trainer/Q2 Predictions Min -87.6579 +trainer/Q Targets Mean -72.7501 +trainer/Q Targets Std 19.4573 +trainer/Q Targets Max -1.93559 +trainer/Q Targets Min -87.3181 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00278517 +trainer/policy/mean Std 0.732047 +trainer/policy/mean Max 0.998823 +trainer/policy/mean Min -0.998704 +trainer/policy/std Mean 0.412905 +trainer/policy/std Std 0.0197421 +trainer/policy/std Max 0.43373 +trainer/policy/std Min 0.381838 +trainer/Advantage Weights Mean 1.93867 +trainer/Advantage Weights Std 12.4821 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.41256e-21 +trainer/Advantage Score Mean -0.562302 +trainer/Advantage Score Std 0.657645 +trainer/Advantage Score Max 1.12324 +trainer/Advantage Score Min -4.80089 +trainer/V1 Predictions Mean -72.5357 +trainer/V1 Predictions Std 19.6563 +trainer/V1 Predictions Max -1.57021 +trainer/V1 Predictions Min -87.3584 +trainer/VF Loss 0.0821666 +expl/num steps total 675000 +expl/num paths total 887 +expl/path length Mean 500 +expl/path length Std 211 +expl/path length Max 711 +expl/path length Min 289 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0140966 +expl/Actions Std 0.842403 +expl/Actions Max 2.26045 +expl/Actions Min -2.55893 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 621452 +eval/num paths total 679 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0144341 +eval/Actions Std 0.761257 +eval/Actions Max 0.999977 +eval/Actions Min -0.999904 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.26711e-05 +time/evaluation sampling (s) 4.7942 +time/exploration sampling (s) 6.1922 +time/logging (s) 0.0113967 +time/saving (s) 0.018021 +time/training (s) 19.516 +time/epoch (s) 30.5318 +time/total (s) 17002.8 +Epoch -326 +------------------------------ ---------------- +2022-05-15 22:46:21.733877 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -325 finished +------------------------------ ---------------- +epoch -325 +replay_buffer/size 999047 +trainer/num train calls 676000 +trainer/QF1 Loss 0.713772 +trainer/QF2 Loss 0.591093 +trainer/Policy Loss 22.7774 +trainer/Q1 Predictions Mean -75.5414 +trainer/Q1 Predictions Std 16.3106 +trainer/Q1 Predictions Max -0.840002 +trainer/Q1 Predictions Min -87.6771 +trainer/Q2 Predictions Mean -75.5219 +trainer/Q2 Predictions Std 16.2852 +trainer/Q2 Predictions Max -1.8001 +trainer/Q2 Predictions Min -87.3062 +trainer/Q Targets Mean -75.347 +trainer/Q Targets Std 16.147 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3739 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0109306 +trainer/policy/mean Std 0.724382 +trainer/policy/mean Max 0.999805 +trainer/policy/mean Min -0.996939 +trainer/policy/std Mean 0.414254 +trainer/policy/std Std 0.0207035 +trainer/policy/std Max 0.437338 +trainer/policy/std Min 0.382378 +trainer/Advantage Weights Mean 4.58092 +trainer/Advantage Weights Std 18.102 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.41959e-24 +trainer/Advantage Score Mean -0.351313 +trainer/Advantage Score Std 0.595068 +trainer/Advantage Score Max 1.51207 +trainer/Advantage Score Min -5.49117 +trainer/V1 Predictions Mean -75.0909 +trainer/V1 Predictions Std 16.1904 +trainer/V1 Predictions Max -1.95147 +trainer/V1 Predictions Min -87.2174 +trainer/VF Loss 0.0758207 +expl/num steps total 676000 +expl/num paths total 889 +expl/path length Mean 500 +expl/path length Std 426 +expl/path length Max 926 +expl/path length Min 74 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0476333 +expl/Actions Std 0.820822 +expl/Actions Max 2.26646 +expl/Actions Min -2.46851 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 622452 +eval/num paths total 680 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.042199 +eval/Actions Std 0.762033 +eval/Actions Max 0.999942 +eval/Actions Min -0.99996 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.3533e-05 +time/evaluation sampling (s) 4.3012 +time/exploration sampling (s) 6.32428 +time/logging (s) 0.0122683 +time/saving (s) 0.0138307 +time/training (s) 19.1236 +time/epoch (s) 29.7752 +time/total (s) 17032.6 +Epoch -325 +------------------------------ ---------------- +2022-05-15 22:46:52.374789 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -324 finished +------------------------------ ---------------- +epoch -324 +replay_buffer/size 999047 +trainer/num train calls 677000 +trainer/QF1 Loss 1.53564 +trainer/QF2 Loss 1.62063 +trainer/Policy Loss 54.0298 +trainer/Q1 Predictions Mean -73.0526 +trainer/Q1 Predictions Std 17.5077 +trainer/Q1 Predictions Max -0.547314 +trainer/Q1 Predictions Min -87.5089 +trainer/Q2 Predictions Mean -73.031 +trainer/Q2 Predictions Std 17.4911 +trainer/Q2 Predictions Max -0.170339 +trainer/Q2 Predictions Min -87.5094 +trainer/Q Targets Mean -73.9309 +trainer/Q Targets Std 17.4877 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.5121 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0342454 +trainer/policy/mean Std 0.722203 +trainer/policy/mean Max 0.998014 +trainer/policy/mean Min -0.999554 +trainer/policy/std Mean 0.414274 +trainer/policy/std Std 0.0206856 +trainer/policy/std Max 0.437806 +trainer/policy/std Min 0.384976 +trainer/Advantage Weights Mean 13.2248 +trainer/Advantage Weights Std 29.0758 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.49201e-14 +trainer/Advantage Score Mean -0.0817705 +trainer/Advantage Score Std 0.530706 +trainer/Advantage Score Max 1.62538 +trainer/Advantage Score Min -3.07339 +trainer/V1 Predictions Mean -73.6532 +trainer/V1 Predictions Std 17.6809 +trainer/V1 Predictions Max 0.161639 +trainer/V1 Predictions Min -88.5742 +trainer/VF Loss 0.0901038 +expl/num steps total 677000 +expl/num paths total 890 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0370315 +expl/Actions Std 0.819827 +expl/Actions Max 2.21845 +expl/Actions Min -2.47175 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 623010 +eval/num paths total 681 +eval/path length Mean 558 +eval/path length Std 0 +eval/path length Max 558 +eval/path length Min 558 +eval/Rewards Mean 0.00179211 +eval/Rewards Std 0.0422954 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0353955 +eval/Actions Std 0.759932 +eval/Actions Max 0.999707 +eval/Actions Min -0.999578 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.22502e-05 +time/evaluation sampling (s) 4.86065 +time/exploration sampling (s) 6.46442 +time/logging (s) 0.0104511 +time/saving (s) 0.0179787 +time/training (s) 19.2694 +time/epoch (s) 30.6229 +time/total (s) 17063.2 +Epoch -324 +------------------------------ ---------------- +2022-05-15 22:47:22.739472 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -323 finished +------------------------------ ---------------- +epoch -323 +replay_buffer/size 999047 +trainer/num train calls 678000 +trainer/QF1 Loss 0.621499 +trainer/QF2 Loss 0.661668 +trainer/Policy Loss 19.0933 +trainer/Q1 Predictions Mean -72.495 +trainer/Q1 Predictions Std 19.9805 +trainer/Q1 Predictions Max -0.582367 +trainer/Q1 Predictions Min -88.5151 +trainer/Q2 Predictions Mean -72.4491 +trainer/Q2 Predictions Std 20.117 +trainer/Q2 Predictions Max -0.380236 +trainer/Q2 Predictions Min -88.435 +trainer/Q Targets Mean -72.1346 +trainer/Q Targets Std 20.1744 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0939 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0176572 +trainer/policy/mean Std 0.721859 +trainer/policy/mean Max 0.999179 +trainer/policy/mean Min -0.998519 +trainer/policy/std Mean 0.412528 +trainer/policy/std Std 0.0210442 +trainer/policy/std Max 0.437065 +trainer/policy/std Min 0.381132 +trainer/Advantage Weights Mean 3.40459 +trainer/Advantage Weights Std 16.1004 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.8194e-21 +trainer/Advantage Score Mean -0.550007 +trainer/Advantage Score Std 0.844435 +trainer/Advantage Score Max 1.27673 +trainer/Advantage Score Min -4.60699 +trainer/V1 Predictions Mean -71.7767 +trainer/V1 Predictions Std 20.4947 +trainer/V1 Predictions Max 0.222908 +trainer/V1 Predictions Min -87.9863 +trainer/VF Loss 0.117527 +expl/num steps total 678000 +expl/num paths total 892 +expl/path length Mean 500 +expl/path length Std 69 +expl/path length Max 569 +expl/path length Min 431 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0264634 +expl/Actions Std 0.845835 +expl/Actions Max 2.31621 +expl/Actions Min -2.39856 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 624010 +eval/num paths total 682 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0251416 +eval/Actions Std 0.792932 +eval/Actions Max 0.999674 +eval/Actions Min -0.999805 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.31014e-05 +time/evaluation sampling (s) 4.53471 +time/exploration sampling (s) 6.34839 +time/logging (s) 0.0126571 +time/saving (s) 0.023125 +time/training (s) 19.4295 +time/epoch (s) 30.3484 +time/total (s) 17093.6 +Epoch -323 +------------------------------ ---------------- +2022-05-15 22:47:54.557990 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -322 finished +------------------------------ ---------------- +epoch -322 +replay_buffer/size 999047 +trainer/num train calls 679000 +trainer/QF1 Loss 0.693881 +trainer/QF2 Loss 0.619929 +trainer/Policy Loss 17.557 +trainer/Q1 Predictions Mean -72.743 +trainer/Q1 Predictions Std 19.3932 +trainer/Q1 Predictions Max -0.231625 +trainer/Q1 Predictions Min -88.2076 +trainer/Q2 Predictions Mean -72.7909 +trainer/Q2 Predictions Std 19.3346 +trainer/Q2 Predictions Max -0.166735 +trainer/Q2 Predictions Min -87.9763 +trainer/Q Targets Mean -73.033 +trainer/Q Targets Std 19.3477 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6734 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00866964 +trainer/policy/mean Std 0.736331 +trainer/policy/mean Max 0.999894 +trainer/policy/mean Min -0.998781 +trainer/policy/std Mean 0.411911 +trainer/policy/std Std 0.0197085 +trainer/policy/std Max 0.433926 +trainer/policy/std Min 0.380619 +trainer/Advantage Weights Mean 5.22191 +trainer/Advantage Weights Std 19.6082 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.76956e-20 +trainer/Advantage Score Mean -0.388614 +trainer/Advantage Score Std 0.678524 +trainer/Advantage Score Max 1.27259 +trainer/Advantage Score Min -4.44895 +trainer/V1 Predictions Mean -72.7089 +trainer/V1 Predictions Std 19.6287 +trainer/V1 Predictions Max -0.974236 +trainer/V1 Predictions Min -87.7111 +trainer/VF Loss 0.0834603 +expl/num steps total 679000 +expl/num paths total 894 +expl/path length Mean 500 +expl/path length Std 226 +expl/path length Max 726 +expl/path length Min 274 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00617516 +expl/Actions Std 0.825501 +expl/Actions Max 2.20067 +expl/Actions Min -2.25363 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 624483 +eval/num paths total 683 +eval/path length Mean 473 +eval/path length Std 0 +eval/path length Max 473 +eval/path length Min 473 +eval/Rewards Mean 0.00211416 +eval/Rewards Std 0.0459314 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0257131 +eval/Actions Std 0.726726 +eval/Actions Max 0.999527 +eval/Actions Min -0.999046 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.27996e-05 +time/evaluation sampling (s) 5.20267 +time/exploration sampling (s) 6.61239 +time/logging (s) 0.00968291 +time/saving (s) 0.0114536 +time/training (s) 19.9597 +time/epoch (s) 31.7959 +time/total (s) 17125.4 +Epoch -322 +------------------------------ ---------------- +2022-05-15 22:48:24.820852 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -321 finished +------------------------------ --------------- +epoch -321 +replay_buffer/size 999047 +trainer/num train calls 680000 +trainer/QF1 Loss 1.82753 +trainer/QF2 Loss 1.88568 +trainer/Policy Loss 6.53916 +trainer/Q1 Predictions Mean -75.45 +trainer/Q1 Predictions Std 16.6838 +trainer/Q1 Predictions Max -0.849871 +trainer/Q1 Predictions Min -87.59 +trainer/Q2 Predictions Mean -75.4087 +trainer/Q2 Predictions Std 16.7396 +trainer/Q2 Predictions Max -0.431422 +trainer/Q2 Predictions Min -87.6804 +trainer/Q Targets Mean -75.2933 +trainer/Q Targets Std 16.8556 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7525 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0119551 +trainer/policy/mean Std 0.731127 +trainer/policy/mean Max 0.998377 +trainer/policy/mean Min -0.999473 +trainer/policy/std Mean 0.412316 +trainer/policy/std Std 0.0198118 +trainer/policy/std Max 0.435129 +trainer/policy/std Min 0.38133 +trainer/Advantage Weights Mean 1.85268 +trainer/Advantage Weights Std 8.97269 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.3648e-10 +trainer/Advantage Score Mean -0.345642 +trainer/Advantage Score Std 0.469809 +trainer/Advantage Score Max 0.85704 +trainer/Advantage Score Min -2.27148 +trainer/V1 Predictions Mean -75.135 +trainer/V1 Predictions Std 17.0163 +trainer/V1 Predictions Max -0.271445 +trainer/V1 Predictions Min -87.5918 +trainer/VF Loss 0.0400819 +expl/num steps total 680000 +expl/num paths total 895 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0527291 +expl/Actions Std 0.8308 +expl/Actions Max 2.35678 +expl/Actions Min -2.11917 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 625217 +eval/num paths total 684 +eval/path length Mean 734 +eval/path length Std 0 +eval/path length Max 734 +eval/path length Min 734 +eval/Rewards Mean 0.0013624 +eval/Rewards Std 0.0368855 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0413748 +eval/Actions Std 0.753823 +eval/Actions Max 0.999407 +eval/Actions Min -0.999375 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.2795e-05 +time/evaluation sampling (s) 4.53242 +time/exploration sampling (s) 6.30073 +time/logging (s) 0.00791201 +time/saving (s) 0.0111463 +time/training (s) 19.3917 +time/epoch (s) 30.244 +time/total (s) 17155.6 +Epoch -321 +------------------------------ --------------- +2022-05-15 22:48:55.025132 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -320 finished +------------------------------ ---------------- +epoch -320 +replay_buffer/size 999047 +trainer/num train calls 681000 +trainer/QF1 Loss 0.658685 +trainer/QF2 Loss 0.759988 +trainer/Policy Loss 18.082 +trainer/Q1 Predictions Mean -75.5049 +trainer/Q1 Predictions Std 15.9226 +trainer/Q1 Predictions Max -0.23084 +trainer/Q1 Predictions Min -87.0701 +trainer/Q2 Predictions Mean -75.5004 +trainer/Q2 Predictions Std 15.9362 +trainer/Q2 Predictions Max -0.340711 +trainer/Q2 Predictions Min -87.1976 +trainer/Q Targets Mean -75.6787 +trainer/Q Targets Std 15.7334 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6186 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00443573 +trainer/policy/mean Std 0.730474 +trainer/policy/mean Max 0.999461 +trainer/policy/mean Min -0.999747 +trainer/policy/std Mean 0.412183 +trainer/policy/std Std 0.019711 +trainer/policy/std Max 0.432727 +trainer/policy/std Min 0.379963 +trainer/Advantage Weights Mean 5.78659 +trainer/Advantage Weights Std 20.5996 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.99323e-14 +trainer/Advantage Score Mean -0.288295 +trainer/Advantage Score Std 0.535182 +trainer/Advantage Score Max 3.43175 +trainer/Advantage Score Min -3.06281 +trainer/V1 Predictions Mean -75.4413 +trainer/V1 Predictions Std 15.8836 +trainer/V1 Predictions Max 0.117469 +trainer/V1 Predictions Min -87.4626 +trainer/VF Loss 0.089659 +expl/num steps total 681000 +expl/num paths total 896 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.202978 +expl/Actions Std 0.833305 +expl/Actions Max 2.35788 +expl/Actions Min -2.51596 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 625668 +eval/num paths total 685 +eval/path length Mean 451 +eval/path length Std 0 +eval/path length Max 451 +eval/path length Min 451 +eval/Rewards Mean 0.00221729 +eval/Rewards Std 0.0470359 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.0108841 +eval/Actions Std 0.73691 +eval/Actions Max 0.999619 +eval/Actions Min -0.9987 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.24612e-06 +time/evaluation sampling (s) 4.62675 +time/exploration sampling (s) 6.01556 +time/logging (s) 0.0094536 +time/saving (s) 0.0163814 +time/training (s) 19.5249 +time/epoch (s) 30.193 +time/total (s) 17185.8 +Epoch -320 +------------------------------ ---------------- +2022-05-15 22:49:26.262626 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -319 finished +------------------------------ ---------------- +epoch -319 +replay_buffer/size 999047 +trainer/num train calls 682000 +trainer/QF1 Loss 12.4136 +trainer/QF2 Loss 12.4529 +trainer/Policy Loss 19.3435 +trainer/Q1 Predictions Mean -72.5998 +trainer/Q1 Predictions Std 19.2199 +trainer/Q1 Predictions Max -0.981767 +trainer/Q1 Predictions Min -87.2812 +trainer/Q2 Predictions Mean -72.6923 +trainer/Q2 Predictions Std 19.2247 +trainer/Q2 Predictions Max -0.975616 +trainer/Q2 Predictions Min -87.2212 +trainer/Q Targets Mean -73.2618 +trainer/Q Targets Std 18.7345 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8863 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00640108 +trainer/policy/mean Std 0.730237 +trainer/policy/mean Max 0.999874 +trainer/policy/mean Min -0.999888 +trainer/policy/std Mean 0.414161 +trainer/policy/std Std 0.0201407 +trainer/policy/std Max 0.435025 +trainer/policy/std Min 0.381047 +trainer/Advantage Weights Mean 5.57789 +trainer/Advantage Weights Std 20.3798 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.59799e-12 +trainer/Advantage Score Mean -0.328045 +trainer/Advantage Score Std 0.569932 +trainer/Advantage Score Max 1.22481 +trainer/Advantage Score Min -2.66763 +trainer/V1 Predictions Mean -72.7579 +trainer/V1 Predictions Std 19.1619 +trainer/V1 Predictions Max -0.586437 +trainer/V1 Predictions Min -87.8011 +trainer/VF Loss 0.0677304 +expl/num steps total 682000 +expl/num paths total 898 +expl/path length Mean 500 +expl/path length Std 252 +expl/path length Max 752 +expl/path length Min 248 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0313773 +expl/Actions Std 0.830075 +expl/Actions Max 2.44554 +expl/Actions Min -2.41674 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 626083 +eval/num paths total 686 +eval/path length Mean 415 +eval/path length Std 0 +eval/path length Max 415 +eval/path length Min 415 +eval/Rewards Mean 0.00240964 +eval/Rewards Std 0.0490289 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0198224 +eval/Actions Std 0.747197 +eval/Actions Max 0.999739 +eval/Actions Min -0.998845 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 6.32321e-06 +time/evaluation sampling (s) 4.59728 +time/exploration sampling (s) 7.10892 +time/logging (s) 0.0102063 +time/saving (s) 0.0182856 +time/training (s) 19.4866 +time/epoch (s) 31.2213 +time/total (s) 17217.1 +Epoch -319 +------------------------------ ---------------- +2022-05-15 22:49:57.111476 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -318 finished +------------------------------ ---------------- +epoch -318 +replay_buffer/size 999047 +trainer/num train calls 683000 +trainer/QF1 Loss 0.684613 +trainer/QF2 Loss 0.674648 +trainer/Policy Loss 7.71971 +trainer/Q1 Predictions Mean -74.4363 +trainer/Q1 Predictions Std 16.726 +trainer/Q1 Predictions Max -1.94732 +trainer/Q1 Predictions Min -87.3405 +trainer/Q2 Predictions Mean -74.3744 +trainer/Q2 Predictions Std 16.6882 +trainer/Q2 Predictions Max -1.94295 +trainer/Q2 Predictions Min -87.369 +trainer/Q Targets Mean -74.6211 +trainer/Q Targets Std 16.7797 +trainer/Q Targets Max -4.90341 +trainer/Q Targets Min -87.4658 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000912493 +trainer/policy/mean Std 0.71222 +trainer/policy/mean Max 0.999181 +trainer/policy/mean Min -0.999375 +trainer/policy/std Mean 0.414182 +trainer/policy/std Std 0.0199671 +trainer/policy/std Max 0.436919 +trainer/policy/std Min 0.382941 +trainer/Advantage Weights Mean 2.02706 +trainer/Advantage Weights Std 12.4656 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.20864e-22 +trainer/Advantage Score Mean -0.445008 +trainer/Advantage Score Std 0.543863 +trainer/Advantage Score Max 1.07367 +trainer/Advantage Score Min -5.04674 +trainer/V1 Predictions Mean -74.3823 +trainer/V1 Predictions Std 16.8785 +trainer/V1 Predictions Max -3.38983 +trainer/V1 Predictions Min -87.5015 +trainer/VF Loss 0.0601743 +expl/num steps total 683000 +expl/num paths total 899 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0108912 +expl/Actions Std 0.838957 +expl/Actions Max 2.40214 +expl/Actions Min -2.48134 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 627083 +eval/num paths total 687 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.205146 +eval/Actions Std 0.70066 +eval/Actions Max 0.999778 +eval/Actions Min -0.999239 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.05291e-05 +time/evaluation sampling (s) 4.92883 +time/exploration sampling (s) 5.97979 +time/logging (s) 0.0117431 +time/saving (s) 0.0162571 +time/training (s) 19.8943 +time/epoch (s) 30.8309 +time/total (s) 17247.9 +Epoch -318 +------------------------------ ---------------- +2022-05-15 22:50:27.685779 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -317 finished +------------------------------ ---------------- +epoch -317 +replay_buffer/size 999047 +trainer/num train calls 684000 +trainer/QF1 Loss 0.563313 +trainer/QF2 Loss 0.537891 +trainer/Policy Loss 10.9891 +trainer/Q1 Predictions Mean -74.0133 +trainer/Q1 Predictions Std 18.8315 +trainer/Q1 Predictions Max -0.526677 +trainer/Q1 Predictions Min -87.5674 +trainer/Q2 Predictions Mean -73.9698 +trainer/Q2 Predictions Std 18.798 +trainer/Q2 Predictions Max -1.70748 +trainer/Q2 Predictions Min -87.4523 +trainer/Q Targets Mean -74.0968 +trainer/Q Targets Std 18.6507 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4618 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0171865 +trainer/policy/mean Std 0.722163 +trainer/policy/mean Max 0.999849 +trainer/policy/mean Min -0.999124 +trainer/policy/std Mean 0.413476 +trainer/policy/std Std 0.0192958 +trainer/policy/std Max 0.436769 +trainer/policy/std Min 0.384323 +trainer/Advantage Weights Mean 1.78265 +trainer/Advantage Weights Std 11.0867 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.83223e-20 +trainer/Advantage Score Mean -0.463348 +trainer/Advantage Score Std 0.695808 +trainer/Advantage Score Max 3.12069 +trainer/Advantage Score Min -4.54462 +trainer/V1 Predictions Mean -73.814 +trainer/V1 Predictions Std 18.9714 +trainer/V1 Predictions Max -0.209075 +trainer/V1 Predictions Min -87.3349 +trainer/VF Loss 0.105471 +expl/num steps total 684000 +expl/num paths total 901 +expl/path length Mean 500 +expl/path length Std 434 +expl/path length Max 934 +expl/path length Min 66 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0281025 +expl/Actions Std 0.825455 +expl/Actions Max 2.28832 +expl/Actions Min -2.55722 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 628083 +eval/num paths total 688 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0472454 +eval/Actions Std 0.670943 +eval/Actions Max 0.999952 +eval/Actions Min -0.999388 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.4314e-05 +time/evaluation sampling (s) 4.53955 +time/exploration sampling (s) 6.54417 +time/logging (s) 0.0117135 +time/saving (s) 0.0139964 +time/training (s) 19.4505 +time/epoch (s) 30.56 +time/total (s) 17278.5 +Epoch -317 +------------------------------ ---------------- +2022-05-15 22:50:58.058522 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -316 finished +------------------------------ ---------------- +epoch -316 +replay_buffer/size 999047 +trainer/num train calls 685000 +trainer/QF1 Loss 0.832036 +trainer/QF2 Loss 0.859774 +trainer/Policy Loss 8.89219 +trainer/Q1 Predictions Mean -74.5439 +trainer/Q1 Predictions Std 16.9993 +trainer/Q1 Predictions Max -0.911656 +trainer/Q1 Predictions Min -87.7898 +trainer/Q2 Predictions Mean -74.5544 +trainer/Q2 Predictions Std 17.0998 +trainer/Q2 Predictions Max 0.35295 +trainer/Q2 Predictions Min -87.5858 +trainer/Q Targets Mean -74.4992 +trainer/Q Targets Std 17.0434 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6836 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0018283 +trainer/policy/mean Std 0.720345 +trainer/policy/mean Max 0.999819 +trainer/policy/mean Min -0.998523 +trainer/policy/std Mean 0.414293 +trainer/policy/std Std 0.0201926 +trainer/policy/std Max 0.439279 +trainer/policy/std Min 0.383854 +trainer/Advantage Weights Mean 2.94329 +trainer/Advantage Weights Std 14.177 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.6682e-10 +trainer/Advantage Score Mean -0.34708 +trainer/Advantage Score Std 0.478273 +trainer/Advantage Score Max 1.96552 +trainer/Advantage Score Min -2.09888 +trainer/V1 Predictions Mean -74.2609 +trainer/V1 Predictions Std 17.1068 +trainer/V1 Predictions Max -1.66152 +trainer/V1 Predictions Min -87.5692 +trainer/VF Loss 0.0598477 +expl/num steps total 685000 +expl/num paths total 902 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00654338 +expl/Actions Std 0.822717 +expl/Actions Max 2.41424 +expl/Actions Min -2.74729 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 628563 +eval/num paths total 689 +eval/path length Mean 480 +eval/path length Std 0 +eval/path length Max 480 +eval/path length Min 480 +eval/Rewards Mean 0.00208333 +eval/Rewards Std 0.045596 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00451842 +eval/Actions Std 0.706573 +eval/Actions Max 0.999253 +eval/Actions Min -0.99884 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.68192e-05 +time/evaluation sampling (s) 4.35898 +time/exploration sampling (s) 6.80301 +time/logging (s) 0.00608896 +time/saving (s) 0.0150339 +time/training (s) 19.1651 +time/epoch (s) 30.3482 +time/total (s) 17308.8 +Epoch -316 +------------------------------ ---------------- +2022-05-15 22:51:28.425718 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -315 finished +------------------------------ ---------------- +epoch -315 +replay_buffer/size 999047 +trainer/num train calls 686000 +trainer/QF1 Loss 0.817355 +trainer/QF2 Loss 0.694778 +trainer/Policy Loss 11.9988 +trainer/Q1 Predictions Mean -75.0583 +trainer/Q1 Predictions Std 16.294 +trainer/Q1 Predictions Max -1.24553 +trainer/Q1 Predictions Min -88.7936 +trainer/Q2 Predictions Mean -75.0415 +trainer/Q2 Predictions Std 16.4472 +trainer/Q2 Predictions Max -2.44997 +trainer/Q2 Predictions Min -88.6698 +trainer/Q Targets Mean -74.7024 +trainer/Q Targets Std 16.5105 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.4769 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0194665 +trainer/policy/mean Std 0.726503 +trainer/policy/mean Max 0.999826 +trainer/policy/mean Min -0.999113 +trainer/policy/std Mean 0.413843 +trainer/policy/std Std 0.0198181 +trainer/policy/std Max 0.435387 +trainer/policy/std Min 0.383812 +trainer/Advantage Weights Mean 2.11936 +trainer/Advantage Weights Std 12.5819 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.19236e-20 +trainer/Advantage Score Mean -0.480518 +trainer/Advantage Score Std 0.60558 +trainer/Advantage Score Max 1.54113 +trainer/Advantage Score Min -4.52667 +trainer/V1 Predictions Mean -74.5077 +trainer/V1 Predictions Std 16.5027 +trainer/V1 Predictions Max 0.0475503 +trainer/V1 Predictions Min -88.0075 +trainer/VF Loss 0.0787143 +expl/num steps total 686000 +expl/num paths total 903 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0889542 +expl/Actions Std 0.813087 +expl/Actions Max 2.32022 +expl/Actions Min -2.25324 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 629563 +eval/num paths total 690 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.269929 +eval/Actions Std 0.638886 +eval/Actions Max 0.999939 +eval/Actions Min -0.999449 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.87876e-06 +time/evaluation sampling (s) 4.45567 +time/exploration sampling (s) 6.74225 +time/logging (s) 0.00767714 +time/saving (s) 0.0107361 +time/training (s) 19.142 +time/epoch (s) 30.3584 +time/total (s) 17339.2 +Epoch -315 +------------------------------ ---------------- +2022-05-15 22:51:59.439489 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -314 finished +------------------------------ ---------------- +epoch -314 +replay_buffer/size 999047 +trainer/num train calls 687000 +trainer/QF1 Loss 1.17072 +trainer/QF2 Loss 1.03047 +trainer/Policy Loss 6.07179 +trainer/Q1 Predictions Mean -74.3506 +trainer/Q1 Predictions Std 17.5657 +trainer/Q1 Predictions Max -0.509119 +trainer/Q1 Predictions Min -88.3982 +trainer/Q2 Predictions Mean -74.3162 +trainer/Q2 Predictions Std 17.591 +trainer/Q2 Predictions Max -0.442727 +trainer/Q2 Predictions Min -88.3658 +trainer/Q Targets Mean -73.8712 +trainer/Q Targets Std 17.5964 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.831 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0177143 +trainer/policy/mean Std 0.727517 +trainer/policy/mean Max 0.999924 +trainer/policy/mean Min -0.999855 +trainer/policy/std Mean 0.41374 +trainer/policy/std Std 0.0198596 +trainer/policy/std Max 0.434042 +trainer/policy/std Min 0.386643 +trainer/Advantage Weights Mean 1.58032 +trainer/Advantage Weights Std 10.3667 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.33999e-15 +trainer/Advantage Score Mean -0.538354 +trainer/Advantage Score Std 0.558142 +trainer/Advantage Score Max 1.58743 +trainer/Advantage Score Min -3.36886 +trainer/V1 Predictions Mean -73.624 +trainer/V1 Predictions Std 17.6191 +trainer/V1 Predictions Max 0.948814 +trainer/V1 Predictions Min -87.5306 +trainer/VF Loss 0.0706847 +expl/num steps total 687000 +expl/num paths total 905 +expl/path length Mean 500 +expl/path length Std 310 +expl/path length Max 810 +expl/path length Min 190 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0140027 +expl/Actions Std 0.837798 +expl/Actions Max 2.21353 +expl/Actions Min -2.48931 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 630563 +eval/num paths total 691 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0952414 +eval/Actions Std 0.707195 +eval/Actions Max 0.999121 +eval/Actions Min -0.99864 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 6.52811e-06 +time/evaluation sampling (s) 5.04109 +time/exploration sampling (s) 6.91443 +time/logging (s) 0.0129849 +time/saving (s) 0.0214451 +time/training (s) 19.0171 +time/epoch (s) 31.007 +time/total (s) 17370.2 +Epoch -314 +------------------------------ ---------------- +2022-05-15 22:52:30.134374 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -313 finished +------------------------------ ---------------- +epoch -313 +replay_buffer/size 999047 +trainer/num train calls 688000 +trainer/QF1 Loss 1.6168 +trainer/QF2 Loss 1.74608 +trainer/Policy Loss 40.812 +trainer/Q1 Predictions Mean -72.9832 +trainer/Q1 Predictions Std 18.8692 +trainer/Q1 Predictions Max -0.984873 +trainer/Q1 Predictions Min -86.6692 +trainer/Q2 Predictions Mean -72.9332 +trainer/Q2 Predictions Std 18.9522 +trainer/Q2 Predictions Max -0.505572 +trainer/Q2 Predictions Min -86.5558 +trainer/Q Targets Mean -73.197 +trainer/Q Targets Std 19.0006 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2826 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0193112 +trainer/policy/mean Std 0.734031 +trainer/policy/mean Max 0.998679 +trainer/policy/mean Min -0.998557 +trainer/policy/std Mean 0.411399 +trainer/policy/std Std 0.0194519 +trainer/policy/std Max 0.431745 +trainer/policy/std Min 0.386455 +trainer/Advantage Weights Mean 8.0318 +trainer/Advantage Weights Std 23.8845 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.3027e-10 +trainer/Advantage Score Mean -0.155664 +trainer/Advantage Score Std 0.497285 +trainer/Advantage Score Max 2.55607 +trainer/Advantage Score Min -2.18311 +trainer/V1 Predictions Mean -72.9441 +trainer/V1 Predictions Std 19.0417 +trainer/V1 Predictions Max -0.255516 +trainer/V1 Predictions Min -87.0225 +trainer/VF Loss 0.0803328 +expl/num steps total 688000 +expl/num paths total 907 +expl/path length Mean 500 +expl/path length Std 239 +expl/path length Max 739 +expl/path length Min 261 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00433578 +expl/Actions Std 0.817507 +expl/Actions Max 2.21757 +expl/Actions Min -2.31569 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 631563 +eval/num paths total 692 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0625838 +eval/Actions Std 0.862256 +eval/Actions Max 0.999513 +eval/Actions Min -0.99985 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.59073e-06 +time/evaluation sampling (s) 5.03841 +time/exploration sampling (s) 6.282 +time/logging (s) 0.0119084 +time/saving (s) 0.0164714 +time/training (s) 19.3298 +time/epoch (s) 30.6786 +time/total (s) 17400.9 +Epoch -313 +------------------------------ ---------------- +2022-05-15 22:53:00.857782 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -312 finished +------------------------------ ---------------- +epoch -312 +replay_buffer/size 999047 +trainer/num train calls 689000 +trainer/QF1 Loss 1.0885 +trainer/QF2 Loss 1.22806 +trainer/Policy Loss 11.8264 +trainer/Q1 Predictions Mean -73.2258 +trainer/Q1 Predictions Std 19.5425 +trainer/Q1 Predictions Max -0.230377 +trainer/Q1 Predictions Min -87.6767 +trainer/Q2 Predictions Mean -73.1806 +trainer/Q2 Predictions Std 19.5552 +trainer/Q2 Predictions Max 0.681991 +trainer/Q2 Predictions Min -87.2235 +trainer/Q Targets Mean -73.2924 +trainer/Q Targets Std 19.4568 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1866 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0157189 +trainer/policy/mean Std 0.720957 +trainer/policy/mean Max 0.99987 +trainer/policy/mean Min -0.999418 +trainer/policy/std Mean 0.412895 +trainer/policy/std Std 0.0183669 +trainer/policy/std Max 0.432603 +trainer/policy/std Min 0.385762 +trainer/Advantage Weights Mean 3.47139 +trainer/Advantage Weights Std 15.9457 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.73258e-15 +trainer/Advantage Score Mean -0.338806 +trainer/Advantage Score Std 0.490538 +trainer/Advantage Score Max 1.77058 +trainer/Advantage Score Min -3.39892 +trainer/V1 Predictions Mean -73.1109 +trainer/V1 Predictions Std 19.4702 +trainer/V1 Predictions Max -0.262549 +trainer/V1 Predictions Min -86.991 +trainer/VF Loss 0.0579178 +expl/num steps total 689000 +expl/num paths total 909 +expl/path length Mean 500 +expl/path length Std 108 +expl/path length Max 608 +expl/path length Min 392 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00552738 +expl/Actions Std 0.818693 +expl/Actions Max 2.1992 +expl/Actions Min -2.54564 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 632563 +eval/num paths total 693 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.142754 +eval/Actions Std 0.718187 +eval/Actions Max 0.999802 +eval/Actions Min -0.999206 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.04657e-05 +time/evaluation sampling (s) 4.96658 +time/exploration sampling (s) 6.10768 +time/logging (s) 0.00857315 +time/saving (s) 0.012194 +time/training (s) 19.6088 +time/epoch (s) 30.7038 +time/total (s) 17431.6 +Epoch -312 +------------------------------ ---------------- +2022-05-15 22:53:31.449432 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -311 finished +------------------------------ ---------------- +epoch -311 +replay_buffer/size 999047 +trainer/num train calls 690000 +trainer/QF1 Loss 0.880323 +trainer/QF2 Loss 0.814646 +trainer/Policy Loss 22.13 +trainer/Q1 Predictions Mean -74.4324 +trainer/Q1 Predictions Std 17.7306 +trainer/Q1 Predictions Max -0.869054 +trainer/Q1 Predictions Min -87.033 +trainer/Q2 Predictions Mean -74.3987 +trainer/Q2 Predictions Std 17.7534 +trainer/Q2 Predictions Max -0.983862 +trainer/Q2 Predictions Min -87.1144 +trainer/Q Targets Mean -74.4086 +trainer/Q Targets Std 17.8585 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4078 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0172814 +trainer/policy/mean Std 0.736198 +trainer/policy/mean Max 0.9993 +trainer/policy/mean Min -0.998633 +trainer/policy/std Mean 0.415046 +trainer/policy/std Std 0.0187149 +trainer/policy/std Max 0.434706 +trainer/policy/std Min 0.384827 +trainer/Advantage Weights Mean 4.39143 +trainer/Advantage Weights Std 16.2501 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.98862e-17 +trainer/Advantage Score Mean -0.372868 +trainer/Advantage Score Std 0.633058 +trainer/Advantage Score Max 1.06171 +trainer/Advantage Score Min -3.84565 +trainer/V1 Predictions Mean -74.1679 +trainer/V1 Predictions Std 17.9974 +trainer/V1 Predictions Max 1.62397 +trainer/V1 Predictions Min -87.3008 +trainer/VF Loss 0.0734042 +expl/num steps total 690000 +expl/num paths total 910 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0517039 +expl/Actions Std 0.822091 +expl/Actions Max 2.45483 +expl/Actions Min -2.15769 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 633563 +eval/num paths total 694 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.130137 +eval/Actions Std 0.750989 +eval/Actions Max 0.999579 +eval/Actions Min -0.999811 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0638e-05 +time/evaluation sampling (s) 4.70458 +time/exploration sampling (s) 6.30758 +time/logging (s) 0.0126033 +time/saving (s) 0.0195667 +time/training (s) 19.5394 +time/epoch (s) 30.5837 +time/total (s) 17462.2 +Epoch -311 +------------------------------ ---------------- +2022-05-15 22:54:02.546920 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -310 finished +------------------------------ --------------- +epoch -310 +replay_buffer/size 999047 +trainer/num train calls 691000 +trainer/QF1 Loss 0.602464 +trainer/QF2 Loss 0.576834 +trainer/Policy Loss 18.465 +trainer/Q1 Predictions Mean -74.1293 +trainer/Q1 Predictions Std 17.8561 +trainer/Q1 Predictions Max -1.25652 +trainer/Q1 Predictions Min -88.1238 +trainer/Q2 Predictions Mean -74.1561 +trainer/Q2 Predictions Std 17.9186 +trainer/Q2 Predictions Max -1.45832 +trainer/Q2 Predictions Min -88.2349 +trainer/Q Targets Mean -74.0765 +trainer/Q Targets Std 18.0477 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9529 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00881176 +trainer/policy/mean Std 0.722767 +trainer/policy/mean Max 0.999649 +trainer/policy/mean Min -0.996489 +trainer/policy/std Mean 0.412912 +trainer/policy/std Std 0.0208504 +trainer/policy/std Max 0.434072 +trainer/policy/std Min 0.379408 +trainer/Advantage Weights Mean 5.04518 +trainer/Advantage Weights Std 20.2536 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.7409e-12 +trainer/Advantage Score Mean -0.322035 +trainer/Advantage Score Std 0.529629 +trainer/Advantage Score Max 1.69143 +trainer/Advantage Score Min -2.66227 +trainer/V1 Predictions Mean -73.8975 +trainer/V1 Predictions Std 17.958 +trainer/V1 Predictions Max -2.03764 +trainer/V1 Predictions Min -87.877 +trainer/VF Loss 0.0695866 +expl/num steps total 691000 +expl/num paths total 912 +expl/path length Mean 500 +expl/path length Std 174 +expl/path length Max 674 +expl/path length Min 326 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0326838 +expl/Actions Std 0.831275 +expl/Actions Max 2.32492 +expl/Actions Min -2.17892 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 634563 +eval/num paths total 695 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.107707 +eval/Actions Std 0.634302 +eval/Actions Max 0.998785 +eval/Actions Min -0.998889 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.3046e-05 +time/evaluation sampling (s) 5.08324 +time/exploration sampling (s) 6.45767 +time/logging (s) 0.0117434 +time/saving (s) 0.0168387 +time/training (s) 19.5037 +time/epoch (s) 31.0732 +time/total (s) 17493.3 +Epoch -310 +------------------------------ --------------- +2022-05-15 22:54:33.025804 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -309 finished +------------------------------ ---------------- +epoch -309 +replay_buffer/size 999047 +trainer/num train calls 692000 +trainer/QF1 Loss 0.822292 +trainer/QF2 Loss 0.808867 +trainer/Policy Loss 22.7693 +trainer/Q1 Predictions Mean -74.6011 +trainer/Q1 Predictions Std 17.2498 +trainer/Q1 Predictions Max -1.32331 +trainer/Q1 Predictions Min -87.5316 +trainer/Q2 Predictions Mean -74.5796 +trainer/Q2 Predictions Std 17.1998 +trainer/Q2 Predictions Max -1.96823 +trainer/Q2 Predictions Min -87.6044 +trainer/Q Targets Mean -74.5084 +trainer/Q Targets Std 17.3988 +trainer/Q Targets Max -1.53135 +trainer/Q Targets Min -87.7129 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0313364 +trainer/policy/mean Std 0.7289 +trainer/policy/mean Max 0.998509 +trainer/policy/mean Min -0.999457 +trainer/policy/std Mean 0.413467 +trainer/policy/std Std 0.0209465 +trainer/policy/std Max 0.437765 +trainer/policy/std Min 0.37967 +trainer/Advantage Weights Mean 4.69484 +trainer/Advantage Weights Std 18.7017 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.25132e-20 +trainer/Advantage Score Mean -0.363596 +trainer/Advantage Score Std 0.54771 +trainer/Advantage Score Max 1.13885 +trainer/Advantage Score Min -4.46045 +trainer/V1 Predictions Mean -74.3535 +trainer/V1 Predictions Std 17.3943 +trainer/V1 Predictions Max -0.15069 +trainer/V1 Predictions Min -88.0499 +trainer/VF Loss 0.064321 +expl/num steps total 692000 +expl/num paths total 914 +expl/path length Mean 500 +expl/path length Std 413 +expl/path length Max 913 +expl/path length Min 87 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0126311 +expl/Actions Std 0.815105 +expl/Actions Max 2.20922 +expl/Actions Min -2.33391 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 635563 +eval/num paths total 696 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0220636 +eval/Actions Std 0.717682 +eval/Actions Max 0.999925 +eval/Actions Min -0.999897 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.07042e-05 +time/evaluation sampling (s) 5.16084 +time/exploration sampling (s) 6.09919 +time/logging (s) 0.0124329 +time/saving (s) 0.0165606 +time/training (s) 19.1738 +time/epoch (s) 30.4629 +time/total (s) 17523.7 +Epoch -309 +------------------------------ ---------------- +2022-05-15 22:55:03.241271 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -308 finished +------------------------------ ---------------- +epoch -308 +replay_buffer/size 999047 +trainer/num train calls 693000 +trainer/QF1 Loss 0.656569 +trainer/QF2 Loss 0.545132 +trainer/Policy Loss 17.6015 +trainer/Q1 Predictions Mean -73.5061 +trainer/Q1 Predictions Std 18.7683 +trainer/Q1 Predictions Max -0.401691 +trainer/Q1 Predictions Min -87.5643 +trainer/Q2 Predictions Mean -73.506 +trainer/Q2 Predictions Std 18.7735 +trainer/Q2 Predictions Max 0.291027 +trainer/Q2 Predictions Min -87.6308 +trainer/Q Targets Mean -73.5697 +trainer/Q Targets Std 18.8508 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3338 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0232707 +trainer/policy/mean Std 0.723719 +trainer/policy/mean Max 0.999868 +trainer/policy/mean Min -0.999163 +trainer/policy/std Mean 0.412769 +trainer/policy/std Std 0.0219845 +trainer/policy/std Max 0.438572 +trainer/policy/std Min 0.377981 +trainer/Advantage Weights Mean 4.38829 +trainer/Advantage Weights Std 17.9095 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.36131e-17 +trainer/Advantage Score Mean -0.318347 +trainer/Advantage Score Std 0.566732 +trainer/Advantage Score Max 1.44539 +trainer/Advantage Score Min -3.82847 +trainer/V1 Predictions Mean -73.3164 +trainer/V1 Predictions Std 18.934 +trainer/V1 Predictions Max -0.102936 +trainer/V1 Predictions Min -87.7811 +trainer/VF Loss 0.0610586 +expl/num steps total 693000 +expl/num paths total 915 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.17523 +expl/Actions Std 0.786346 +expl/Actions Max 2.30771 +expl/Actions Min -2.08431 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 636048 +eval/num paths total 697 +eval/path length Mean 485 +eval/path length Std 0 +eval/path length Max 485 +eval/path length Min 485 +eval/Rewards Mean 0.00206186 +eval/Rewards Std 0.0453608 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00215801 +eval/Actions Std 0.737807 +eval/Actions Max 0.999848 +eval/Actions Min -0.99963 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.39968e-05 +time/evaluation sampling (s) 4.59767 +time/exploration sampling (s) 6.2609 +time/logging (s) 0.0101793 +time/saving (s) 0.018625 +time/training (s) 19.3125 +time/epoch (s) 30.1998 +time/total (s) 17553.9 +Epoch -308 +------------------------------ ---------------- +2022-05-15 22:55:33.816897 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -307 finished +------------------------------ ---------------- +epoch -307 +replay_buffer/size 999047 +trainer/num train calls 694000 +trainer/QF1 Loss 0.581335 +trainer/QF2 Loss 0.662453 +trainer/Policy Loss 35.7114 +trainer/Q1 Predictions Mean -73.2057 +trainer/Q1 Predictions Std 19.879 +trainer/Q1 Predictions Max -0.293407 +trainer/Q1 Predictions Min -86.9267 +trainer/Q2 Predictions Mean -73.2618 +trainer/Q2 Predictions Std 19.7585 +trainer/Q2 Predictions Max -0.176336 +trainer/Q2 Predictions Min -87.383 +trainer/Q Targets Mean -73.1643 +trainer/Q Targets Std 19.9494 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.701 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0150335 +trainer/policy/mean Std 0.738763 +trainer/policy/mean Max 0.999505 +trainer/policy/mean Min -0.999203 +trainer/policy/std Mean 0.412352 +trainer/policy/std Std 0.0205273 +trainer/policy/std Max 0.436944 +trainer/policy/std Min 0.381737 +trainer/Advantage Weights Mean 4.79506 +trainer/Advantage Weights Std 19.1484 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.40855e-14 +trainer/Advantage Score Mean -0.327681 +trainer/Advantage Score Std 0.493497 +trainer/Advantage Score Max 1.48438 +trainer/Advantage Score Min -3.05482 +trainer/V1 Predictions Mean -72.977 +trainer/V1 Predictions Std 19.9018 +trainer/V1 Predictions Max 0.704156 +trainer/V1 Predictions Min -87.5407 +trainer/VF Loss 0.0594855 +expl/num steps total 694000 +expl/num paths total 917 +expl/path length Mean 500 +expl/path length Std 214 +expl/path length Max 714 +expl/path length Min 286 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0382123 +expl/Actions Std 0.842396 +expl/Actions Max 2.30556 +expl/Actions Min -2.26354 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 636656 +eval/num paths total 698 +eval/path length Mean 608 +eval/path length Std 0 +eval/path length Max 608 +eval/path length Min 608 +eval/Rewards Mean 0.00164474 +eval/Rewards Std 0.040522 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0358783 +eval/Actions Std 0.730621 +eval/Actions Max 0.999479 +eval/Actions Min -0.999309 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.15765e-06 +time/evaluation sampling (s) 4.84019 +time/exploration sampling (s) 6.14929 +time/logging (s) 0.00952739 +time/saving (s) 0.0137436 +time/training (s) 19.5422 +time/epoch (s) 30.5549 +time/total (s) 17584.5 +Epoch -307 +------------------------------ ---------------- +2022-05-15 22:56:03.728151 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -306 finished +------------------------------ ---------------- +epoch -306 +replay_buffer/size 999047 +trainer/num train calls 695000 +trainer/QF1 Loss 0.916984 +trainer/QF2 Loss 0.982448 +trainer/Policy Loss 9.95931 +trainer/Q1 Predictions Mean -73.9079 +trainer/Q1 Predictions Std 19.2332 +trainer/Q1 Predictions Max -0.832259 +trainer/Q1 Predictions Min -88.0028 +trainer/Q2 Predictions Mean -73.9044 +trainer/Q2 Predictions Std 19.22 +trainer/Q2 Predictions Max -1.1175 +trainer/Q2 Predictions Min -87.2089 +trainer/Q Targets Mean -73.5649 +trainer/Q Targets Std 19.6003 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7819 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00306955 +trainer/policy/mean Std 0.732343 +trainer/policy/mean Max 0.999829 +trainer/policy/mean Min -0.999624 +trainer/policy/std Mean 0.410233 +trainer/policy/std Std 0.0206315 +trainer/policy/std Max 0.436183 +trainer/policy/std Min 0.377447 +trainer/Advantage Weights Mean 3.41189 +trainer/Advantage Weights Std 16.5878 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.77629e-22 +trainer/Advantage Score Mean -0.399562 +trainer/Advantage Score Std 0.684228 +trainer/Advantage Score Max 1.86091 +trainer/Advantage Score Min -4.87434 +trainer/V1 Predictions Mean -73.3917 +trainer/V1 Predictions Std 19.5401 +trainer/V1 Predictions Max 0.526042 +trainer/V1 Predictions Min -87.4241 +trainer/VF Loss 0.0843649 +expl/num steps total 695000 +expl/num paths total 918 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0504103 +expl/Actions Std 0.850156 +expl/Actions Max 2.13822 +expl/Actions Min -2.24091 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 637656 +eval/num paths total 699 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0802963 +eval/Actions Std 0.789773 +eval/Actions Max 0.999891 +eval/Actions Min -0.999834 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.2476e-05 +time/evaluation sampling (s) 4.48488 +time/exploration sampling (s) 5.85639 +time/logging (s) 0.0143393 +time/saving (s) 0.0196959 +time/training (s) 19.5223 +time/epoch (s) 29.8976 +time/total (s) 17614.4 +Epoch -306 +------------------------------ ---------------- +2022-05-15 22:56:34.403791 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -305 finished +------------------------------ ---------------- +epoch -305 +replay_buffer/size 999047 +trainer/num train calls 696000 +trainer/QF1 Loss 0.600644 +trainer/QF2 Loss 0.537964 +trainer/Policy Loss 22.9277 +trainer/Q1 Predictions Mean -75.5312 +trainer/Q1 Predictions Std 15.8725 +trainer/Q1 Predictions Max -1.34953 +trainer/Q1 Predictions Min -87.0396 +trainer/Q2 Predictions Mean -75.5123 +trainer/Q2 Predictions Std 15.993 +trainer/Q2 Predictions Max -0.325218 +trainer/Q2 Predictions Min -87.4849 +trainer/Q Targets Mean -75.7227 +trainer/Q Targets Std 16.0572 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2416 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.014555 +trainer/policy/mean Std 0.728662 +trainer/policy/mean Max 0.998933 +trainer/policy/mean Min -0.999784 +trainer/policy/std Mean 0.410711 +trainer/policy/std Std 0.0211079 +trainer/policy/std Max 0.434945 +trainer/policy/std Min 0.377439 +trainer/Advantage Weights Mean 4.016 +trainer/Advantage Weights Std 16.2367 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.56656e-12 +trainer/Advantage Score Mean -0.339573 +trainer/Advantage Score Std 0.527435 +trainer/Advantage Score Max 0.954792 +trainer/Advantage Score Min -2.71821 +trainer/V1 Predictions Mean -75.4961 +trainer/V1 Predictions Std 16.27 +trainer/V1 Predictions Max 0.855588 +trainer/V1 Predictions Min -87.1085 +trainer/VF Loss 0.0520724 +expl/num steps total 696000 +expl/num paths total 919 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0488408 +expl/Actions Std 0.804706 +expl/Actions Max 2.57332 +expl/Actions Min -2.14527 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 638656 +eval/num paths total 700 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.147022 +eval/Actions Std 0.642511 +eval/Actions Max 0.999972 +eval/Actions Min -0.999238 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.08052e-05 +time/evaluation sampling (s) 5.15654 +time/exploration sampling (s) 6.44859 +time/logging (s) 0.00896667 +time/saving (s) 0.0168199 +time/training (s) 19.0193 +time/epoch (s) 30.6502 +time/total (s) 17645.1 +Epoch -305 +------------------------------ ---------------- +2022-05-15 22:57:05.216368 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -304 finished +------------------------------ ---------------- +epoch -304 +replay_buffer/size 999047 +trainer/num train calls 697000 +trainer/QF1 Loss 2.00871 +trainer/QF2 Loss 1.99936 +trainer/Policy Loss 15.9796 +trainer/Q1 Predictions Mean -73.4473 +trainer/Q1 Predictions Std 19.1554 +trainer/Q1 Predictions Max -0.0164513 +trainer/Q1 Predictions Min -86.8979 +trainer/Q2 Predictions Mean -73.3955 +trainer/Q2 Predictions Std 19.1794 +trainer/Q2 Predictions Max 0.39109 +trainer/Q2 Predictions Min -86.9379 +trainer/Q Targets Mean -73.4275 +trainer/Q Targets Std 19.2849 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2173 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00224533 +trainer/policy/mean Std 0.724916 +trainer/policy/mean Max 0.999208 +trainer/policy/mean Min -0.998378 +trainer/policy/std Mean 0.410061 +trainer/policy/std Std 0.0218884 +trainer/policy/std Max 0.434895 +trainer/policy/std Min 0.375709 +trainer/Advantage Weights Mean 3.30719 +trainer/Advantage Weights Std 13.8527 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.25228e-19 +trainer/Advantage Score Mean -0.322773 +trainer/Advantage Score Std 0.552208 +trainer/Advantage Score Max 0.685295 +trainer/Advantage Score Min -4.35241 +trainer/V1 Predictions Mean -73.2313 +trainer/V1 Predictions Std 19.5253 +trainer/V1 Predictions Max -0.10099 +trainer/V1 Predictions Min -87.0695 +trainer/VF Loss 0.0493082 +expl/num steps total 697000 +expl/num paths total 920 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0813848 +expl/Actions Std 0.790253 +expl/Actions Max 2.35705 +expl/Actions Min -2.31634 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 639167 +eval/num paths total 701 +eval/path length Mean 511 +eval/path length Std 0 +eval/path length Max 511 +eval/path length Min 511 +eval/Rewards Mean 0.00195695 +eval/Rewards Std 0.0441941 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.020321 +eval/Actions Std 0.735124 +eval/Actions Max 0.998703 +eval/Actions Min -0.999016 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.33081e-05 +time/evaluation sampling (s) 4.72604 +time/exploration sampling (s) 6.31619 +time/logging (s) 0.0101801 +time/saving (s) 0.0178405 +time/training (s) 19.7291 +time/epoch (s) 30.7994 +time/total (s) 17675.9 +Epoch -304 +------------------------------ ---------------- +2022-05-15 22:57:35.680015 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -303 finished +------------------------------ ---------------- +epoch -303 +replay_buffer/size 999047 +trainer/num train calls 698000 +trainer/QF1 Loss 0.920463 +trainer/QF2 Loss 0.990813 +trainer/Policy Loss 29.4514 +trainer/Q1 Predictions Mean -72.7969 +trainer/Q1 Predictions Std 17.8174 +trainer/Q1 Predictions Max -0.23902 +trainer/Q1 Predictions Min -86.5227 +trainer/Q2 Predictions Mean -72.8717 +trainer/Q2 Predictions Std 17.8859 +trainer/Q2 Predictions Max -0.0591531 +trainer/Q2 Predictions Min -86.6728 +trainer/Q Targets Mean -73.2155 +trainer/Q Targets Std 18.0286 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1523 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0200612 +trainer/policy/mean Std 0.730679 +trainer/policy/mean Max 0.999706 +trainer/policy/mean Min -0.998121 +trainer/policy/std Mean 0.411069 +trainer/policy/std Std 0.0203893 +trainer/policy/std Max 0.434136 +trainer/policy/std Min 0.378342 +trainer/Advantage Weights Mean 6.3465 +trainer/Advantage Weights Std 21.7757 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.2848e-14 +trainer/Advantage Score Mean -0.363859 +trainer/Advantage Score Std 0.60943 +trainer/Advantage Score Max 1.02399 +trainer/Advantage Score Min -3.14099 +trainer/V1 Predictions Mean -72.9536 +trainer/V1 Predictions Std 18.1753 +trainer/V1 Predictions Max 0.70233 +trainer/V1 Predictions Min -87.0221 +trainer/VF Loss 0.0707527 +expl/num steps total 698000 +expl/num paths total 922 +expl/path length Mean 500 +expl/path length Std 258 +expl/path length Max 758 +expl/path length Min 242 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0261036 +expl/Actions Std 0.830478 +expl/Actions Max 2.21598 +expl/Actions Min -2.29846 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 640167 +eval/num paths total 702 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.224953 +eval/Actions Std 0.534888 +eval/Actions Max 0.99985 +eval/Actions Min -0.999151 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.48572e-06 +time/evaluation sampling (s) 4.48372 +time/exploration sampling (s) 6.46243 +time/logging (s) 0.0116752 +time/saving (s) 0.0154713 +time/training (s) 19.4718 +time/epoch (s) 30.4451 +time/total (s) 17706.3 +Epoch -303 +------------------------------ ---------------- +2022-05-15 22:58:06.013673 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -302 finished +------------------------------ ---------------- +epoch -302 +replay_buffer/size 999047 +trainer/num train calls 699000 +trainer/QF1 Loss 0.948411 +trainer/QF2 Loss 1.01298 +trainer/Policy Loss 15.7501 +trainer/Q1 Predictions Mean -73.4233 +trainer/Q1 Predictions Std 18.4903 +trainer/Q1 Predictions Max -0.496057 +trainer/Q1 Predictions Min -87.8756 +trainer/Q2 Predictions Mean -73.4272 +trainer/Q2 Predictions Std 18.4382 +trainer/Q2 Predictions Max -0.59245 +trainer/Q2 Predictions Min -87.7831 +trainer/Q Targets Mean -73.3152 +trainer/Q Targets Std 18.8677 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7096 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.015771 +trainer/policy/mean Std 0.726543 +trainer/policy/mean Max 0.999195 +trainer/policy/mean Min -0.998762 +trainer/policy/std Mean 0.412015 +trainer/policy/std Std 0.0203567 +trainer/policy/std Max 0.434652 +trainer/policy/std Min 0.380797 +trainer/Advantage Weights Mean 3.88832 +trainer/Advantage Weights Std 15.3514 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.33283e-17 +trainer/Advantage Score Mean -0.408615 +trainer/Advantage Score Std 0.715496 +trainer/Advantage Score Max 1.98316 +trainer/Advantage Score Min -3.71516 +trainer/V1 Predictions Mean -73.1269 +trainer/V1 Predictions Std 18.7693 +trainer/V1 Predictions Max 0.302172 +trainer/V1 Predictions Min -87.5934 +trainer/VF Loss 0.104435 +expl/num steps total 699000 +expl/num paths total 923 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0584798 +expl/Actions Std 0.883397 +expl/Actions Max 2.29457 +expl/Actions Min -2.41251 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 640699 +eval/num paths total 703 +eval/path length Mean 532 +eval/path length Std 0 +eval/path length Max 532 +eval/path length Min 532 +eval/Rewards Mean 0.0018797 +eval/Rewards Std 0.0433147 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0323019 +eval/Actions Std 0.758949 +eval/Actions Max 0.999862 +eval/Actions Min -0.999507 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 6.41309e-06 +time/evaluation sampling (s) 4.46963 +time/exploration sampling (s) 7.02528 +time/logging (s) 0.0105049 +time/saving (s) 0.0153273 +time/training (s) 18.7944 +time/epoch (s) 30.3151 +time/total (s) 17736.6 +Epoch -302 +------------------------------ ---------------- +2022-05-15 22:58:36.675082 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -301 finished +------------------------------ ---------------- +epoch -301 +replay_buffer/size 999047 +trainer/num train calls 700000 +trainer/QF1 Loss 1.10728 +trainer/QF2 Loss 1.09709 +trainer/Policy Loss 7.44692 +trainer/Q1 Predictions Mean -72.7446 +trainer/Q1 Predictions Std 18.1745 +trainer/Q1 Predictions Max -0.829685 +trainer/Q1 Predictions Min -86.9532 +trainer/Q2 Predictions Mean -72.6532 +trainer/Q2 Predictions Std 18.1296 +trainer/Q2 Predictions Max -0.347562 +trainer/Q2 Predictions Min -86.6871 +trainer/Q Targets Mean -72.6146 +trainer/Q Targets Std 18.1403 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8815 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00284607 +trainer/policy/mean Std 0.728414 +trainer/policy/mean Max 0.999778 +trainer/policy/mean Min -0.999427 +trainer/policy/std Mean 0.412922 +trainer/policy/std Std 0.019985 +trainer/policy/std Max 0.435923 +trainer/policy/std Min 0.38505 +trainer/Advantage Weights Mean 1.7904 +trainer/Advantage Weights Std 10.5852 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.22703e-18 +trainer/Advantage Score Mean -0.532046 +trainer/Advantage Score Std 0.630801 +trainer/Advantage Score Max 1.71935 +trainer/Advantage Score Min -4.12419 +trainer/V1 Predictions Mean -72.3162 +trainer/V1 Predictions Std 18.4144 +trainer/V1 Predictions Max -0.777135 +trainer/V1 Predictions Min -86.6114 +trainer/VF Loss 0.0807405 +expl/num steps total 700000 +expl/num paths total 925 +expl/path length Mean 500 +expl/path length Std 308 +expl/path length Max 808 +expl/path length Min 192 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0330515 +expl/Actions Std 0.826536 +expl/Actions Max 2.12442 +expl/Actions Min -2.50163 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 641699 +eval/num paths total 704 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0584039 +eval/Actions Std 0.726238 +eval/Actions Max 0.999888 +eval/Actions Min -0.999608 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.35209e-05 +time/evaluation sampling (s) 4.87647 +time/exploration sampling (s) 6.46088 +time/logging (s) 0.00764698 +time/saving (s) 0.0107234 +time/training (s) 19.2872 +time/epoch (s) 30.6429 +time/total (s) 17767.3 +Epoch -301 +------------------------------ ---------------- +2022-05-15 22:59:07.102251 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -300 finished +------------------------------ ---------------- +epoch -300 +replay_buffer/size 999047 +trainer/num train calls 701000 +trainer/QF1 Loss 1.38692 +trainer/QF2 Loss 1.38112 +trainer/Policy Loss 13.5691 +trainer/Q1 Predictions Mean -72.8268 +trainer/Q1 Predictions Std 19.0121 +trainer/Q1 Predictions Max -0.51663 +trainer/Q1 Predictions Min -87.6488 +trainer/Q2 Predictions Mean -72.7762 +trainer/Q2 Predictions Std 18.9992 +trainer/Q2 Predictions Max 0.288719 +trainer/Q2 Predictions Min -87.4227 +trainer/Q Targets Mean -72.7736 +trainer/Q Targets Std 18.7924 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3257 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0206575 +trainer/policy/mean Std 0.728246 +trainer/policy/mean Max 0.999286 +trainer/policy/mean Min -0.999617 +trainer/policy/std Mean 0.412953 +trainer/policy/std Std 0.0206218 +trainer/policy/std Max 0.437093 +trainer/policy/std Min 0.384076 +trainer/Advantage Weights Mean 3.53707 +trainer/Advantage Weights Std 16.7952 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.21025e-19 +trainer/Advantage Score Mean -0.420622 +trainer/Advantage Score Std 0.578914 +trainer/Advantage Score Max 1.26315 +trainer/Advantage Score Min -4.2956 +trainer/V1 Predictions Mean -72.5644 +trainer/V1 Predictions Std 18.9276 +trainer/V1 Predictions Max 0.465361 +trainer/V1 Predictions Min -87.1984 +trainer/VF Loss 0.0702834 +expl/num steps total 701000 +expl/num paths total 926 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.173279 +expl/Actions Std 0.831064 +expl/Actions Max 2.44015 +expl/Actions Min -2.3476 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 642699 +eval/num paths total 705 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0330285 +eval/Actions Std 0.721162 +eval/Actions Max 0.999826 +eval/Actions Min -0.999833 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.39192e-06 +time/evaluation sampling (s) 4.78453 +time/exploration sampling (s) 6.6738 +time/logging (s) 0.0106252 +time/saving (s) 0.0310129 +time/training (s) 18.9183 +time/epoch (s) 30.4183 +time/total (s) 17797.7 +Epoch -300 +------------------------------ ---------------- +2022-05-15 22:59:36.272086 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -299 finished +------------------------------ ---------------- +epoch -299 +replay_buffer/size 999047 +trainer/num train calls 702000 +trainer/QF1 Loss 0.8753 +trainer/QF2 Loss 1.11984 +trainer/Policy Loss 21.7378 +trainer/Q1 Predictions Mean -75.5228 +trainer/Q1 Predictions Std 15.5261 +trainer/Q1 Predictions Max -0.227092 +trainer/Q1 Predictions Min -86.6933 +trainer/Q2 Predictions Mean -75.584 +trainer/Q2 Predictions Std 15.4936 +trainer/Q2 Predictions Max 0.157855 +trainer/Q2 Predictions Min -86.8931 +trainer/Q Targets Mean -75.9219 +trainer/Q Targets Std 15.6958 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3606 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0433026 +trainer/policy/mean Std 0.726234 +trainer/policy/mean Max 0.99929 +trainer/policy/mean Min -0.999033 +trainer/policy/std Mean 0.41162 +trainer/policy/std Std 0.0199534 +trainer/policy/std Max 0.433176 +trainer/policy/std Min 0.382923 +trainer/Advantage Weights Mean 6.12616 +trainer/Advantage Weights Std 21.0574 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.1898e-17 +trainer/Advantage Score Mean -0.280352 +trainer/Advantage Score Std 0.587272 +trainer/Advantage Score Max 1.52225 +trainer/Advantage Score Min -3.7984 +trainer/V1 Predictions Mean -75.7126 +trainer/V1 Predictions Std 15.7226 +trainer/V1 Predictions Max 0.086151 +trainer/V1 Predictions Min -87.2249 +trainer/VF Loss 0.0688088 +expl/num steps total 702000 +expl/num paths total 927 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0351174 +expl/Actions Std 0.836943 +expl/Actions Max 2.65472 +expl/Actions Min -2.19336 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 643619 +eval/num paths total 706 +eval/path length Mean 920 +eval/path length Std 0 +eval/path length Max 920 +eval/path length Min 920 +eval/Rewards Mean 0.00108696 +eval/Rewards Std 0.0329511 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0191799 +eval/Actions Std 0.731002 +eval/Actions Max 0.999964 +eval/Actions Min -0.999776 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.26127e-06 +time/evaluation sampling (s) 4.48588 +time/exploration sampling (s) 6.09105 +time/logging (s) 0.0109694 +time/saving (s) 0.0146775 +time/training (s) 18.5489 +time/epoch (s) 29.1515 +time/total (s) 17826.9 +Epoch -299 +------------------------------ ---------------- +2022-05-15 23:00:06.973447 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -298 finished +------------------------------ ---------------- +epoch -298 +replay_buffer/size 999047 +trainer/num train calls 703000 +trainer/QF1 Loss 0.603746 +trainer/QF2 Loss 0.767251 +trainer/Policy Loss 14.9377 +trainer/Q1 Predictions Mean -71.9974 +trainer/Q1 Predictions Std 20.9141 +trainer/Q1 Predictions Max -1.95128 +trainer/Q1 Predictions Min -87.6912 +trainer/Q2 Predictions Mean -72.0567 +trainer/Q2 Predictions Std 21.1173 +trainer/Q2 Predictions Max -1.21813 +trainer/Q2 Predictions Min -87.4345 +trainer/Q Targets Mean -72.0782 +trainer/Q Targets Std 20.7709 +trainer/Q Targets Max -3.13955 +trainer/Q Targets Min -87.1974 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0120989 +trainer/policy/mean Std 0.729469 +trainer/policy/mean Max 0.999298 +trainer/policy/mean Min -0.999708 +trainer/policy/std Mean 0.410624 +trainer/policy/std Std 0.0197584 +trainer/policy/std Max 0.432142 +trainer/policy/std Min 0.37973 +trainer/Advantage Weights Mean 5.54794 +trainer/Advantage Weights Std 20.4343 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.90421e-13 +trainer/Advantage Score Mean -0.258429 +trainer/Advantage Score Std 0.483489 +trainer/Advantage Score Max 1.28789 +trainer/Advantage Score Min -2.78662 +trainer/V1 Predictions Mean -71.7618 +trainer/V1 Predictions Std 20.979 +trainer/V1 Predictions Max -2.29827 +trainer/V1 Predictions Min -87.0624 +trainer/VF Loss 0.0553312 +expl/num steps total 703000 +expl/num paths total 929 +expl/path length Mean 500 +expl/path length Std 284 +expl/path length Max 784 +expl/path length Min 216 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0190514 +expl/Actions Std 0.824437 +expl/Actions Max 2.32717 +expl/Actions Min -2.17875 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 644619 +eval/num paths total 707 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.389807 +eval/Actions Std 0.69784 +eval/Actions Max 0.999759 +eval/Actions Min -0.999441 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.12867e-05 +time/evaluation sampling (s) 4.95662 +time/exploration sampling (s) 6.68612 +time/logging (s) 0.011722 +time/saving (s) 0.0162301 +time/training (s) 19.0093 +time/epoch (s) 30.68 +time/total (s) 17857.6 +Epoch -298 +------------------------------ ---------------- +2022-05-15 23:00:37.557023 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -297 finished +------------------------------ ---------------- +epoch -297 +replay_buffer/size 999047 +trainer/num train calls 704000 +trainer/QF1 Loss 0.508632 +trainer/QF2 Loss 0.565913 +trainer/Policy Loss 10.8093 +trainer/Q1 Predictions Mean -74.3417 +trainer/Q1 Predictions Std 15.9897 +trainer/Q1 Predictions Max -0.238785 +trainer/Q1 Predictions Min -87.0253 +trainer/Q2 Predictions Mean -74.2524 +trainer/Q2 Predictions Std 15.9792 +trainer/Q2 Predictions Max 0.119042 +trainer/Q2 Predictions Min -86.9841 +trainer/Q Targets Mean -74.6011 +trainer/Q Targets Std 15.8645 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.8841 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0115477 +trainer/policy/mean Std 0.723578 +trainer/policy/mean Max 0.998376 +trainer/policy/mean Min -0.998913 +trainer/policy/std Mean 0.411233 +trainer/policy/std Std 0.0196223 +trainer/policy/std Max 0.434911 +trainer/policy/std Min 0.379303 +trainer/Advantage Weights Mean 3.57051 +trainer/Advantage Weights Std 14.4265 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.01636e-08 +trainer/Advantage Score Mean -0.289941 +trainer/Advantage Score Std 0.405715 +trainer/Advantage Score Max 1.0716 +trainer/Advantage Score Min -1.6808 +trainer/V1 Predictions Mean -74.3956 +trainer/V1 Predictions Std 15.903 +trainer/V1 Predictions Max 0.664583 +trainer/V1 Predictions Min -86.7285 +trainer/VF Loss 0.0371877 +expl/num steps total 704000 +expl/num paths total 930 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.125668 +expl/Actions Std 0.880226 +expl/Actions Max 2.55544 +expl/Actions Min -2.36409 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 645619 +eval/num paths total 708 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.130843 +eval/Actions Std 0.742131 +eval/Actions Max 0.999818 +eval/Actions Min -0.999811 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.03381e-05 +time/evaluation sampling (s) 4.71258 +time/exploration sampling (s) 6.32839 +time/logging (s) 0.0117977 +time/saving (s) 0.0157854 +time/training (s) 19.4988 +time/epoch (s) 30.5674 +time/total (s) 17888.1 +Epoch -297 +------------------------------ ---------------- +2022-05-15 23:01:08.940277 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -296 finished +------------------------------ ---------------- +epoch -296 +replay_buffer/size 999047 +trainer/num train calls 705000 +trainer/QF1 Loss 0.576732 +trainer/QF2 Loss 0.526455 +trainer/Policy Loss 24.3066 +trainer/Q1 Predictions Mean -73.5072 +trainer/Q1 Predictions Std 18.8332 +trainer/Q1 Predictions Max -1.38716 +trainer/Q1 Predictions Min -86.6987 +trainer/Q2 Predictions Mean -73.5602 +trainer/Q2 Predictions Std 18.7558 +trainer/Q2 Predictions Max -1.80399 +trainer/Q2 Predictions Min -86.6054 +trainer/Q Targets Mean -73.7419 +trainer/Q Targets Std 18.804 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.7094 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0279339 +trainer/policy/mean Std 0.730406 +trainer/policy/mean Max 0.999116 +trainer/policy/mean Min -0.999627 +trainer/policy/std Mean 0.414395 +trainer/policy/std Std 0.0193031 +trainer/policy/std Max 0.435839 +trainer/policy/std Min 0.381149 +trainer/Advantage Weights Mean 3.34329 +trainer/Advantage Weights Std 14.7975 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.7015e-18 +trainer/Advantage Score Mean -0.421333 +trainer/Advantage Score Std 0.656548 +trainer/Advantage Score Max 1.08566 +trainer/Advantage Score Min -3.91743 +trainer/V1 Predictions Mean -73.4763 +trainer/V1 Predictions Std 18.8853 +trainer/V1 Predictions Max -2.24305 +trainer/V1 Predictions Min -86.8722 +trainer/VF Loss 0.0747192 +expl/num steps total 705000 +expl/num paths total 932 +expl/path length Mean 500 +expl/path length Std 111 +expl/path length Max 611 +expl/path length Min 389 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.013033 +expl/Actions Std 0.82688 +expl/Actions Max 2.51905 +expl/Actions Min -2.31226 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 646619 +eval/num paths total 709 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0553972 +eval/Actions Std 0.722736 +eval/Actions Max 0.999671 +eval/Actions Min -0.999691 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.33298e-06 +time/evaluation sampling (s) 4.65794 +time/exploration sampling (s) 7.48946 +time/logging (s) 0.0127753 +time/saving (s) 0.0175945 +time/training (s) 19.1804 +time/epoch (s) 31.3582 +time/total (s) 17919.5 +Epoch -296 +------------------------------ ---------------- +2022-05-15 23:01:38.441642 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -295 finished +------------------------------ ---------------- +epoch -295 +replay_buffer/size 999047 +trainer/num train calls 706000 +trainer/QF1 Loss 0.659515 +trainer/QF2 Loss 0.724592 +trainer/Policy Loss 4.19781 +trainer/Q1 Predictions Mean -75.5562 +trainer/Q1 Predictions Std 15.9036 +trainer/Q1 Predictions Max -0.633918 +trainer/Q1 Predictions Min -87.4857 +trainer/Q2 Predictions Mean -75.5538 +trainer/Q2 Predictions Std 15.9861 +trainer/Q2 Predictions Max -0.30974 +trainer/Q2 Predictions Min -87.2295 +trainer/Q Targets Mean -75.4261 +trainer/Q Targets Std 16.0943 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1134 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00810156 +trainer/policy/mean Std 0.723792 +trainer/policy/mean Max 0.998888 +trainer/policy/mean Min -0.999177 +trainer/policy/std Mean 0.413017 +trainer/policy/std Std 0.018698 +trainer/policy/std Max 0.434017 +trainer/policy/std Min 0.383437 +trainer/Advantage Weights Mean 0.881177 +trainer/Advantage Weights Std 6.85748 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.04292e-15 +trainer/Advantage Score Mean -0.458511 +trainer/Advantage Score Std 0.496211 +trainer/Advantage Score Max 1.17383 +trainer/Advantage Score Min -3.29208 +trainer/V1 Predictions Mean -75.1365 +trainer/V1 Predictions Std 16.2948 +trainer/V1 Predictions Max 0.279047 +trainer/V1 Predictions Min -87.2216 +trainer/VF Loss 0.0511148 +expl/num steps total 706000 +expl/num paths total 934 +expl/path length Mean 500 +expl/path length Std 143 +expl/path length Max 643 +expl/path length Min 357 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0236184 +expl/Actions Std 0.835944 +expl/Actions Max 2.35092 +expl/Actions Min -2.28225 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 647619 +eval/num paths total 710 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0139902 +eval/Actions Std 0.730069 +eval/Actions Max 0.999739 +eval/Actions Min -0.999657 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.79165e-06 +time/evaluation sampling (s) 4.33986 +time/exploration sampling (s) 6.37411 +time/logging (s) 0.0124815 +time/saving (s) 0.0172367 +time/training (s) 18.7379 +time/epoch (s) 29.4816 +time/total (s) 17949 +Epoch -295 +------------------------------ ---------------- +2022-05-15 23:02:09.107249 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -294 finished +------------------------------ ---------------- +epoch -294 +replay_buffer/size 999047 +trainer/num train calls 707000 +trainer/QF1 Loss 0.712156 +trainer/QF2 Loss 0.672859 +trainer/Policy Loss 10.1056 +trainer/Q1 Predictions Mean -73.5044 +trainer/Q1 Predictions Std 17.3999 +trainer/Q1 Predictions Max -0.225799 +trainer/Q1 Predictions Min -87.0745 +trainer/Q2 Predictions Mean -73.4451 +trainer/Q2 Predictions Std 17.3272 +trainer/Q2 Predictions Max -0.0636876 +trainer/Q2 Predictions Min -86.9989 +trainer/Q Targets Mean -73.4965 +trainer/Q Targets Std 17.4811 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0119 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00488769 +trainer/policy/mean Std 0.721214 +trainer/policy/mean Max 0.998926 +trainer/policy/mean Min -0.999418 +trainer/policy/std Mean 0.412357 +trainer/policy/std Std 0.0187798 +trainer/policy/std Max 0.432324 +trainer/policy/std Min 0.380993 +trainer/Advantage Weights Mean 3.69212 +trainer/Advantage Weights Std 17.6167 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.51769e-15 +trainer/Advantage Score Mean -0.468891 +trainer/Advantage Score Std 0.655824 +trainer/Advantage Score Max 3.20875 +trainer/Advantage Score Min -3.41216 +trainer/V1 Predictions Mean -73.2383 +trainer/V1 Predictions Std 17.5624 +trainer/V1 Predictions Max 0.341088 +trainer/V1 Predictions Min -86.8911 +trainer/VF Loss 0.137371 +expl/num steps total 707000 +expl/num paths total 935 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00236932 +expl/Actions Std 0.786506 +expl/Actions Max 2.29973 +expl/Actions Min -2.30512 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 648619 +eval/num paths total 711 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0749762 +eval/Actions Std 0.764567 +eval/Actions Max 0.998347 +eval/Actions Min -0.997862 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.36499e-05 +time/evaluation sampling (s) 5.07473 +time/exploration sampling (s) 6.39784 +time/logging (s) 0.010996 +time/saving (s) 0.0160612 +time/training (s) 19.1448 +time/epoch (s) 30.6444 +time/total (s) 17979.7 +Epoch -294 +------------------------------ ---------------- +2022-05-15 23:02:39.686785 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -293 finished +------------------------------ ---------------- +epoch -293 +replay_buffer/size 999047 +trainer/num train calls 708000 +trainer/QF1 Loss 0.668725 +trainer/QF2 Loss 0.700801 +trainer/Policy Loss 19.329 +trainer/Q1 Predictions Mean -72.7062 +trainer/Q1 Predictions Std 19.6812 +trainer/Q1 Predictions Max -0.751579 +trainer/Q1 Predictions Min -87.5419 +trainer/Q2 Predictions Mean -72.7821 +trainer/Q2 Predictions Std 19.584 +trainer/Q2 Predictions Max -0.342141 +trainer/Q2 Predictions Min -87.5798 +trainer/Q Targets Mean -72.6651 +trainer/Q Targets Std 19.6001 +trainer/Q Targets Max -1.74962 +trainer/Q Targets Min -87.3315 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0198533 +trainer/policy/mean Std 0.723832 +trainer/policy/mean Max 0.999625 +trainer/policy/mean Min -0.999636 +trainer/policy/std Mean 0.41172 +trainer/policy/std Std 0.0200077 +trainer/policy/std Max 0.43172 +trainer/policy/std Min 0.38154 +trainer/Advantage Weights Mean 4.28042 +trainer/Advantage Weights Std 16.2562 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.5029e-17 +trainer/Advantage Score Mean -0.404448 +trainer/Advantage Score Std 0.669498 +trainer/Advantage Score Max 2.11676 +trainer/Advantage Score Min -3.72717 +trainer/V1 Predictions Mean -72.3966 +trainer/V1 Predictions Std 19.7771 +trainer/V1 Predictions Max -0.808308 +trainer/V1 Predictions Min -87.4814 +trainer/VF Loss 0.0882186 +expl/num steps total 708000 +expl/num paths total 936 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.17201 +expl/Actions Std 0.805078 +expl/Actions Max 2.12041 +expl/Actions Min -2.51474 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 649242 +eval/num paths total 712 +eval/path length Mean 623 +eval/path length Std 0 +eval/path length Max 623 +eval/path length Min 623 +eval/Rewards Mean 0.00160514 +eval/Rewards Std 0.040032 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00199911 +eval/Actions Std 0.734785 +eval/Actions Max 0.999748 +eval/Actions Min -0.999699 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.35377e-05 +time/evaluation sampling (s) 4.45585 +time/exploration sampling (s) 7.26132 +time/logging (s) 0.0123427 +time/saving (s) 0.0169961 +time/training (s) 18.8168 +time/epoch (s) 30.5633 +time/total (s) 18010.2 +Epoch -293 +------------------------------ ---------------- +2022-05-15 23:03:09.893105 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -292 finished +------------------------------ ---------------- +epoch -292 +replay_buffer/size 999047 +trainer/num train calls 709000 +trainer/QF1 Loss 0.940183 +trainer/QF2 Loss 0.956021 +trainer/Policy Loss 23.2032 +trainer/Q1 Predictions Mean -73.0469 +trainer/Q1 Predictions Std 18.9136 +trainer/Q1 Predictions Max -1.00769 +trainer/Q1 Predictions Min -86.9244 +trainer/Q2 Predictions Mean -73.0302 +trainer/Q2 Predictions Std 18.8186 +trainer/Q2 Predictions Max -1.08695 +trainer/Q2 Predictions Min -87.0073 +trainer/Q Targets Mean -73.092 +trainer/Q Targets Std 18.9115 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0738 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0211087 +trainer/policy/mean Std 0.728112 +trainer/policy/mean Max 0.999116 +trainer/policy/mean Min -0.998958 +trainer/policy/std Mean 0.412751 +trainer/policy/std Std 0.0193358 +trainer/policy/std Max 0.433417 +trainer/policy/std Min 0.38395 +trainer/Advantage Weights Mean 5.07376 +trainer/Advantage Weights Std 18.7692 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.07788e-18 +trainer/Advantage Score Mean -0.303132 +trainer/Advantage Score Std 0.60158 +trainer/Advantage Score Max 2.00401 +trainer/Advantage Score Min -4.13715 +trainer/V1 Predictions Mean -72.8424 +trainer/V1 Predictions Std 19.063 +trainer/V1 Predictions Max -1.5007 +trainer/V1 Predictions Min -86.9377 +trainer/VF Loss 0.077595 +expl/num steps total 709000 +expl/num paths total 938 +expl/path length Mean 500 +expl/path length Std 276 +expl/path length Max 776 +expl/path length Min 224 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.042944 +expl/Actions Std 0.813682 +expl/Actions Max 2.48161 +expl/Actions Min -2.58065 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 650242 +eval/num paths total 713 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.258145 +eval/Actions Std 0.665496 +eval/Actions Max 0.999572 +eval/Actions Min -0.999749 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.32374e-05 +time/evaluation sampling (s) 5.04864 +time/exploration sampling (s) 6.46858 +time/logging (s) 0.013025 +time/saving (s) 0.0174185 +time/training (s) 18.6457 +time/epoch (s) 30.1933 +time/total (s) 18040.4 +Epoch -292 +------------------------------ ---------------- +2022-05-15 23:03:40.992214 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -291 finished +------------------------------ ---------------- +epoch -291 +replay_buffer/size 999047 +trainer/num train calls 710000 +trainer/QF1 Loss 0.938531 +trainer/QF2 Loss 0.878063 +trainer/Policy Loss 26.3922 +trainer/Q1 Predictions Mean -72.9001 +trainer/Q1 Predictions Std 19.5609 +trainer/Q1 Predictions Max -0.966171 +trainer/Q1 Predictions Min -87.152 +trainer/Q2 Predictions Mean -72.9445 +trainer/Q2 Predictions Std 19.6166 +trainer/Q2 Predictions Max -0.0717744 +trainer/Q2 Predictions Min -86.9395 +trainer/Q Targets Mean -73.2856 +trainer/Q Targets Std 19.6171 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4146 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0164649 +trainer/policy/mean Std 0.733632 +trainer/policy/mean Max 0.999111 +trainer/policy/mean Min -0.99875 +trainer/policy/std Mean 0.411117 +trainer/policy/std Std 0.0197533 +trainer/policy/std Max 0.433074 +trainer/policy/std Min 0.384252 +trainer/Advantage Weights Mean 7.39834 +trainer/Advantage Weights Std 21.1746 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.57844e-15 +trainer/Advantage Score Mean -0.211838 +trainer/Advantage Score Std 0.636843 +trainer/Advantage Score Max 2.13831 +trainer/Advantage Score Min -3.30174 +trainer/V1 Predictions Mean -72.9665 +trainer/V1 Predictions Std 19.9704 +trainer/V1 Predictions Max 1.29268 +trainer/V1 Predictions Min -87.292 +trainer/VF Loss 0.0806657 +expl/num steps total 710000 +expl/num paths total 940 +expl/path length Mean 500 +expl/path length Std 462 +expl/path length Max 962 +expl/path length Min 38 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0473261 +expl/Actions Std 0.820516 +expl/Actions Max 2.66441 +expl/Actions Min -2.19112 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 650718 +eval/num paths total 714 +eval/path length Mean 476 +eval/path length Std 0 +eval/path length Max 476 +eval/path length Min 476 +eval/Rewards Mean 0.00210084 +eval/Rewards Std 0.0457868 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0512125 +eval/Actions Std 0.744873 +eval/Actions Max 0.999814 +eval/Actions Min -0.998414 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.37012e-05 +time/evaluation sampling (s) 4.6951 +time/exploration sampling (s) 6.79385 +time/logging (s) 0.0102036 +time/saving (s) 0.0183379 +time/training (s) 19.5574 +time/epoch (s) 31.0749 +time/total (s) 18071.5 +Epoch -291 +------------------------------ ---------------- +2022-05-15 23:04:11.652772 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -290 finished +------------------------------ ---------------- +epoch -290 +replay_buffer/size 999047 +trainer/num train calls 711000 +trainer/QF1 Loss 0.868519 +trainer/QF2 Loss 0.830604 +trainer/Policy Loss 7.93996 +trainer/Q1 Predictions Mean -72.3857 +trainer/Q1 Predictions Std 21.0221 +trainer/Q1 Predictions Max -0.225427 +trainer/Q1 Predictions Min -87.669 +trainer/Q2 Predictions Mean -72.3857 +trainer/Q2 Predictions Std 20.9713 +trainer/Q2 Predictions Max 0.28554 +trainer/Q2 Predictions Min -87.5055 +trainer/Q Targets Mean -72.2081 +trainer/Q Targets Std 21.0204 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2982 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0235272 +trainer/policy/mean Std 0.735279 +trainer/policy/mean Max 0.999508 +trainer/policy/mean Min -0.998504 +trainer/policy/std Mean 0.410846 +trainer/policy/std Std 0.0196285 +trainer/policy/std Max 0.433175 +trainer/policy/std Min 0.382273 +trainer/Advantage Weights Mean 1.98551 +trainer/Advantage Weights Std 11.3101 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.28489e-18 +trainer/Advantage Score Mean -0.518363 +trainer/Advantage Score Std 0.718759 +trainer/Advantage Score Max 0.516914 +trainer/Advantage Score Min -3.96084 +trainer/V1 Predictions Mean -71.8582 +trainer/V1 Predictions Std 21.1943 +trainer/V1 Predictions Max 0.659009 +trainer/V1 Predictions Min -87.0744 +trainer/VF Loss 0.0828932 +expl/num steps total 711000 +expl/num paths total 941 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.076549 +expl/Actions Std 0.85825 +expl/Actions Max 2.21602 +expl/Actions Min -2.29095 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 651718 +eval/num paths total 715 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0140704 +eval/Actions Std 0.750659 +eval/Actions Max 0.999969 +eval/Actions Min -0.999823 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.98121e-06 +time/evaluation sampling (s) 4.88339 +time/exploration sampling (s) 6.4795 +time/logging (s) 0.0126229 +time/saving (s) 0.0181565 +time/training (s) 19.2498 +time/epoch (s) 30.6435 +time/total (s) 18102.2 +Epoch -290 +------------------------------ ---------------- +2022-05-15 23:04:42.820208 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -289 finished +------------------------------ ---------------- +epoch -289 +replay_buffer/size 999047 +trainer/num train calls 712000 +trainer/QF1 Loss 1.04165 +trainer/QF2 Loss 0.859334 +trainer/Policy Loss 24.7318 +trainer/Q1 Predictions Mean -73.606 +trainer/Q1 Predictions Std 17.9943 +trainer/Q1 Predictions Max -0.581662 +trainer/Q1 Predictions Min -86.9362 +trainer/Q2 Predictions Mean -73.6008 +trainer/Q2 Predictions Std 17.9217 +trainer/Q2 Predictions Max -0.585988 +trainer/Q2 Predictions Min -86.6483 +trainer/Q Targets Mean -73.3631 +trainer/Q Targets Std 17.8203 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.5841 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00390992 +trainer/policy/mean Std 0.739759 +trainer/policy/mean Max 0.99913 +trainer/policy/mean Min -0.999425 +trainer/policy/std Mean 0.41177 +trainer/policy/std Std 0.0199427 +trainer/policy/std Max 0.437148 +trainer/policy/std Min 0.381217 +trainer/Advantage Weights Mean 2.74086 +trainer/Advantage Weights Std 14.1176 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.85193e-21 +trainer/Advantage Score Mean -0.450638 +trainer/Advantage Score Std 0.583371 +trainer/Advantage Score Max 1.22192 +trainer/Advantage Score Min -4.73063 +trainer/V1 Predictions Mean -73.0539 +trainer/V1 Predictions Std 18.0405 +trainer/V1 Predictions Max -0.649916 +trainer/V1 Predictions Min -86.4398 +trainer/VF Loss 0.0658792 +expl/num steps total 712000 +expl/num paths total 942 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.015219 +expl/Actions Std 0.833377 +expl/Actions Max 2.18644 +expl/Actions Min -2.39117 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 652718 +eval/num paths total 716 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.125964 +eval/Actions Std 0.790785 +eval/Actions Max 0.999857 +eval/Actions Min -0.999449 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.17884e-06 +time/evaluation sampling (s) 4.82302 +time/exploration sampling (s) 7.19312 +time/logging (s) 0.00808242 +time/saving (s) 0.0173155 +time/training (s) 19.1017 +time/epoch (s) 31.1432 +time/total (s) 18133.3 +Epoch -289 +------------------------------ ---------------- +2022-05-15 23:05:14.324486 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -288 finished +------------------------------ ---------------- +epoch -288 +replay_buffer/size 999047 +trainer/num train calls 713000 +trainer/QF1 Loss 0.722959 +trainer/QF2 Loss 0.662844 +trainer/Policy Loss 18.6059 +trainer/Q1 Predictions Mean -73.4222 +trainer/Q1 Predictions Std 17.6097 +trainer/Q1 Predictions Max -2.34884 +trainer/Q1 Predictions Min -86.931 +trainer/Q2 Predictions Mean -73.4527 +trainer/Q2 Predictions Std 17.625 +trainer/Q2 Predictions Max -3.64769 +trainer/Q2 Predictions Min -87.1082 +trainer/Q Targets Mean -73.3455 +trainer/Q Targets Std 17.5521 +trainer/Q Targets Max -3.74071 +trainer/Q Targets Min -86.9687 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.03308 +trainer/policy/mean Std 0.731131 +trainer/policy/mean Max 0.998263 +trainer/policy/mean Min -0.999133 +trainer/policy/std Mean 0.410857 +trainer/policy/std Std 0.0198106 +trainer/policy/std Max 0.434408 +trainer/policy/std Min 0.381534 +trainer/Advantage Weights Mean 5.88284 +trainer/Advantage Weights Std 21.0138 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.48744e-16 +trainer/Advantage Score Mean -0.345865 +trainer/Advantage Score Std 0.607641 +trainer/Advantage Score Max 1.74431 +trainer/Advantage Score Min -3.55922 +trainer/V1 Predictions Mean -73.1125 +trainer/V1 Predictions Std 17.6984 +trainer/V1 Predictions Max -1.79968 +trainer/V1 Predictions Min -86.8468 +trainer/VF Loss 0.0798479 +expl/num steps total 713000 +expl/num paths total 943 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.115572 +expl/Actions Std 0.80412 +expl/Actions Max 2.36755 +expl/Actions Min -2.4159 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 653299 +eval/num paths total 717 +eval/path length Mean 581 +eval/path length Std 0 +eval/path length Max 581 +eval/path length Min 581 +eval/Rewards Mean 0.00172117 +eval/Rewards Std 0.0414513 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0119729 +eval/Actions Std 0.735395 +eval/Actions Max 0.998445 +eval/Actions Min -0.999028 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.30041e-05 +time/evaluation sampling (s) 5.35708 +time/exploration sampling (s) 7.29771 +time/logging (s) 0.00738104 +time/saving (s) 0.0133472 +time/training (s) 18.8139 +time/epoch (s) 31.4894 +time/total (s) 18164.8 +Epoch -288 +------------------------------ ---------------- +2022-05-15 23:05:45.332485 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -287 finished +------------------------------ ---------------- +epoch -287 +replay_buffer/size 999047 +trainer/num train calls 714000 +trainer/QF1 Loss 0.526281 +trainer/QF2 Loss 0.444549 +trainer/Policy Loss 7.13187 +trainer/Q1 Predictions Mean -75.0429 +trainer/Q1 Predictions Std 17.507 +trainer/Q1 Predictions Max -1.2509 +trainer/Q1 Predictions Min -87.859 +trainer/Q2 Predictions Mean -75.0681 +trainer/Q2 Predictions Std 17.5239 +trainer/Q2 Predictions Max -0.998656 +trainer/Q2 Predictions Min -87.6442 +trainer/Q Targets Mean -75.0023 +trainer/Q Targets Std 17.3532 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8232 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0138766 +trainer/policy/mean Std 0.731169 +trainer/policy/mean Max 0.998145 +trainer/policy/mean Min -0.998104 +trainer/policy/std Mean 0.411613 +trainer/policy/std Std 0.0201343 +trainer/policy/std Max 0.433617 +trainer/policy/std Min 0.381164 +trainer/Advantage Weights Mean 2.22815 +trainer/Advantage Weights Std 11.3972 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.8762e-15 +trainer/Advantage Score Mean -0.378981 +trainer/Advantage Score Std 0.537244 +trainer/Advantage Score Max 2.56193 +trainer/Advantage Score Min -3.39095 +trainer/V1 Predictions Mean -74.8191 +trainer/V1 Predictions Std 17.5171 +trainer/V1 Predictions Max -1.03485 +trainer/V1 Predictions Min -87.567 +trainer/VF Loss 0.073946 +expl/num steps total 714000 +expl/num paths total 945 +expl/path length Mean 500 +expl/path length Std 269 +expl/path length Max 769 +expl/path length Min 231 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0555367 +expl/Actions Std 0.817865 +expl/Actions Max 2.27943 +expl/Actions Min -2.42547 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 654299 +eval/num paths total 718 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.175486 +eval/Actions Std 0.777666 +eval/Actions Max 0.999431 +eval/Actions Min -0.999662 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.27861e-05 +time/evaluation sampling (s) 4.82297 +time/exploration sampling (s) 6.74104 +time/logging (s) 0.0102938 +time/saving (s) 0.0151702 +time/training (s) 19.4082 +time/epoch (s) 30.9977 +time/total (s) 18195.8 +Epoch -287 +------------------------------ ---------------- +2022-05-15 23:06:16.393656 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -286 finished +------------------------------ ---------------- +epoch -286 +replay_buffer/size 999047 +trainer/num train calls 715000 +trainer/QF1 Loss 0.807189 +trainer/QF2 Loss 0.78637 +trainer/Policy Loss 9.27443 +trainer/Q1 Predictions Mean -72.808 +trainer/Q1 Predictions Std 19.7113 +trainer/Q1 Predictions Max -1.27169 +trainer/Q1 Predictions Min -88.3516 +trainer/Q2 Predictions Mean -72.7934 +trainer/Q2 Predictions Std 19.5943 +trainer/Q2 Predictions Max -2.3926 +trainer/Q2 Predictions Min -88.1995 +trainer/Q Targets Mean -72.578 +trainer/Q Targets Std 19.5941 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8819 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0176159 +trainer/policy/mean Std 0.738263 +trainer/policy/mean Max 0.999864 +trainer/policy/mean Min -0.998668 +trainer/policy/std Mean 0.410889 +trainer/policy/std Std 0.0201593 +trainer/policy/std Max 0.435509 +trainer/policy/std Min 0.378085 +trainer/Advantage Weights Mean 2.50585 +trainer/Advantage Weights Std 14.2032 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.07706e-20 +trainer/Advantage Score Mean -0.594961 +trainer/Advantage Score Std 0.721754 +trainer/Advantage Score Max 1.87252 +trainer/Advantage Score Min -4.53207 +trainer/V1 Predictions Mean -72.2289 +trainer/V1 Predictions Std 19.9021 +trainer/V1 Predictions Max -1.85679 +trainer/V1 Predictions Min -88.4029 +trainer/VF Loss 0.105686 +expl/num steps total 715000 +expl/num paths total 947 +expl/path length Mean 500 +expl/path length Std 173 +expl/path length Max 673 +expl/path length Min 327 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0244246 +expl/Actions Std 0.834522 +expl/Actions Max 2.62447 +expl/Actions Min -2.50308 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 655299 +eval/num paths total 719 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.22023 +eval/Actions Std 0.764166 +eval/Actions Max 0.99965 +eval/Actions Min -0.999937 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.10879e-05 +time/evaluation sampling (s) 4.94406 +time/exploration sampling (s) 6.93397 +time/logging (s) 0.00949758 +time/saving (s) 0.0166217 +time/training (s) 19.1422 +time/epoch (s) 31.0463 +time/total (s) 18226.9 +Epoch -286 +------------------------------ ---------------- +2022-05-15 23:06:46.673307 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -285 finished +------------------------------ ---------------- +epoch -285 +replay_buffer/size 999047 +trainer/num train calls 716000 +trainer/QF1 Loss 0.87013 +trainer/QF2 Loss 0.897977 +trainer/Policy Loss 39.4757 +trainer/Q1 Predictions Mean -75.7349 +trainer/Q1 Predictions Std 17.2802 +trainer/Q1 Predictions Max -3.7402 +trainer/Q1 Predictions Min -87.1973 +trainer/Q2 Predictions Mean -75.7111 +trainer/Q2 Predictions Std 17.2798 +trainer/Q2 Predictions Max -2.37953 +trainer/Q2 Predictions Min -87.165 +trainer/Q Targets Mean -75.7386 +trainer/Q Targets Std 17.4898 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1055 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00870982 +trainer/policy/mean Std 0.735751 +trainer/policy/mean Max 0.999356 +trainer/policy/mean Min -0.999704 +trainer/policy/std Mean 0.411426 +trainer/policy/std Std 0.0196942 +trainer/policy/std Max 0.435843 +trainer/policy/std Min 0.381985 +trainer/Advantage Weights Mean 7.93106 +trainer/Advantage Weights Std 21.7956 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.59646e-18 +trainer/Advantage Score Mean -0.190305 +trainer/Advantage Score Std 0.658328 +trainer/Advantage Score Max 0.904549 +trainer/Advantage Score Min -4.09787 +trainer/V1 Predictions Mean -75.5574 +trainer/V1 Predictions Std 17.5241 +trainer/V1 Predictions Max -1.22918 +trainer/V1 Predictions Min -87.0561 +trainer/VF Loss 0.0729898 +expl/num steps total 716000 +expl/num paths total 949 +expl/path length Mean 500 +expl/path length Std 465 +expl/path length Max 965 +expl/path length Min 35 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0314064 +expl/Actions Std 0.812453 +expl/Actions Max 2.18137 +expl/Actions Min -2.31683 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 656299 +eval/num paths total 720 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0535879 +eval/Actions Std 0.692413 +eval/Actions Max 0.999527 +eval/Actions Min -0.999283 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.58721e-06 +time/evaluation sampling (s) 4.85902 +time/exploration sampling (s) 6.3508 +time/logging (s) 0.0116004 +time/saving (s) 0.0171593 +time/training (s) 19.0271 +time/epoch (s) 30.2656 +time/total (s) 18257.1 +Epoch -285 +------------------------------ ---------------- +2022-05-15 23:07:17.522033 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -284 finished +------------------------------ ---------------- +epoch -284 +replay_buffer/size 999047 +trainer/num train calls 717000 +trainer/QF1 Loss 5.60304 +trainer/QF2 Loss 5.61368 +trainer/Policy Loss 9.21658 +trainer/Q1 Predictions Mean -74.1468 +trainer/Q1 Predictions Std 18.4422 +trainer/Q1 Predictions Max -0.543927 +trainer/Q1 Predictions Min -87.914 +trainer/Q2 Predictions Mean -74.1235 +trainer/Q2 Predictions Std 18.4293 +trainer/Q2 Predictions Max -0.350594 +trainer/Q2 Predictions Min -87.8084 +trainer/Q Targets Mean -73.7605 +trainer/Q Targets Std 18.7853 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0558 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00582857 +trainer/policy/mean Std 0.723803 +trainer/policy/mean Max 0.999599 +trainer/policy/mean Min -0.999317 +trainer/policy/std Mean 0.41355 +trainer/policy/std Std 0.019703 +trainer/policy/std Max 0.435107 +trainer/policy/std Min 0.383376 +trainer/Advantage Weights Mean 1.97558 +trainer/Advantage Weights Std 12.5041 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.83166e-19 +trainer/Advantage Score Mean -0.609932 +trainer/Advantage Score Std 0.662766 +trainer/Advantage Score Max 0.88044 +trainer/Advantage Score Min -4.14635 +trainer/V1 Predictions Mean -73.6623 +trainer/V1 Predictions Std 18.7535 +trainer/V1 Predictions Max -0.136092 +trainer/V1 Predictions Min -87.8227 +trainer/VF Loss 0.0886805 +expl/num steps total 717000 +expl/num paths total 950 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0233967 +expl/Actions Std 0.847297 +expl/Actions Max 2.36762 +expl/Actions Min -2.2163 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 657299 +eval/num paths total 721 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0383288 +eval/Actions Std 0.72546 +eval/Actions Max 0.999584 +eval/Actions Min -0.999181 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.3974e-05 +time/evaluation sampling (s) 4.71238 +time/exploration sampling (s) 6.76449 +time/logging (s) 0.0119373 +time/saving (s) 0.0156805 +time/training (s) 19.3283 +time/epoch (s) 30.8328 +time/total (s) 18288 +Epoch -284 +------------------------------ ---------------- +2022-05-15 23:07:48.238520 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -283 finished +------------------------------ ---------------- +epoch -283 +replay_buffer/size 999047 +trainer/num train calls 718000 +trainer/QF1 Loss 1.3176 +trainer/QF2 Loss 1.36626 +trainer/Policy Loss 21.6437 +trainer/Q1 Predictions Mean -73.7675 +trainer/Q1 Predictions Std 18.6293 +trainer/Q1 Predictions Max -0.357654 +trainer/Q1 Predictions Min -87.2784 +trainer/Q2 Predictions Mean -73.7469 +trainer/Q2 Predictions Std 18.6382 +trainer/Q2 Predictions Max 0.0787367 +trainer/Q2 Predictions Min -87.4288 +trainer/Q Targets Mean -74.0958 +trainer/Q Targets Std 18.6716 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1098 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00627283 +trainer/policy/mean Std 0.72229 +trainer/policy/mean Max 0.99896 +trainer/policy/mean Min -0.998798 +trainer/policy/std Mean 0.41298 +trainer/policy/std Std 0.0193378 +trainer/policy/std Max 0.435392 +trainer/policy/std Min 0.383559 +trainer/Advantage Weights Mean 4.80207 +trainer/Advantage Weights Std 16.9601 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.51509e-15 +trainer/Advantage Score Mean -0.2623 +trainer/Advantage Score Std 0.567999 +trainer/Advantage Score Max 1.61994 +trainer/Advantage Score Min -3.36165 +trainer/V1 Predictions Mean -73.765 +trainer/V1 Predictions Std 18.8801 +trainer/V1 Predictions Max 1.29475 +trainer/V1 Predictions Min -87.9847 +trainer/VF Loss 0.0590214 +expl/num steps total 718000 +expl/num paths total 952 +expl/path length Mean 500 +expl/path length Std 442 +expl/path length Max 942 +expl/path length Min 58 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0234472 +expl/Actions Std 0.823947 +expl/Actions Max 2.37619 +expl/Actions Min -2.286 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 658299 +eval/num paths total 722 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.33311 +eval/Actions Std 0.696239 +eval/Actions Max 0.999895 +eval/Actions Min -0.997311 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.07042e-05 +time/evaluation sampling (s) 4.7915 +time/exploration sampling (s) 6.70945 +time/logging (s) 0.0109851 +time/saving (s) 0.0148088 +time/training (s) 19.1719 +time/epoch (s) 30.6987 +time/total (s) 18318.7 +Epoch -283 +------------------------------ ---------------- +2022-05-15 23:08:18.338224 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -282 finished +------------------------------ ---------------- +epoch -282 +replay_buffer/size 999047 +trainer/num train calls 719000 +trainer/QF1 Loss 2.02239 +trainer/QF2 Loss 1.8816 +trainer/Policy Loss 12.6778 +trainer/Q1 Predictions Mean -71.4594 +trainer/Q1 Predictions Std 20.7305 +trainer/Q1 Predictions Max -0.571754 +trainer/Q1 Predictions Min -87.3195 +trainer/Q2 Predictions Mean -71.4174 +trainer/Q2 Predictions Std 20.8159 +trainer/Q2 Predictions Max -0.635928 +trainer/Q2 Predictions Min -87.1817 +trainer/Q Targets Mean -71.156 +trainer/Q Targets Std 21.2985 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2103 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0158958 +trainer/policy/mean Std 0.717306 +trainer/policy/mean Max 0.999418 +trainer/policy/mean Min -0.998732 +trainer/policy/std Mean 0.41336 +trainer/policy/std Std 0.0198146 +trainer/policy/std Max 0.43738 +trainer/policy/std Min 0.382649 +trainer/Advantage Weights Mean 2.33367 +trainer/Advantage Weights Std 11.308 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.47278e-18 +trainer/Advantage Score Mean -0.561028 +trainer/Advantage Score Std 0.724165 +trainer/Advantage Score Max 0.687057 +trainer/Advantage Score Min -3.95789 +trainer/V1 Predictions Mean -70.9629 +trainer/V1 Predictions Std 21.1235 +trainer/V1 Predictions Max -0.225753 +trainer/V1 Predictions Min -87.1933 +trainer/VF Loss 0.0907259 +expl/num steps total 719000 +expl/num paths total 954 +expl/path length Mean 500 +expl/path length Std 205 +expl/path length Max 705 +expl/path length Min 295 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0353983 +expl/Actions Std 0.821804 +expl/Actions Max 2.39791 +expl/Actions Min -2.37199 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 659230 +eval/num paths total 723 +eval/path length Mean 931 +eval/path length Std 0 +eval/path length Max 931 +eval/path length Min 931 +eval/Rewards Mean 0.00107411 +eval/Rewards Std 0.0327561 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0566011 +eval/Actions Std 0.720448 +eval/Actions Max 0.999702 +eval/Actions Min -0.999181 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.25482e-05 +time/evaluation sampling (s) 4.53693 +time/exploration sampling (s) 6.13138 +time/logging (s) 0.00855913 +time/saving (s) 0.0155148 +time/training (s) 19.3865 +time/epoch (s) 30.0789 +time/total (s) 18348.8 +Epoch -282 +------------------------------ ---------------- +2022-05-15 23:08:49.209584 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -281 finished +------------------------------ ---------------- +epoch -281 +replay_buffer/size 999047 +trainer/num train calls 720000 +trainer/QF1 Loss 0.83276 +trainer/QF2 Loss 0.851425 +trainer/Policy Loss 6.10703 +trainer/Q1 Predictions Mean -73.9125 +trainer/Q1 Predictions Std 18.5808 +trainer/Q1 Predictions Max -0.812904 +trainer/Q1 Predictions Min -87.9474 +trainer/Q2 Predictions Mean -73.9337 +trainer/Q2 Predictions Std 18.6075 +trainer/Q2 Predictions Max -1.27748 +trainer/Q2 Predictions Min -88.0441 +trainer/Q Targets Mean -73.5915 +trainer/Q Targets Std 18.5982 +trainer/Q Targets Max -1.69305 +trainer/Q Targets Min -87.4411 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0037621 +trainer/policy/mean Std 0.724992 +trainer/policy/mean Max 0.999176 +trainer/policy/mean Min -0.999881 +trainer/policy/std Mean 0.41424 +trainer/policy/std Std 0.0192712 +trainer/policy/std Max 0.438898 +trainer/policy/std Min 0.385756 +trainer/Advantage Weights Mean 1.67377 +trainer/Advantage Weights Std 10.4539 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.08617e-20 +trainer/Advantage Score Mean -0.69876 +trainer/Advantage Score Std 0.606724 +trainer/Advantage Score Max 1.06381 +trainer/Advantage Score Min -4.5969 +trainer/V1 Predictions Mean -73.3394 +trainer/V1 Predictions Std 18.787 +trainer/V1 Predictions Max -0.81418 +trainer/V1 Predictions Min -86.9711 +trainer/VF Loss 0.0925521 +expl/num steps total 720000 +expl/num paths total 955 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0322124 +expl/Actions Std 0.801028 +expl/Actions Max 2.42965 +expl/Actions Min -2.32755 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 660230 +eval/num paths total 724 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00551625 +eval/Actions Std 0.793315 +eval/Actions Max 0.999741 +eval/Actions Min -0.999634 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.21497e-06 +time/evaluation sampling (s) 4.69497 +time/exploration sampling (s) 7.1071 +time/logging (s) 0.0115164 +time/saving (s) 0.0164975 +time/training (s) 19.0331 +time/epoch (s) 30.8632 +time/total (s) 18379.6 +Epoch -281 +------------------------------ ---------------- +2022-05-15 23:09:18.835967 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -280 finished +------------------------------ ---------------- +epoch -280 +replay_buffer/size 999047 +trainer/num train calls 721000 +trainer/QF1 Loss 0.692951 +trainer/QF2 Loss 0.657706 +trainer/Policy Loss 9.81442 +trainer/Q1 Predictions Mean -73.6502 +trainer/Q1 Predictions Std 18.4511 +trainer/Q1 Predictions Max -1.41108 +trainer/Q1 Predictions Min -86.9069 +trainer/Q2 Predictions Mean -73.5754 +trainer/Q2 Predictions Std 18.4961 +trainer/Q2 Predictions Max -0.0304581 +trainer/Q2 Predictions Min -86.8359 +trainer/Q Targets Mean -73.7732 +trainer/Q Targets Std 18.5975 +trainer/Q Targets Max -2.4693 +trainer/Q Targets Min -87.8295 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00452755 +trainer/policy/mean Std 0.723876 +trainer/policy/mean Max 0.999525 +trainer/policy/mean Min -0.999572 +trainer/policy/std Mean 0.412783 +trainer/policy/std Std 0.0186249 +trainer/policy/std Max 0.432425 +trainer/policy/std Min 0.384122 +trainer/Advantage Weights Mean 2.47965 +trainer/Advantage Weights Std 13.0747 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.22105e-10 +trainer/Advantage Score Mean -0.400823 +trainer/Advantage Score Std 0.474667 +trainer/Advantage Score Max 1.67154 +trainer/Advantage Score Min -2.10488 +trainer/V1 Predictions Mean -73.5413 +trainer/V1 Predictions Std 18.6452 +trainer/V1 Predictions Max -0.828205 +trainer/V1 Predictions Min -87.6958 +trainer/VF Loss 0.0544265 +expl/num steps total 721000 +expl/num paths total 957 +expl/path length Mean 500 +expl/path length Std 97 +expl/path length Max 597 +expl/path length Min 403 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0367341 +expl/Actions Std 0.827298 +expl/Actions Max 2.33478 +expl/Actions Min -2.68777 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 660716 +eval/num paths total 725 +eval/path length Mean 486 +eval/path length Std 0 +eval/path length Max 486 +eval/path length Min 486 +eval/Rewards Mean 0.00205761 +eval/Rewards Std 0.0453142 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00720094 +eval/Actions Std 0.752247 +eval/Actions Max 0.999778 +eval/Actions Min -0.999396 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.10189e-05 +time/evaluation sampling (s) 4.78794 +time/exploration sampling (s) 6.7288 +time/logging (s) 0.00961952 +time/saving (s) 0.0126178 +time/training (s) 18.0705 +time/epoch (s) 29.6095 +time/total (s) 18409.2 +Epoch -280 +------------------------------ ---------------- +2022-05-15 23:09:48.204259 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -279 finished +------------------------------ ---------------- +epoch -279 +replay_buffer/size 999047 +trainer/num train calls 722000 +trainer/QF1 Loss 0.735606 +trainer/QF2 Loss 0.666549 +trainer/Policy Loss 31.5105 +trainer/Q1 Predictions Mean -71.7054 +trainer/Q1 Predictions Std 20.1515 +trainer/Q1 Predictions Max -1.53365 +trainer/Q1 Predictions Min -86.9096 +trainer/Q2 Predictions Mean -71.7409 +trainer/Q2 Predictions Std 20.2513 +trainer/Q2 Predictions Max -1.17357 +trainer/Q2 Predictions Min -87.258 +trainer/Q Targets Mean -72.1617 +trainer/Q Targets Std 20.1355 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9838 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0167058 +trainer/policy/mean Std 0.730667 +trainer/policy/mean Max 0.999665 +trainer/policy/mean Min -0.999745 +trainer/policy/std Mean 0.41353 +trainer/policy/std Std 0.0185542 +trainer/policy/std Max 0.432546 +trainer/policy/std Min 0.38448 +trainer/Advantage Weights Mean 6.50013 +trainer/Advantage Weights Std 20.8459 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.83134e-18 +trainer/Advantage Score Mean -0.276504 +trainer/Advantage Score Std 0.662511 +trainer/Advantage Score Max 2.024 +trainer/Advantage Score Min -3.96833 +trainer/V1 Predictions Mean -71.8913 +trainer/V1 Predictions Std 20.2148 +trainer/V1 Predictions Max -1.28893 +trainer/V1 Predictions Min -87.7743 +trainer/VF Loss 0.0900757 +expl/num steps total 722000 +expl/num paths total 959 +expl/path length Mean 500 +expl/path length Std 284 +expl/path length Max 784 +expl/path length Min 216 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0217936 +expl/Actions Std 0.825355 +expl/Actions Max 2.40215 +expl/Actions Min -2.34303 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 661621 +eval/num paths total 726 +eval/path length Mean 905 +eval/path length Std 0 +eval/path length Max 905 +eval/path length Min 905 +eval/Rewards Mean 0.00110497 +eval/Rewards Std 0.0332228 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0102127 +eval/Actions Std 0.751773 +eval/Actions Max 0.999806 +eval/Actions Min -0.999882 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.49363e-06 +time/evaluation sampling (s) 4.71633 +time/exploration sampling (s) 6.6059 +time/logging (s) 0.0122137 +time/saving (s) 0.0177927 +time/training (s) 18.0006 +time/epoch (s) 29.3529 +time/total (s) 18438.6 +Epoch -279 +------------------------------ ---------------- +2022-05-15 23:10:18.846089 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -278 finished +------------------------------ ---------------- +epoch -278 +replay_buffer/size 999047 +trainer/num train calls 723000 +trainer/QF1 Loss 0.921157 +trainer/QF2 Loss 0.919114 +trainer/Policy Loss 19.7412 +trainer/Q1 Predictions Mean -72.5407 +trainer/Q1 Predictions Std 20.6576 +trainer/Q1 Predictions Max -0.749822 +trainer/Q1 Predictions Min -87.8467 +trainer/Q2 Predictions Mean -72.4319 +trainer/Q2 Predictions Std 20.6701 +trainer/Q2 Predictions Max 0.0812772 +trainer/Q2 Predictions Min -87.6652 +trainer/Q Targets Mean -72.42 +trainer/Q Targets Std 20.6444 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9507 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.016111 +trainer/policy/mean Std 0.736992 +trainer/policy/mean Max 0.999479 +trainer/policy/mean Min -0.998618 +trainer/policy/std Mean 0.411706 +trainer/policy/std Std 0.0184296 +trainer/policy/std Max 0.435257 +trainer/policy/std Min 0.381994 +trainer/Advantage Weights Mean 6.25195 +trainer/Advantage Weights Std 21.0613 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.68079e-16 +trainer/Advantage Score Mean -0.277899 +trainer/Advantage Score Std 0.572469 +trainer/Advantage Score Max 1.95193 +trainer/Advantage Score Min -3.63221 +trainer/V1 Predictions Mean -72.1728 +trainer/V1 Predictions Std 20.7116 +trainer/V1 Predictions Max 0.129654 +trainer/V1 Predictions Min -87.8649 +trainer/VF Loss 0.0742493 +expl/num steps total 723000 +expl/num paths total 960 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.172133 +expl/Actions Std 0.820078 +expl/Actions Max 2.47717 +expl/Actions Min -2.3099 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 662621 +eval/num paths total 727 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.400047 +eval/Actions Std 0.637181 +eval/Actions Max 0.999835 +eval/Actions Min -0.999666 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.19794e-06 +time/evaluation sampling (s) 5.31404 +time/exploration sampling (s) 6.3709 +time/logging (s) 0.0129827 +time/saving (s) 0.0195871 +time/training (s) 18.9063 +time/epoch (s) 30.6238 +time/total (s) 18469.2 +Epoch -278 +------------------------------ ---------------- +2022-05-15 23:10:49.804059 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -277 finished +------------------------------ ---------------- +epoch -277 +replay_buffer/size 999047 +trainer/num train calls 724000 +trainer/QF1 Loss 0.910523 +trainer/QF2 Loss 0.783036 +trainer/Policy Loss 14.8141 +trainer/Q1 Predictions Mean -74.665 +trainer/Q1 Predictions Std 17.3047 +trainer/Q1 Predictions Max -0.234746 +trainer/Q1 Predictions Min -87.4632 +trainer/Q2 Predictions Mean -74.6862 +trainer/Q2 Predictions Std 17.3618 +trainer/Q2 Predictions Max -0.1584 +trainer/Q2 Predictions Min -87.6909 +trainer/Q Targets Mean -74.6011 +trainer/Q Targets Std 17.4283 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3388 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.022317 +trainer/policy/mean Std 0.725211 +trainer/policy/mean Max 0.999724 +trainer/policy/mean Min -0.99971 +trainer/policy/std Mean 0.411526 +trainer/policy/std Std 0.0200679 +trainer/policy/std Max 0.436292 +trainer/policy/std Min 0.379814 +trainer/Advantage Weights Mean 3.18502 +trainer/Advantage Weights Std 14.7312 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.08794e-23 +trainer/Advantage Score Mean -0.467007 +trainer/Advantage Score Std 0.734804 +trainer/Advantage Score Max 0.920391 +trainer/Advantage Score Min -5.28752 +trainer/V1 Predictions Mean -74.2739 +trainer/V1 Predictions Std 17.8003 +trainer/V1 Predictions Max -0.223693 +trainer/V1 Predictions Min -87.4611 +trainer/VF Loss 0.0869844 +expl/num steps total 724000 +expl/num paths total 962 +expl/path length Mean 500 +expl/path length Std 327 +expl/path length Max 827 +expl/path length Min 173 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0558774 +expl/Actions Std 0.829984 +expl/Actions Max 2.29762 +expl/Actions Min -2.28046 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 663621 +eval/num paths total 728 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.234661 +eval/Actions Std 0.664417 +eval/Actions Max 0.999142 +eval/Actions Min -0.999182 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.31628e-05 +time/evaluation sampling (s) 4.58883 +time/exploration sampling (s) 6.77656 +time/logging (s) 0.0118084 +time/saving (s) 0.0150085 +time/training (s) 19.5441 +time/epoch (s) 30.9363 +time/total (s) 18500.2 +Epoch -277 +------------------------------ ---------------- +2022-05-15 23:11:20.150675 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -276 finished +------------------------------ ---------------- +epoch -276 +replay_buffer/size 999047 +trainer/num train calls 725000 +trainer/QF1 Loss 0.834542 +trainer/QF2 Loss 0.929662 +trainer/Policy Loss 11.0511 +trainer/Q1 Predictions Mean -72.2488 +trainer/Q1 Predictions Std 20.5772 +trainer/Q1 Predictions Max -0.26254 +trainer/Q1 Predictions Min -87.6918 +trainer/Q2 Predictions Mean -72.3422 +trainer/Q2 Predictions Std 20.6107 +trainer/Q2 Predictions Max 0.306137 +trainer/Q2 Predictions Min -87.4843 +trainer/Q Targets Mean -72.4616 +trainer/Q Targets Std 20.5751 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5932 +trainer/rewards Mean -0.976562 +trainer/rewards Std 0.151288 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0234375 +trainer/terminals Std 0.151288 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00384254 +trainer/policy/mean Std 0.726504 +trainer/policy/mean Max 0.999548 +trainer/policy/mean Min -0.99809 +trainer/policy/std Mean 0.411413 +trainer/policy/std Std 0.0196165 +trainer/policy/std Max 0.434754 +trainer/policy/std Min 0.380429 +trainer/Advantage Weights Mean 2.93932 +trainer/Advantage Weights Std 14.417 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.24134e-22 +trainer/Advantage Score Mean -0.530211 +trainer/Advantage Score Std 0.760608 +trainer/Advantage Score Max 1.86986 +trainer/Advantage Score Min -5.04407 +trainer/V1 Predictions Mean -72.1794 +trainer/V1 Predictions Std 20.5861 +trainer/V1 Predictions Max -0.273357 +trainer/V1 Predictions Min -87.4891 +trainer/VF Loss 0.113735 +expl/num steps total 725000 +expl/num paths total 963 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.025487 +expl/Actions Std 0.802958 +expl/Actions Max 2.39924 +expl/Actions Min -2.25256 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 664621 +eval/num paths total 729 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0208987 +eval/Actions Std 0.755181 +eval/Actions Max 0.999859 +eval/Actions Min -0.998655 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.04508e-05 +time/evaluation sampling (s) 4.78961 +time/exploration sampling (s) 6.31983 +time/logging (s) 0.0125447 +time/saving (s) 0.014743 +time/training (s) 19.1924 +time/epoch (s) 30.3291 +time/total (s) 18530.5 +Epoch -276 +------------------------------ ---------------- +2022-05-15 23:11:50.593976 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -275 finished +------------------------------ ---------------- +epoch -275 +replay_buffer/size 999047 +trainer/num train calls 726000 +trainer/QF1 Loss 0.476842 +trainer/QF2 Loss 0.518265 +trainer/Policy Loss 19.2868 +trainer/Q1 Predictions Mean -74.5091 +trainer/Q1 Predictions Std 18.2603 +trainer/Q1 Predictions Max -0.819982 +trainer/Q1 Predictions Min -87.5291 +trainer/Q2 Predictions Mean -74.4783 +trainer/Q2 Predictions Std 18.1934 +trainer/Q2 Predictions Max -1.4048 +trainer/Q2 Predictions Min -87.819 +trainer/Q Targets Mean -74.6844 +trainer/Q Targets Std 18.3519 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5334 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0322646 +trainer/policy/mean Std 0.723578 +trainer/policy/mean Max 0.999473 +trainer/policy/mean Min -0.997881 +trainer/policy/std Mean 0.413025 +trainer/policy/std Std 0.0194625 +trainer/policy/std Max 0.434315 +trainer/policy/std Min 0.383344 +trainer/Advantage Weights Mean 3.70813 +trainer/Advantage Weights Std 15.9552 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.60269e-19 +trainer/Advantage Score Mean -0.321394 +trainer/Advantage Score Std 0.544685 +trainer/Advantage Score Max 1.50864 +trainer/Advantage Score Min -4.32774 +trainer/V1 Predictions Mean -74.4588 +trainer/V1 Predictions Std 18.3807 +trainer/V1 Predictions Max 0.316469 +trainer/V1 Predictions Min -87.5057 +trainer/VF Loss 0.062551 +expl/num steps total 726000 +expl/num paths total 964 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0437903 +expl/Actions Std 0.8416 +expl/Actions Max 2.39527 +expl/Actions Min -2.30483 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 665621 +eval/num paths total 730 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0891741 +eval/Actions Std 0.715352 +eval/Actions Max 0.999 +eval/Actions Min -0.99894 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.7232e-06 +time/evaluation sampling (s) 4.72271 +time/exploration sampling (s) 6.62948 +time/logging (s) 0.0126201 +time/saving (s) 0.0182375 +time/training (s) 19.0432 +time/epoch (s) 30.4262 +time/total (s) 18560.9 +Epoch -275 +------------------------------ ---------------- +2022-05-15 23:12:21.010037 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -274 finished +------------------------------ ---------------- +epoch -274 +replay_buffer/size 999047 +trainer/num train calls 727000 +trainer/QF1 Loss 0.752228 +trainer/QF2 Loss 0.702312 +trainer/Policy Loss 15.5254 +trainer/Q1 Predictions Mean -74.3567 +trainer/Q1 Predictions Std 17.1836 +trainer/Q1 Predictions Max -1.20137 +trainer/Q1 Predictions Min -87.8723 +trainer/Q2 Predictions Mean -74.4113 +trainer/Q2 Predictions Std 17.0284 +trainer/Q2 Predictions Max -1.66444 +trainer/Q2 Predictions Min -88.1571 +trainer/Q Targets Mean -74.5338 +trainer/Q Targets Std 17.0155 +trainer/Q Targets Max -1.50002 +trainer/Q Targets Min -87.7094 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0030339 +trainer/policy/mean Std 0.730941 +trainer/policy/mean Max 0.99925 +trainer/policy/mean Min -0.999155 +trainer/policy/std Mean 0.41311 +trainer/policy/std Std 0.0199911 +trainer/policy/std Max 0.432927 +trainer/policy/std Min 0.382138 +trainer/Advantage Weights Mean 4.88711 +trainer/Advantage Weights Std 17.6793 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.62496e-13 +trainer/Advantage Score Mean -0.288289 +trainer/Advantage Score Std 0.554376 +trainer/Advantage Score Max 1.35616 +trainer/Advantage Score Min -2.77789 +trainer/V1 Predictions Mean -74.2483 +trainer/V1 Predictions Std 17.2872 +trainer/V1 Predictions Max -0.373729 +trainer/V1 Predictions Min -87.6921 +trainer/VF Loss 0.0611516 +expl/num steps total 727000 +expl/num paths total 965 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -9.47697e-05 +expl/Actions Std 0.832873 +expl/Actions Max 2.31387 +expl/Actions Min -2.239 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 666001 +eval/num paths total 731 +eval/path length Mean 380 +eval/path length Std 0 +eval/path length Max 380 +eval/path length Min 380 +eval/Rewards Mean 0.00263158 +eval/Rewards Std 0.0512314 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00887895 +eval/Actions Std 0.757329 +eval/Actions Max 0.99967 +eval/Actions Min -0.999542 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.19002e-06 +time/evaluation sampling (s) 5.16952 +time/exploration sampling (s) 6.31305 +time/logging (s) 0.00998629 +time/saving (s) 0.0143582 +time/training (s) 18.8862 +time/epoch (s) 30.3931 +time/total (s) 18591.3 +Epoch -274 +------------------------------ ---------------- +2022-05-15 23:12:51.036451 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -273 finished +------------------------------ ---------------- +epoch -273 +replay_buffer/size 999047 +trainer/num train calls 728000 +trainer/QF1 Loss 0.77538 +trainer/QF2 Loss 0.701406 +trainer/Policy Loss 14.2427 +trainer/Q1 Predictions Mean -74.2741 +trainer/Q1 Predictions Std 17.0043 +trainer/Q1 Predictions Max -6.92016 +trainer/Q1 Predictions Min -87.734 +trainer/Q2 Predictions Mean -74.3358 +trainer/Q2 Predictions Std 17.0392 +trainer/Q2 Predictions Max -5.84215 +trainer/Q2 Predictions Min -87.7381 +trainer/Q Targets Mean -74.3157 +trainer/Q Targets Std 17.1252 +trainer/Q Targets Max -6.7132 +trainer/Q Targets Min -87.9882 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00189681 +trainer/policy/mean Std 0.737111 +trainer/policy/mean Max 0.997615 +trainer/policy/mean Min -0.998538 +trainer/policy/std Mean 0.41164 +trainer/policy/std Std 0.0204145 +trainer/policy/std Max 0.433968 +trainer/policy/std Min 0.381834 +trainer/Advantage Weights Mean 3.01445 +trainer/Advantage Weights Std 13.6564 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.04932e-26 +trainer/Advantage Score Mean -0.439404 +trainer/Advantage Score Std 0.776906 +trainer/Advantage Score Max 0.853266 +trainer/Advantage Score Min -5.98191 +trainer/V1 Predictions Mean -73.9684 +trainer/V1 Predictions Std 17.4124 +trainer/V1 Predictions Max -5.06787 +trainer/V1 Predictions Min -87.8792 +trainer/VF Loss 0.0903244 +expl/num steps total 728000 +expl/num paths total 967 +expl/path length Mean 500 +expl/path length Std 310 +expl/path length Max 810 +expl/path length Min 190 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0402512 +expl/Actions Std 0.833051 +expl/Actions Max 2.28898 +expl/Actions Min -2.19316 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 667001 +eval/num paths total 732 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.337677 +eval/Actions Std 0.681609 +eval/Actions Max 0.999975 +eval/Actions Min -0.999304 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.08292e-06 +time/evaluation sampling (s) 5.07705 +time/exploration sampling (s) 5.8712 +time/logging (s) 0.00974754 +time/saving (s) 0.020113 +time/training (s) 19.0354 +time/epoch (s) 30.0135 +time/total (s) 18621.4 +Epoch -273 +------------------------------ ---------------- +2022-05-15 23:13:22.330430 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -272 finished +------------------------------ ---------------- +epoch -272 +replay_buffer/size 999047 +trainer/num train calls 729000 +trainer/QF1 Loss 0.541333 +trainer/QF2 Loss 0.575222 +trainer/Policy Loss 27.0052 +trainer/Q1 Predictions Mean -74.4883 +trainer/Q1 Predictions Std 17.4519 +trainer/Q1 Predictions Max -1.15884 +trainer/Q1 Predictions Min -88.4698 +trainer/Q2 Predictions Mean -74.5082 +trainer/Q2 Predictions Std 17.4471 +trainer/Q2 Predictions Max -1.23406 +trainer/Q2 Predictions Min -88.4193 +trainer/Q Targets Mean -74.2694 +trainer/Q Targets Std 17.2791 +trainer/Q Targets Max -2.02785 +trainer/Q Targets Min -87.5034 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0206868 +trainer/policy/mean Std 0.725327 +trainer/policy/mean Max 0.999786 +trainer/policy/mean Min -0.999708 +trainer/policy/std Mean 0.409908 +trainer/policy/std Std 0.020771 +trainer/policy/std Max 0.431294 +trainer/policy/std Min 0.37773 +trainer/Advantage Weights Mean 6.83132 +trainer/Advantage Weights Std 22.1666 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.18084e-19 +trainer/Advantage Score Mean -0.299764 +trainer/Advantage Score Std 0.577395 +trainer/Advantage Score Max 1.25682 +trainer/Advantage Score Min -4.19277 +trainer/V1 Predictions Mean -73.9916 +trainer/V1 Predictions Std 17.5117 +trainer/V1 Predictions Max -1.15518 +trainer/V1 Predictions Min -87.444 +trainer/VF Loss 0.0692852 +expl/num steps total 729000 +expl/num paths total 968 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.274563 +expl/Actions Std 0.849022 +expl/Actions Max 2.243 +expl/Actions Min -2.47029 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 668001 +eval/num paths total 733 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00112951 +eval/Actions Std 0.671721 +eval/Actions Max 0.999777 +eval/Actions Min -0.999117 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.29002e-05 +time/evaluation sampling (s) 4.52422 +time/exploration sampling (s) 7.08358 +time/logging (s) 0.0121896 +time/saving (s) 0.0167396 +time/training (s) 19.6456 +time/epoch (s) 31.2824 +time/total (s) 18652.6 +Epoch -272 +------------------------------ ---------------- +2022-05-15 23:13:53.878807 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -271 finished +------------------------------ ---------------- +epoch -271 +replay_buffer/size 999047 +trainer/num train calls 730000 +trainer/QF1 Loss 0.622559 +trainer/QF2 Loss 0.635203 +trainer/Policy Loss 6.29149 +trainer/Q1 Predictions Mean -75.5564 +trainer/Q1 Predictions Std 17.0383 +trainer/Q1 Predictions Max -1.15867 +trainer/Q1 Predictions Min -87.8445 +trainer/Q2 Predictions Mean -75.5452 +trainer/Q2 Predictions Std 16.9289 +trainer/Q2 Predictions Max -0.579599 +trainer/Q2 Predictions Min -87.7579 +trainer/Q Targets Mean -75.2897 +trainer/Q Targets Std 17.2163 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9397 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00698516 +trainer/policy/mean Std 0.743673 +trainer/policy/mean Max 0.999982 +trainer/policy/mean Min -0.999775 +trainer/policy/std Mean 0.412418 +trainer/policy/std Std 0.0195313 +trainer/policy/std Max 0.432981 +trainer/policy/std Min 0.383784 +trainer/Advantage Weights Mean 2.48971 +trainer/Advantage Weights Std 13.3133 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.73199e-22 +trainer/Advantage Score Mean -0.538239 +trainer/Advantage Score Std 0.669651 +trainer/Advantage Score Max 1.13334 +trainer/Advantage Score Min -4.89108 +trainer/V1 Predictions Mean -75.1068 +trainer/V1 Predictions Std 17.1688 +trainer/V1 Predictions Max -1.1389 +trainer/V1 Predictions Min -87.8092 +trainer/VF Loss 0.0847901 +expl/num steps total 730000 +expl/num paths total 970 +expl/path length Mean 500 +expl/path length Std 117 +expl/path length Max 617 +expl/path length Min 383 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0144952 +expl/Actions Std 0.83112 +expl/Actions Max 2.99624 +expl/Actions Min -2.35297 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 668409 +eval/num paths total 734 +eval/path length Mean 408 +eval/path length Std 0 +eval/path length Max 408 +eval/path length Min 408 +eval/Rewards Mean 0.00245098 +eval/Rewards Std 0.0494467 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00147245 +eval/Actions Std 0.73452 +eval/Actions Max 0.999814 +eval/Actions Min -0.999433 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.13919e-05 +time/evaluation sampling (s) 4.63037 +time/exploration sampling (s) 7.19231 +time/logging (s) 0.0101311 +time/saving (s) 0.0188739 +time/training (s) 19.6791 +time/epoch (s) 31.5308 +time/total (s) 18684.2 +Epoch -271 +------------------------------ ---------------- +2022-05-15 23:14:24.601400 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -270 finished +------------------------------ ---------------- +epoch -270 +replay_buffer/size 999047 +trainer/num train calls 731000 +trainer/QF1 Loss 0.976858 +trainer/QF2 Loss 0.979818 +trainer/Policy Loss 25.4299 +trainer/Q1 Predictions Mean -72.0906 +trainer/Q1 Predictions Std 19.7181 +trainer/Q1 Predictions Max -0.353125 +trainer/Q1 Predictions Min -87.7501 +trainer/Q2 Predictions Mean -72.07 +trainer/Q2 Predictions Std 19.6547 +trainer/Q2 Predictions Max -0.830068 +trainer/Q2 Predictions Min -87.7068 +trainer/Q Targets Mean -72.4783 +trainer/Q Targets Std 19.541 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9564 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0130768 +trainer/policy/mean Std 0.729372 +trainer/policy/mean Max 0.997487 +trainer/policy/mean Min -0.998943 +trainer/policy/std Mean 0.411844 +trainer/policy/std Std 0.0203936 +trainer/policy/std Max 0.436652 +trainer/policy/std Min 0.381568 +trainer/Advantage Weights Mean 6.51821 +trainer/Advantage Weights Std 21.1965 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.14978e-13 +trainer/Advantage Score Mean -0.298546 +trainer/Advantage Score Std 0.570188 +trainer/Advantage Score Max 1.68759 +trainer/Advantage Score Min -2.85106 +trainer/V1 Predictions Mean -72.2153 +trainer/V1 Predictions Std 19.7597 +trainer/V1 Predictions Max -0.299127 +trainer/V1 Predictions Min -87.7573 +trainer/VF Loss 0.0758658 +expl/num steps total 731000 +expl/num paths total 971 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0719393 +expl/Actions Std 0.78575 +expl/Actions Max 2.37466 +expl/Actions Min -2.21475 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 669058 +eval/num paths total 735 +eval/path length Mean 649 +eval/path length Std 0 +eval/path length Max 649 +eval/path length Min 649 +eval/Rewards Mean 0.00154083 +eval/Rewards Std 0.0392232 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0115787 +eval/Actions Std 0.738433 +eval/Actions Max 0.999894 +eval/Actions Min -0.999523 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.27777e-05 +time/evaluation sampling (s) 4.75326 +time/exploration sampling (s) 7.11755 +time/logging (s) 0.0105141 +time/saving (s) 0.0180265 +time/training (s) 18.8031 +time/epoch (s) 30.7024 +time/total (s) 18714.9 +Epoch -270 +------------------------------ ---------------- +2022-05-15 23:14:55.082232 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -269 finished +------------------------------ ---------------- +epoch -269 +replay_buffer/size 999047 +trainer/num train calls 732000 +trainer/QF1 Loss 1.66648 +trainer/QF2 Loss 1.36195 +trainer/Policy Loss 6.50924 +trainer/Q1 Predictions Mean -72.9412 +trainer/Q1 Predictions Std 20.2177 +trainer/Q1 Predictions Max -1.49948 +trainer/Q1 Predictions Min -87.5022 +trainer/Q2 Predictions Mean -72.9247 +trainer/Q2 Predictions Std 20.2468 +trainer/Q2 Predictions Max -0.879298 +trainer/Q2 Predictions Min -87.3007 +trainer/Q Targets Mean -72.8315 +trainer/Q Targets Std 20.4373 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.673 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0112006 +trainer/policy/mean Std 0.72188 +trainer/policy/mean Max 0.999163 +trainer/policy/mean Min -0.998224 +trainer/policy/std Mean 0.414178 +trainer/policy/std Std 0.0191199 +trainer/policy/std Max 0.435736 +trainer/policy/std Min 0.386664 +trainer/Advantage Weights Mean 2.49564 +trainer/Advantage Weights Std 13.8436 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.79786e-15 +trainer/Advantage Score Mean -0.398076 +trainer/Advantage Score Std 0.583278 +trainer/Advantage Score Max 1.60503 +trainer/Advantage Score Min -3.35099 +trainer/V1 Predictions Mean -72.6499 +trainer/V1 Predictions Std 20.1231 +trainer/V1 Predictions Max -1.18542 +trainer/V1 Predictions Min -87.4436 +trainer/VF Loss 0.0645456 +expl/num steps total 732000 +expl/num paths total 973 +expl/path length Mean 500 +expl/path length Std 167 +expl/path length Max 667 +expl/path length Min 333 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0217453 +expl/Actions Std 0.833964 +expl/Actions Max 2.29677 +expl/Actions Min -2.6847 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 670058 +eval/num paths total 736 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0741397 +eval/Actions Std 0.66183 +eval/Actions Max 0.999449 +eval/Actions Min -0.99903 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.23982e-05 +time/evaluation sampling (s) 4.72871 +time/exploration sampling (s) 6.74408 +time/logging (s) 0.0102354 +time/saving (s) 0.0127686 +time/training (s) 18.969 +time/epoch (s) 30.4648 +time/total (s) 18745.4 +Epoch -269 +------------------------------ ---------------- +2022-05-15 23:15:25.500363 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -268 finished +------------------------------ ---------------- +epoch -268 +replay_buffer/size 999047 +trainer/num train calls 733000 +trainer/QF1 Loss 0.779263 +trainer/QF2 Loss 0.81979 +trainer/Policy Loss 12.8803 +trainer/Q1 Predictions Mean -72.5351 +trainer/Q1 Predictions Std 19.5148 +trainer/Q1 Predictions Max -1.37621 +trainer/Q1 Predictions Min -88.2321 +trainer/Q2 Predictions Mean -72.6146 +trainer/Q2 Predictions Std 19.5156 +trainer/Q2 Predictions Max -0.671719 +trainer/Q2 Predictions Min -88.3266 +trainer/Q Targets Mean -72.6569 +trainer/Q Targets Std 19.4151 +trainer/Q Targets Max -2.1509 +trainer/Q Targets Min -87.9091 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0139083 +trainer/policy/mean Std 0.728926 +trainer/policy/mean Max 0.999799 +trainer/policy/mean Min -0.999218 +trainer/policy/std Mean 0.411294 +trainer/policy/std Std 0.0191076 +trainer/policy/std Max 0.433926 +trainer/policy/std Min 0.381803 +trainer/Advantage Weights Mean 3.11554 +trainer/Advantage Weights Std 13.7542 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.88306e-17 +trainer/Advantage Score Mean -0.433001 +trainer/Advantage Score Std 0.635526 +trainer/Advantage Score Max 0.920401 +trainer/Advantage Score Min -3.8511 +trainer/V1 Predictions Mean -72.363 +trainer/V1 Predictions Std 19.5879 +trainer/V1 Predictions Max -1.6196 +trainer/V1 Predictions Min -87.7873 +trainer/VF Loss 0.0693967 +expl/num steps total 733000 +expl/num paths total 975 +expl/path length Mean 500 +expl/path length Std 120 +expl/path length Max 620 +expl/path length Min 380 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0532299 +expl/Actions Std 0.819669 +expl/Actions Max 2.27423 +expl/Actions Min -2.16614 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 670658 +eval/num paths total 737 +eval/path length Mean 600 +eval/path length Std 0 +eval/path length Max 600 +eval/path length Min 600 +eval/Rewards Mean 0.00166667 +eval/Rewards Std 0.0407908 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0223785 +eval/Actions Std 0.722951 +eval/Actions Max 0.999965 +eval/Actions Min -0.999686 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.1866e-05 +time/evaluation sampling (s) 4.52033 +time/exploration sampling (s) 6.95084 +time/logging (s) 0.0110525 +time/saving (s) 0.0191405 +time/training (s) 18.903 +time/epoch (s) 30.4044 +time/total (s) 18775.8 +Epoch -268 +------------------------------ ---------------- +2022-05-15 23:15:56.724911 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -267 finished +------------------------------ ---------------- +epoch -267 +replay_buffer/size 999047 +trainer/num train calls 734000 +trainer/QF1 Loss 7.90615 +trainer/QF2 Loss 7.89244 +trainer/Policy Loss 10.4814 +trainer/Q1 Predictions Mean -75.6407 +trainer/Q1 Predictions Std 15.7065 +trainer/Q1 Predictions Max -0.842925 +trainer/Q1 Predictions Min -86.9844 +trainer/Q2 Predictions Mean -75.6142 +trainer/Q2 Predictions Std 15.5896 +trainer/Q2 Predictions Max -0.389583 +trainer/Q2 Predictions Min -86.8719 +trainer/Q Targets Mean -75.5508 +trainer/Q Targets Std 16.0255 +trainer/Q Targets Max -1.60971 +trainer/Q Targets Min -87.0891 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0484583 +trainer/policy/mean Std 0.727531 +trainer/policy/mean Max 0.999446 +trainer/policy/mean Min -0.999689 +trainer/policy/std Mean 0.410834 +trainer/policy/std Std 0.0195896 +trainer/policy/std Max 0.430954 +trainer/policy/std Min 0.376711 +trainer/Advantage Weights Mean 2.98006 +trainer/Advantage Weights Std 14.5635 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.74102e-16 +trainer/Advantage Score Mean -0.392843 +trainer/Advantage Score Std 0.548198 +trainer/Advantage Score Max 1.35207 +trainer/Advantage Score Min -3.62869 +trainer/V1 Predictions Mean -75.485 +trainer/V1 Predictions Std 15.8389 +trainer/V1 Predictions Max -1.06875 +trainer/V1 Predictions Min -86.949 +trainer/VF Loss 0.0609638 +expl/num steps total 734000 +expl/num paths total 976 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.16956 +expl/Actions Std 0.794093 +expl/Actions Max 2.37172 +expl/Actions Min -2.37276 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 671658 +eval/num paths total 738 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0542339 +eval/Actions Std 0.705747 +eval/Actions Max 0.998652 +eval/Actions Min -0.999775 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.36211e-05 +time/evaluation sampling (s) 5.52229 +time/exploration sampling (s) 6.56746 +time/logging (s) 0.0117989 +time/saving (s) 0.0137292 +time/training (s) 19.0857 +time/epoch (s) 31.201 +time/total (s) 18807 +Epoch -267 +------------------------------ ---------------- +2022-05-15 23:16:27.351525 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -266 finished +------------------------------ ---------------- +epoch -266 +replay_buffer/size 999047 +trainer/num train calls 735000 +trainer/QF1 Loss 0.68317 +trainer/QF2 Loss 0.61784 +trainer/Policy Loss 15.1331 +trainer/Q1 Predictions Mean -72.567 +trainer/Q1 Predictions Std 19.1589 +trainer/Q1 Predictions Max -3.33927 +trainer/Q1 Predictions Min -87.2755 +trainer/Q2 Predictions Mean -72.6702 +trainer/Q2 Predictions Std 19.1905 +trainer/Q2 Predictions Max -2.35576 +trainer/Q2 Predictions Min -87.3405 +trainer/Q Targets Mean -72.6075 +trainer/Q Targets Std 19.1512 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3367 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0469528 +trainer/policy/mean Std 0.727063 +trainer/policy/mean Max 0.999385 +trainer/policy/mean Min -0.998566 +trainer/policy/std Mean 0.411854 +trainer/policy/std Std 0.0191273 +trainer/policy/std Max 0.43257 +trainer/policy/std Min 0.383042 +trainer/Advantage Weights Mean 3.34352 +trainer/Advantage Weights Std 15.259 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.32582e-16 +trainer/Advantage Score Mean -0.376813 +trainer/Advantage Score Std 0.632094 +trainer/Advantage Score Max 1.48844 +trainer/Advantage Score Min -3.65593 +trainer/V1 Predictions Mean -72.3976 +trainer/V1 Predictions Std 19.3139 +trainer/V1 Predictions Max -2.17492 +trainer/V1 Predictions Min -87.2073 +trainer/VF Loss 0.0769536 +expl/num steps total 735000 +expl/num paths total 978 +expl/path length Mean 500 +expl/path length Std 13 +expl/path length Max 513 +expl/path length Min 487 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0189377 +expl/Actions Std 0.834547 +expl/Actions Max 2.20214 +expl/Actions Min -2.28598 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 672475 +eval/num paths total 739 +eval/path length Mean 817 +eval/path length Std 0 +eval/path length Max 817 +eval/path length Min 817 +eval/Rewards Mean 0.00122399 +eval/Rewards Std 0.0349642 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0503923 +eval/Actions Std 0.756662 +eval/Actions Max 0.999611 +eval/Actions Min -0.999679 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.82611e-06 +time/evaluation sampling (s) 4.81178 +time/exploration sampling (s) 6.12431 +time/logging (s) 0.00889637 +time/saving (s) 0.0113856 +time/training (s) 19.6499 +time/epoch (s) 30.6063 +time/total (s) 18837.6 +Epoch -266 +------------------------------ ---------------- +2022-05-15 23:16:57.317855 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -265 finished +------------------------------ ---------------- +epoch -265 +replay_buffer/size 999047 +trainer/num train calls 736000 +trainer/QF1 Loss 5.0666 +trainer/QF2 Loss 4.82105 +trainer/Policy Loss 11.4149 +trainer/Q1 Predictions Mean -73.5229 +trainer/Q1 Predictions Std 18.7662 +trainer/Q1 Predictions Max -1.63669 +trainer/Q1 Predictions Min -87.5215 +trainer/Q2 Predictions Mean -73.4603 +trainer/Q2 Predictions Std 18.7474 +trainer/Q2 Predictions Max -1.43095 +trainer/Q2 Predictions Min -87.7057 +trainer/Q Targets Mean -73.2645 +trainer/Q Targets Std 18.7651 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0946 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.025252 +trainer/policy/mean Std 0.726174 +trainer/policy/mean Max 0.998388 +trainer/policy/mean Min -0.998911 +trainer/policy/std Mean 0.412856 +trainer/policy/std Std 0.0203422 +trainer/policy/std Max 0.433371 +trainer/policy/std Min 0.382726 +trainer/Advantage Weights Mean 1.46883 +trainer/Advantage Weights Std 10.0488 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.6068e-32 +trainer/Advantage Score Mean -0.618502 +trainer/Advantage Score Std 0.797753 +trainer/Advantage Score Max 0.671692 +trainer/Advantage Score Min -7.23999 +trainer/V1 Predictions Mean -72.8445 +trainer/V1 Predictions Std 19.2776 +trainer/V1 Predictions Max -1.09814 +trainer/V1 Predictions Min -86.996 +trainer/VF Loss 0.105821 +expl/num steps total 736000 +expl/num paths total 980 +expl/path length Mean 500 +expl/path length Std 228 +expl/path length Max 728 +expl/path length Min 272 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0388394 +expl/Actions Std 0.810748 +expl/Actions Max 2.16899 +expl/Actions Min -2.37417 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 673475 +eval/num paths total 740 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.193228 +eval/Actions Std 0.770383 +eval/Actions Max 0.999981 +eval/Actions Min -0.999744 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.34869e-05 +time/evaluation sampling (s) 4.65953 +time/exploration sampling (s) 6.38803 +time/logging (s) 0.0112623 +time/saving (s) 0.0154122 +time/training (s) 18.8804 +time/epoch (s) 29.9546 +time/total (s) 18867.6 +Epoch -265 +------------------------------ ---------------- +2022-05-15 23:17:27.968625 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -264 finished +------------------------------ ---------------- +epoch -264 +replay_buffer/size 999047 +trainer/num train calls 737000 +trainer/QF1 Loss 0.878403 +trainer/QF2 Loss 0.866836 +trainer/Policy Loss 39.1528 +trainer/Q1 Predictions Mean -73.3835 +trainer/Q1 Predictions Std 19.7508 +trainer/Q1 Predictions Max -0.883694 +trainer/Q1 Predictions Min -88.5761 +trainer/Q2 Predictions Mean -73.3944 +trainer/Q2 Predictions Std 19.7337 +trainer/Q2 Predictions Max -1.18746 +trainer/Q2 Predictions Min -88.4301 +trainer/Q Targets Mean -73.8046 +trainer/Q Targets Std 19.7382 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.5135 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0435127 +trainer/policy/mean Std 0.733548 +trainer/policy/mean Max 0.999695 +trainer/policy/mean Min -0.997523 +trainer/policy/std Mean 0.411154 +trainer/policy/std Std 0.0213715 +trainer/policy/std Max 0.434867 +trainer/policy/std Min 0.378865 +trainer/Advantage Weights Mean 6.62961 +trainer/Advantage Weights Std 20.75 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.43667e-15 +trainer/Advantage Score Mean -0.232014 +trainer/Advantage Score Std 0.574766 +trainer/Advantage Score Max 2.42625 +trainer/Advantage Score Min -3.24062 +trainer/V1 Predictions Mean -73.5498 +trainer/V1 Predictions Std 19.9111 +trainer/V1 Predictions Max 0.190761 +trainer/V1 Predictions Min -88.4885 +trainer/VF Loss 0.0809638 +expl/num steps total 737000 +expl/num paths total 982 +expl/path length Mean 500 +expl/path length Std 302 +expl/path length Max 802 +expl/path length Min 198 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.012989 +expl/Actions Std 0.821383 +expl/Actions Max 2.46141 +expl/Actions Min -2.15755 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 674475 +eval/num paths total 741 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.09197 +eval/Actions Std 0.741504 +eval/Actions Max 0.999113 +eval/Actions Min -0.999947 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.36533e-06 +time/evaluation sampling (s) 4.08053 +time/exploration sampling (s) 6.87913 +time/logging (s) 0.0126007 +time/saving (s) 0.0184923 +time/training (s) 19.6432 +time/epoch (s) 30.6339 +time/total (s) 18898.2 +Epoch -264 +------------------------------ ---------------- +2022-05-15 23:17:57.999428 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -263 finished +------------------------------ ---------------- +epoch -263 +replay_buffer/size 999047 +trainer/num train calls 738000 +trainer/QF1 Loss 1.50542 +trainer/QF2 Loss 1.60666 +trainer/Policy Loss 32.1056 +trainer/Q1 Predictions Mean -72.9163 +trainer/Q1 Predictions Std 18.8564 +trainer/Q1 Predictions Max -0.925606 +trainer/Q1 Predictions Min -87.8087 +trainer/Q2 Predictions Mean -72.9247 +trainer/Q2 Predictions Std 18.7582 +trainer/Q2 Predictions Max -0.944923 +trainer/Q2 Predictions Min -87.877 +trainer/Q Targets Mean -73.0424 +trainer/Q Targets Std 18.4283 +trainer/Q Targets Max -1.30392 +trainer/Q Targets Min -87.3752 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0257963 +trainer/policy/mean Std 0.720475 +trainer/policy/mean Max 0.998082 +trainer/policy/mean Min -0.999167 +trainer/policy/std Mean 0.409502 +trainer/policy/std Std 0.0209434 +trainer/policy/std Max 0.43129 +trainer/policy/std Min 0.375985 +trainer/Advantage Weights Mean 6.34919 +trainer/Advantage Weights Std 22.2761 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.01489e-19 +trainer/Advantage Score Mean -0.382271 +trainer/Advantage Score Std 0.715344 +trainer/Advantage Score Max 2.17711 +trainer/Advantage Score Min -4.30485 +trainer/V1 Predictions Mean -72.6425 +trainer/V1 Predictions Std 18.9152 +trainer/V1 Predictions Max -0.576657 +trainer/V1 Predictions Min -87.1001 +trainer/VF Loss 0.129791 +expl/num steps total 738000 +expl/num paths total 984 +expl/path length Mean 500 +expl/path length Std 353 +expl/path length Max 853 +expl/path length Min 147 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0249495 +expl/Actions Std 0.829447 +expl/Actions Max 2.4892 +expl/Actions Min -2.26496 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 675268 +eval/num paths total 742 +eval/path length Mean 793 +eval/path length Std 0 +eval/path length Max 793 +eval/path length Min 793 +eval/Rewards Mean 0.00126103 +eval/Rewards Std 0.0354886 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0402919 +eval/Actions Std 0.7344 +eval/Actions Max 0.99996 +eval/Actions Min -0.999649 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.08383e-05 +time/evaluation sampling (s) 4.86066 +time/exploration sampling (s) 6.31192 +time/logging (s) 0.0108229 +time/saving (s) 0.0169615 +time/training (s) 18.8085 +time/epoch (s) 30.0089 +time/total (s) 18928.2 +Epoch -263 +------------------------------ ---------------- +2022-05-15 23:18:28.641502 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -262 finished +------------------------------ ---------------- +epoch -262 +replay_buffer/size 999047 +trainer/num train calls 739000 +trainer/QF1 Loss 0.577514 +trainer/QF2 Loss 0.512596 +trainer/Policy Loss 7.46864 +trainer/Q1 Predictions Mean -73.3817 +trainer/Q1 Predictions Std 18.128 +trainer/Q1 Predictions Max -0.308709 +trainer/Q1 Predictions Min -87.5508 +trainer/Q2 Predictions Mean -73.3378 +trainer/Q2 Predictions Std 18.1744 +trainer/Q2 Predictions Max -0.19131 +trainer/Q2 Predictions Min -87.2028 +trainer/Q Targets Mean -73.2061 +trainer/Q Targets Std 18.1524 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3062 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0158273 +trainer/policy/mean Std 0.722058 +trainer/policy/mean Max 0.998624 +trainer/policy/mean Min -0.999612 +trainer/policy/std Mean 0.410777 +trainer/policy/std Std 0.0209119 +trainer/policy/std Max 0.4327 +trainer/policy/std Min 0.375818 +trainer/Advantage Weights Mean 1.83073 +trainer/Advantage Weights Std 11.5871 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.4434e-15 +trainer/Advantage Score Mean -0.538207 +trainer/Advantage Score Std 0.561571 +trainer/Advantage Score Max 0.972607 +trainer/Advantage Score Min -3.41718 +trainer/V1 Predictions Mean -72.9359 +trainer/V1 Predictions Std 18.3197 +trainer/V1 Predictions Max 0.611293 +trainer/V1 Predictions Min -87.1777 +trainer/VF Loss 0.0665764 +expl/num steps total 739000 +expl/num paths total 986 +expl/path length Mean 500 +expl/path length Std 453 +expl/path length Max 953 +expl/path length Min 47 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0242272 +expl/Actions Std 0.819701 +expl/Actions Max 2.44554 +expl/Actions Min -2.06152 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 676268 +eval/num paths total 743 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00994504 +eval/Actions Std 0.707323 +eval/Actions Max 0.999889 +eval/Actions Min -0.999463 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.08629e-05 +time/evaluation sampling (s) 4.44157 +time/exploration sampling (s) 7.04345 +time/logging (s) 0.0144117 +time/saving (s) 0.0196548 +time/training (s) 19.109 +time/epoch (s) 30.6281 +time/total (s) 18958.8 +Epoch -262 +------------------------------ ---------------- +2022-05-15 23:18:59.869358 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -261 finished +------------------------------ ---------------- +epoch -261 +replay_buffer/size 999047 +trainer/num train calls 740000 +trainer/QF1 Loss 0.955584 +trainer/QF2 Loss 0.886592 +trainer/Policy Loss 7.35024 +trainer/Q1 Predictions Mean -72.935 +trainer/Q1 Predictions Std 19.5888 +trainer/Q1 Predictions Max -1.58582 +trainer/Q1 Predictions Min -87.7951 +trainer/Q2 Predictions Mean -72.9269 +trainer/Q2 Predictions Std 19.5798 +trainer/Q2 Predictions Max -2.09144 +trainer/Q2 Predictions Min -87.8655 +trainer/Q Targets Mean -72.7959 +trainer/Q Targets Std 19.8073 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8619 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0113493 +trainer/policy/mean Std 0.729519 +trainer/policy/mean Max 0.999621 +trainer/policy/mean Min -0.99907 +trainer/policy/std Mean 0.411989 +trainer/policy/std Std 0.0199705 +trainer/policy/std Max 0.433318 +trainer/policy/std Min 0.379901 +trainer/Advantage Weights Mean 2.16173 +trainer/Advantage Weights Std 11.2745 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.70204e-17 +trainer/Advantage Score Mean -0.430456 +trainer/Advantage Score Std 0.611487 +trainer/Advantage Score Max 0.997436 +trainer/Advantage Score Min -3.78351 +trainer/V1 Predictions Mean -72.5223 +trainer/V1 Predictions Std 19.9772 +trainer/V1 Predictions Max 0.29541 +trainer/V1 Predictions Min -87.5959 +trainer/VF Loss 0.0639124 +expl/num steps total 740000 +expl/num paths total 988 +expl/path length Mean 500 +expl/path length Std 0 +expl/path length Max 500 +expl/path length Min 500 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.000274058 +expl/Actions Std 0.830271 +expl/Actions Max 2.63324 +expl/Actions Min -2.32504 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 677027 +eval/num paths total 744 +eval/path length Mean 759 +eval/path length Std 0 +eval/path length Max 759 +eval/path length Min 759 +eval/Rewards Mean 0.00131752 +eval/Rewards Std 0.0362738 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0381132 +eval/Actions Std 0.735621 +eval/Actions Max 0.999344 +eval/Actions Min -0.999221 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.30041e-05 +time/evaluation sampling (s) 4.85218 +time/exploration sampling (s) 7.12463 +time/logging (s) 0.00710441 +time/saving (s) 0.0112954 +time/training (s) 19.2037 +time/epoch (s) 31.1989 +time/total (s) 18990.1 +Epoch -261 +------------------------------ ---------------- +2022-05-15 23:19:30.100196 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -260 finished +------------------------------ ---------------- +epoch -260 +replay_buffer/size 999047 +trainer/num train calls 741000 +trainer/QF1 Loss 0.485827 +trainer/QF2 Loss 0.479965 +trainer/Policy Loss 21.7135 +trainer/Q1 Predictions Mean -73.054 +trainer/Q1 Predictions Std 19.4042 +trainer/Q1 Predictions Max -1.04053 +trainer/Q1 Predictions Min -87.5381 +trainer/Q2 Predictions Mean -73.1041 +trainer/Q2 Predictions Std 19.4444 +trainer/Q2 Predictions Max -0.1079 +trainer/Q2 Predictions Min -87.3034 +trainer/Q Targets Mean -73.0741 +trainer/Q Targets Std 19.4911 +trainer/Q Targets Max -0.254747 +trainer/Q Targets Min -87.0603 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0105268 +trainer/policy/mean Std 0.731801 +trainer/policy/mean Max 0.999925 +trainer/policy/mean Min -0.99778 +trainer/policy/std Mean 0.41159 +trainer/policy/std Std 0.0196065 +trainer/policy/std Max 0.435011 +trainer/policy/std Min 0.380204 +trainer/Advantage Weights Mean 5.25905 +trainer/Advantage Weights Std 20.5593 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.59698e-14 +trainer/Advantage Score Mean -0.405174 +trainer/Advantage Score Std 0.584057 +trainer/Advantage Score Max 1.64238 +trainer/Advantage Score Min -3.03496 +trainer/V1 Predictions Mean -72.829 +trainer/V1 Predictions Std 19.6296 +trainer/V1 Predictions Max 0.884118 +trainer/V1 Predictions Min -86.934 +trainer/VF Loss 0.0809514 +expl/num steps total 741000 +expl/num paths total 990 +expl/path length Mean 500 +expl/path length Std 49 +expl/path length Max 549 +expl/path length Min 451 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0189944 +expl/Actions Std 0.830822 +expl/Actions Max 2.1685 +expl/Actions Min -2.62894 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 677640 +eval/num paths total 745 +eval/path length Mean 613 +eval/path length Std 0 +eval/path length Max 613 +eval/path length Min 613 +eval/Rewards Mean 0.00163132 +eval/Rewards Std 0.0403567 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.028753 +eval/Actions Std 0.736329 +eval/Actions Max 0.9999 +eval/Actions Min -0.999838 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.98792e-06 +time/evaluation sampling (s) 4.62677 +time/exploration sampling (s) 7.04927 +time/logging (s) 0.011518 +time/saving (s) 0.0187761 +time/training (s) 18.5137 +time/epoch (s) 30.22 +time/total (s) 19020.3 +Epoch -260 +------------------------------ ---------------- +2022-05-15 23:20:00.564417 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -259 finished +------------------------------ ---------------- +epoch -259 +replay_buffer/size 999047 +trainer/num train calls 742000 +trainer/QF1 Loss 0.759651 +trainer/QF2 Loss 0.775123 +trainer/Policy Loss 7.98369 +trainer/Q1 Predictions Mean -75.8725 +trainer/Q1 Predictions Std 15.9682 +trainer/Q1 Predictions Max -1.35583 +trainer/Q1 Predictions Min -87.9755 +trainer/Q2 Predictions Mean -75.8798 +trainer/Q2 Predictions Std 16.0293 +trainer/Q2 Predictions Max -1.19692 +trainer/Q2 Predictions Min -87.9924 +trainer/Q Targets Mean -75.4585 +trainer/Q Targets Std 15.9473 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5414 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00997439 +trainer/policy/mean Std 0.733517 +trainer/policy/mean Max 0.999598 +trainer/policy/mean Min -0.998254 +trainer/policy/std Mean 0.411562 +trainer/policy/std Std 0.019216 +trainer/policy/std Max 0.432814 +trainer/policy/std Min 0.379592 +trainer/Advantage Weights Mean 2.4539 +trainer/Advantage Weights Std 12.9815 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.01122e-15 +trainer/Advantage Score Mean -0.507715 +trainer/Advantage Score Std 0.622487 +trainer/Advantage Score Max 0.998085 +trainer/Advantage Score Min -3.34364 +trainer/V1 Predictions Mean -75.2013 +trainer/V1 Predictions Std 16.1994 +trainer/V1 Predictions Max -0.105403 +trainer/V1 Predictions Min -87.4149 +trainer/VF Loss 0.0744913 +expl/num steps total 742000 +expl/num paths total 992 +expl/path length Mean 500 +expl/path length Std 93 +expl/path length Max 593 +expl/path length Min 407 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0415061 +expl/Actions Std 0.811173 +expl/Actions Max 2.21025 +expl/Actions Min -2.10325 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 678640 +eval/num paths total 746 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.153379 +eval/Actions Std 0.700106 +eval/Actions Max 0.99946 +eval/Actions Min -0.998382 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.56229e-05 +time/evaluation sampling (s) 4.87726 +time/exploration sampling (s) 6.36344 +time/logging (s) 0.0136111 +time/saving (s) 0.0154784 +time/training (s) 19.1755 +time/epoch (s) 30.4453 +time/total (s) 19050.7 +Epoch -259 +------------------------------ ---------------- +2022-05-15 23:20:32.088868 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -258 finished +------------------------------ ---------------- +epoch -258 +replay_buffer/size 999047 +trainer/num train calls 743000 +trainer/QF1 Loss 0.607496 +trainer/QF2 Loss 0.603152 +trainer/Policy Loss 22.9521 +trainer/Q1 Predictions Mean -74.8405 +trainer/Q1 Predictions Std 16.1485 +trainer/Q1 Predictions Max -1.92247 +trainer/Q1 Predictions Min -87.3034 +trainer/Q2 Predictions Mean -74.8466 +trainer/Q2 Predictions Std 16.1554 +trainer/Q2 Predictions Max -2.58608 +trainer/Q2 Predictions Min -87.2628 +trainer/Q Targets Mean -75.0688 +trainer/Q Targets Std 16.3597 +trainer/Q Targets Max -1.51493 +trainer/Q Targets Min -87.4646 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0274329 +trainer/policy/mean Std 0.732851 +trainer/policy/mean Max 0.999671 +trainer/policy/mean Min -0.999408 +trainer/policy/std Mean 0.411638 +trainer/policy/std Std 0.0194838 +trainer/policy/std Max 0.430457 +trainer/policy/std Min 0.378733 +trainer/Advantage Weights Mean 4.84996 +trainer/Advantage Weights Std 17.0698 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.59827e-16 +trainer/Advantage Score Mean -0.253208 +trainer/Advantage Score Std 0.562575 +trainer/Advantage Score Max 2.35266 +trainer/Advantage Score Min -3.51189 +trainer/V1 Predictions Mean -74.8294 +trainer/V1 Predictions Std 16.4934 +trainer/V1 Predictions Max -0.51235 +trainer/V1 Predictions Min -87.3851 +trainer/VF Loss 0.0733139 +expl/num steps total 743000 +expl/num paths total 994 +expl/path length Mean 500 +expl/path length Std 380 +expl/path length Max 880 +expl/path length Min 120 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0116574 +expl/Actions Std 0.819983 +expl/Actions Max 2.73626 +expl/Actions Min -2.53232 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 679204 +eval/num paths total 747 +eval/path length Mean 564 +eval/path length Std 0 +eval/path length Max 564 +eval/path length Min 564 +eval/Rewards Mean 0.00177305 +eval/Rewards Std 0.0420703 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00927477 +eval/Actions Std 0.738963 +eval/Actions Max 0.999947 +eval/Actions Min -0.99947 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.31191e-05 +time/evaluation sampling (s) 4.90895 +time/exploration sampling (s) 7.29978 +time/logging (s) 0.010491 +time/saving (s) 0.0183809 +time/training (s) 19.263 +time/epoch (s) 31.5006 +time/total (s) 19082.2 +Epoch -258 +------------------------------ ---------------- +2022-05-15 23:21:02.677767 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -257 finished +------------------------------ ---------------- +epoch -257 +replay_buffer/size 999047 +trainer/num train calls 744000 +trainer/QF1 Loss 0.751924 +trainer/QF2 Loss 0.795202 +trainer/Policy Loss 16.8426 +trainer/Q1 Predictions Mean -73.3112 +trainer/Q1 Predictions Std 17.9839 +trainer/Q1 Predictions Max -0.330351 +trainer/Q1 Predictions Min -87.8186 +trainer/Q2 Predictions Mean -73.2481 +trainer/Q2 Predictions Std 18.0675 +trainer/Q2 Predictions Max 0.112961 +trainer/Q2 Predictions Min -87.6331 +trainer/Q Targets Mean -73.2646 +trainer/Q Targets Std 17.9417 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8578 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00979572 +trainer/policy/mean Std 0.733848 +trainer/policy/mean Max 0.99868 +trainer/policy/mean Min -0.999599 +trainer/policy/std Mean 0.411706 +trainer/policy/std Std 0.0193792 +trainer/policy/std Max 0.432985 +trainer/policy/std Min 0.380008 +trainer/Advantage Weights Mean 4.20517 +trainer/Advantage Weights Std 15.8672 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.09946e-17 +trainer/Advantage Score Mean -0.324437 +trainer/Advantage Score Std 0.611293 +trainer/Advantage Score Max 2.1604 +trainer/Advantage Score Min -3.90491 +trainer/V1 Predictions Mean -72.9826 +trainer/V1 Predictions Std 18.1662 +trainer/V1 Predictions Max -0.0974309 +trainer/V1 Predictions Min -87.6021 +trainer/VF Loss 0.0733544 +expl/num steps total 744000 +expl/num paths total 996 +expl/path length Mean 500 +expl/path length Std 422 +expl/path length Max 922 +expl/path length Min 78 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0299061 +expl/Actions Std 0.821844 +expl/Actions Max 2.34082 +expl/Actions Min -2.27976 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 680204 +eval/num paths total 748 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0400147 +eval/Actions Std 0.756699 +eval/Actions Max 0.99965 +eval/Actions Min -0.998772 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 6.4401e-06 +time/evaluation sampling (s) 4.83625 +time/exploration sampling (s) 6.4993 +time/logging (s) 0.0121022 +time/saving (s) 0.0170647 +time/training (s) 19.2056 +time/epoch (s) 30.5704 +time/total (s) 19112.8 +Epoch -257 +------------------------------ ---------------- +2022-05-15 23:21:33.286583 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -256 finished +------------------------------ ---------------- +epoch -256 +replay_buffer/size 999047 +trainer/num train calls 745000 +trainer/QF1 Loss 0.443587 +trainer/QF2 Loss 0.550388 +trainer/Policy Loss 22.7028 +trainer/Q1 Predictions Mean -74.8846 +trainer/Q1 Predictions Std 17.2844 +trainer/Q1 Predictions Max -0.273737 +trainer/Q1 Predictions Min -87.1729 +trainer/Q2 Predictions Mean -74.8001 +trainer/Q2 Predictions Std 17.249 +trainer/Q2 Predictions Max -0.113573 +trainer/Q2 Predictions Min -86.9823 +trainer/Q Targets Mean -74.9128 +trainer/Q Targets Std 17.4254 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2595 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0282143 +trainer/policy/mean Std 0.728121 +trainer/policy/mean Max 0.999968 +trainer/policy/mean Min -0.999677 +trainer/policy/std Mean 0.409668 +trainer/policy/std Std 0.0198137 +trainer/policy/std Max 0.431826 +trainer/policy/std Min 0.380563 +trainer/Advantage Weights Mean 5.81237 +trainer/Advantage Weights Std 21.142 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.87987e-18 +trainer/Advantage Score Mean -0.268917 +trainer/Advantage Score Std 0.480757 +trainer/Advantage Score Max 1.66684 +trainer/Advantage Score Min -3.98614 +trainer/V1 Predictions Mean -74.7717 +trainer/V1 Predictions Std 17.335 +trainer/V1 Predictions Max -0.263043 +trainer/V1 Predictions Min -87.3731 +trainer/VF Loss 0.0629833 +expl/num steps total 745000 +expl/num paths total 998 +expl/path length Mean 500 +expl/path length Std 235 +expl/path length Max 735 +expl/path length Min 265 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.032926 +expl/Actions Std 0.831022 +expl/Actions Max 2.3412 +expl/Actions Min -2.166 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 681204 +eval/num paths total 749 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.100739 +eval/Actions Std 0.726646 +eval/Actions Max 0.999942 +eval/Actions Min -0.999602 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.77698e-06 +time/evaluation sampling (s) 4.88939 +time/exploration sampling (s) 6.81686 +time/logging (s) 0.0110033 +time/saving (s) 0.0128615 +time/training (s) 18.8585 +time/epoch (s) 30.5886 +time/total (s) 19143.4 +Epoch -256 +------------------------------ ---------------- +2022-05-15 23:22:04.545142 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -255 finished +------------------------------ ---------------- +epoch -255 +replay_buffer/size 999047 +trainer/num train calls 746000 +trainer/QF1 Loss 1.25005 +trainer/QF2 Loss 1.02826 +trainer/Policy Loss 41.51 +trainer/Q1 Predictions Mean -74.1536 +trainer/Q1 Predictions Std 17.7113 +trainer/Q1 Predictions Max -2.24355 +trainer/Q1 Predictions Min -86.9677 +trainer/Q2 Predictions Mean -74.1853 +trainer/Q2 Predictions Std 17.6807 +trainer/Q2 Predictions Max -2.26006 +trainer/Q2 Predictions Min -87.0219 +trainer/Q Targets Mean -74.6356 +trainer/Q Targets Std 17.7533 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5594 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000928023 +trainer/policy/mean Std 0.718778 +trainer/policy/mean Max 0.999595 +trainer/policy/mean Min -0.999807 +trainer/policy/std Mean 0.41054 +trainer/policy/std Std 0.0215792 +trainer/policy/std Max 0.43583 +trainer/policy/std Min 0.376909 +trainer/Advantage Weights Mean 11.1589 +trainer/Advantage Weights Std 27.4654 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.04026e-17 +trainer/Advantage Score Mean -0.242226 +trainer/Advantage Score Std 0.637076 +trainer/Advantage Score Max 2.06428 +trainer/Advantage Score Min -3.84309 +trainer/V1 Predictions Mean -74.361 +trainer/V1 Predictions Std 17.8286 +trainer/V1 Predictions Max -4.71826 +trainer/V1 Predictions Min -87.4342 +trainer/VF Loss 0.106842 +expl/num steps total 746000 +expl/num paths total 1000 +expl/path length Mean 500 +expl/path length Std 262 +expl/path length Max 762 +expl/path length Min 238 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0435552 +expl/Actions Std 0.824932 +expl/Actions Max 2.30285 +expl/Actions Min -2.3295 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 682204 +eval/num paths total 750 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0467196 +eval/Actions Std 0.7018 +eval/Actions Max 0.999976 +eval/Actions Min -0.999843 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.06343e-05 +time/evaluation sampling (s) 5.40305 +time/exploration sampling (s) 7.07059 +time/logging (s) 0.0111447 +time/saving (s) 0.0161529 +time/training (s) 18.7392 +time/epoch (s) 31.2401 +time/total (s) 19174.7 +Epoch -255 +------------------------------ ---------------- +2022-05-15 23:22:34.872924 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -254 finished +------------------------------ ---------------- +epoch -254 +replay_buffer/size 999047 +trainer/num train calls 747000 +trainer/QF1 Loss 1.22201 +trainer/QF2 Loss 1.12388 +trainer/Policy Loss 4.43753 +trainer/Q1 Predictions Mean -75.0779 +trainer/Q1 Predictions Std 17.3845 +trainer/Q1 Predictions Max -0.331605 +trainer/Q1 Predictions Min -87.8851 +trainer/Q2 Predictions Mean -75.1022 +trainer/Q2 Predictions Std 17.4127 +trainer/Q2 Predictions Max -0.287411 +trainer/Q2 Predictions Min -87.9246 +trainer/Q Targets Mean -74.5119 +trainer/Q Targets Std 17.8178 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4487 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0236586 +trainer/policy/mean Std 0.734507 +trainer/policy/mean Max 0.999573 +trainer/policy/mean Min -0.999688 +trainer/policy/std Mean 0.411992 +trainer/policy/std Std 0.0201784 +trainer/policy/std Max 0.435872 +trainer/policy/std Min 0.377699 +trainer/Advantage Weights Mean 1.59258 +trainer/Advantage Weights Std 10.863 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.02184e-24 +trainer/Advantage Score Mean -0.570053 +trainer/Advantage Score Std 0.748882 +trainer/Advantage Score Max 1.07097 +trainer/Advantage Score Min -5.52404 +trainer/V1 Predictions Mean -74.2419 +trainer/V1 Predictions Std 17.8191 +trainer/V1 Predictions Max 0.259624 +trainer/V1 Predictions Min -87.3222 +trainer/VF Loss 0.0953455 +expl/num steps total 747000 +expl/num paths total 1002 +expl/path length Mean 500 +expl/path length Std 118 +expl/path length Max 618 +expl/path length Min 382 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0278649 +expl/Actions Std 0.826356 +expl/Actions Max 2.16673 +expl/Actions Min -2.12171 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 682716 +eval/num paths total 751 +eval/path length Mean 512 +eval/path length Std 0 +eval/path length Max 512 +eval/path length Min 512 +eval/Rewards Mean 0.00195312 +eval/Rewards Std 0.044151 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0405662 +eval/Actions Std 0.734034 +eval/Actions Max 0.999973 +eval/Actions Min -0.999856 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.32668e-06 +time/evaluation sampling (s) 4.88174 +time/exploration sampling (s) 6.97741 +time/logging (s) 0.00979105 +time/saving (s) 0.0131631 +time/training (s) 18.4287 +time/epoch (s) 30.3108 +time/total (s) 19205 +Epoch -254 +------------------------------ ---------------- +2022-05-15 23:23:05.469313 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -253 finished +------------------------------ ---------------- +epoch -253 +replay_buffer/size 999047 +trainer/num train calls 748000 +trainer/QF1 Loss 0.456257 +trainer/QF2 Loss 0.539641 +trainer/Policy Loss 6.1949 +trainer/Q1 Predictions Mean -73.5617 +trainer/Q1 Predictions Std 18.4298 +trainer/Q1 Predictions Max -0.426781 +trainer/Q1 Predictions Min -87.1349 +trainer/Q2 Predictions Mean -73.5741 +trainer/Q2 Predictions Std 18.5052 +trainer/Q2 Predictions Max 0.223589 +trainer/Q2 Predictions Min -87.2795 +trainer/Q Targets Mean -73.4755 +trainer/Q Targets Std 18.5283 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6779 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0129669 +trainer/policy/mean Std 0.730224 +trainer/policy/mean Max 0.999948 +trainer/policy/mean Min -0.999143 +trainer/policy/std Mean 0.411896 +trainer/policy/std Std 0.0197303 +trainer/policy/std Max 0.434444 +trainer/policy/std Min 0.382818 +trainer/Advantage Weights Mean 1.81257 +trainer/Advantage Weights Std 11.2104 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.4117e-14 +trainer/Advantage Score Mean -0.48769 +trainer/Advantage Score Std 0.587754 +trainer/Advantage Score Max 2.24957 +trainer/Advantage Score Min -3.13559 +trainer/V1 Predictions Mean -73.1887 +trainer/V1 Predictions Std 18.6462 +trainer/V1 Predictions Max 0.311447 +trainer/V1 Predictions Min -87.5309 +trainer/VF Loss 0.0805851 +expl/num steps total 748000 +expl/num paths total 1004 +expl/path length Mean 500 +expl/path length Std 491 +expl/path length Max 991 +expl/path length Min 9 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0135422 +expl/Actions Std 0.813253 +expl/Actions Max 2.26209 +expl/Actions Min -2.42776 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 683289 +eval/num paths total 752 +eval/path length Mean 573 +eval/path length Std 0 +eval/path length Max 573 +eval/path length Min 573 +eval/Rewards Mean 0.0017452 +eval/Rewards Std 0.0417391 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.013945 +eval/Actions Std 0.733478 +eval/Actions Max 0.999843 +eval/Actions Min -0.99907 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.05328e-05 +time/evaluation sampling (s) 4.79565 +time/exploration sampling (s) 6.69899 +time/logging (s) 0.00934134 +time/saving (s) 0.0156216 +time/training (s) 19.0602 +time/epoch (s) 30.5798 +time/total (s) 19235.6 +Epoch -253 +------------------------------ ---------------- +2022-05-15 23:23:35.425581 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -252 finished +------------------------------ ---------------- +epoch -252 +replay_buffer/size 999047 +trainer/num train calls 749000 +trainer/QF1 Loss 0.610593 +trainer/QF2 Loss 0.607758 +trainer/Policy Loss 24.991 +trainer/Q1 Predictions Mean -73.5748 +trainer/Q1 Predictions Std 18.47 +trainer/Q1 Predictions Max -0.302933 +trainer/Q1 Predictions Min -87.5671 +trainer/Q2 Predictions Mean -73.6208 +trainer/Q2 Predictions Std 18.4265 +trainer/Q2 Predictions Max 0.374447 +trainer/Q2 Predictions Min -87.3892 +trainer/Q Targets Mean -74.048 +trainer/Q Targets Std 18.3112 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8986 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0118452 +trainer/policy/mean Std 0.734619 +trainer/policy/mean Max 0.999281 +trainer/policy/mean Min -0.999113 +trainer/policy/std Mean 0.411779 +trainer/policy/std Std 0.0201121 +trainer/policy/std Max 0.436287 +trainer/policy/std Min 0.380487 +trainer/Advantage Weights Mean 7.40279 +trainer/Advantage Weights Std 22.8648 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.60564e-17 +trainer/Advantage Score Mean -0.207623 +trainer/Advantage Score Std 0.565086 +trainer/Advantage Score Max 2.10279 +trainer/Advantage Score Min -3.86704 +trainer/V1 Predictions Mean -73.7891 +trainer/V1 Predictions Std 18.3755 +trainer/V1 Predictions Max -0.63628 +trainer/V1 Predictions Min -87.7766 +trainer/VF Loss 0.0791154 +expl/num steps total 749000 +expl/num paths total 1006 +expl/path length Mean 500 +expl/path length Std 467 +expl/path length Max 967 +expl/path length Min 33 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00684357 +expl/Actions Std 0.818828 +expl/Actions Max 2.34986 +expl/Actions Min -2.34885 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 684289 +eval/num paths total 753 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.219524 +eval/Actions Std 0.767878 +eval/Actions Max 0.99998 +eval/Actions Min -0.999821 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.00508e-05 +time/evaluation sampling (s) 4.80339 +time/exploration sampling (s) 6.24035 +time/logging (s) 0.00761763 +time/saving (s) 0.0161016 +time/training (s) 18.8762 +time/epoch (s) 29.9437 +time/total (s) 19265.5 +Epoch -252 +------------------------------ ---------------- +2022-05-15 23:24:06.285542 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -251 finished +------------------------------ ---------------- +epoch -251 +replay_buffer/size 999047 +trainer/num train calls 750000 +trainer/QF1 Loss 0.461036 +trainer/QF2 Loss 0.500375 +trainer/Policy Loss 22.3709 +trainer/Q1 Predictions Mean -73.5335 +trainer/Q1 Predictions Std 18.8248 +trainer/Q1 Predictions Max -0.330665 +trainer/Q1 Predictions Min -87.2838 +trainer/Q2 Predictions Mean -73.5328 +trainer/Q2 Predictions Std 18.8154 +trainer/Q2 Predictions Max -0.147269 +trainer/Q2 Predictions Min -87.4451 +trainer/Q Targets Mean -73.5614 +trainer/Q Targets Std 18.9689 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7119 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0109239 +trainer/policy/mean Std 0.735255 +trainer/policy/mean Max 0.998996 +trainer/policy/mean Min -0.999782 +trainer/policy/std Mean 0.413243 +trainer/policy/std Std 0.0196941 +trainer/policy/std Max 0.435523 +trainer/policy/std Min 0.38151 +trainer/Advantage Weights Mean 2.58183 +trainer/Advantage Weights Std 13.0349 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.18794e-14 +trainer/Advantage Score Mean -0.404936 +trainer/Advantage Score Std 0.549556 +trainer/Advantage Score Max 0.986321 +trainer/Advantage Score Min -3.04136 +trainer/V1 Predictions Mean -73.2741 +trainer/V1 Predictions Std 18.9885 +trainer/V1 Predictions Max -0.620162 +trainer/V1 Predictions Min -87.5777 +trainer/VF Loss 0.0554183 +expl/num steps total 750000 +expl/num paths total 1008 +expl/path length Mean 500 +expl/path length Std 253 +expl/path length Max 753 +expl/path length Min 247 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0134882 +expl/Actions Std 0.845131 +expl/Actions Max 2.42213 +expl/Actions Min -2.36088 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 685289 +eval/num paths total 754 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0786805 +eval/Actions Std 0.690231 +eval/Actions Max 0.999672 +eval/Actions Min -0.999592 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.13747e-05 +time/evaluation sampling (s) 4.76264 +time/exploration sampling (s) 6.54384 +time/logging (s) 0.0150095 +time/saving (s) 0.014409 +time/training (s) 19.5155 +time/epoch (s) 30.8514 +time/total (s) 19296.4 +Epoch -251 +------------------------------ ---------------- +2022-05-15 23:24:37.081592 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -250 finished +------------------------------ ---------------- +epoch -250 +replay_buffer/size 999047 +trainer/num train calls 751000 +trainer/QF1 Loss 0.897174 +trainer/QF2 Loss 0.877563 +trainer/Policy Loss 33.2719 +trainer/Q1 Predictions Mean -74.184 +trainer/Q1 Predictions Std 17.6333 +trainer/Q1 Predictions Max -0.619153 +trainer/Q1 Predictions Min -87.9805 +trainer/Q2 Predictions Mean -74.2438 +trainer/Q2 Predictions Std 17.6161 +trainer/Q2 Predictions Max 0.0349001 +trainer/Q2 Predictions Min -87.9921 +trainer/Q Targets Mean -74.4748 +trainer/Q Targets Std 17.6403 +trainer/Q Targets Max -0.920882 +trainer/Q Targets Min -87.7094 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0103136 +trainer/policy/mean Std 0.734988 +trainer/policy/mean Max 0.998403 +trainer/policy/mean Min -0.999127 +trainer/policy/std Mean 0.411143 +trainer/policy/std Std 0.0191284 +trainer/policy/std Max 0.431736 +trainer/policy/std Min 0.380174 +trainer/Advantage Weights Mean 6.74318 +trainer/Advantage Weights Std 19.8185 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.23195e-15 +trainer/Advantage Score Mean -0.258207 +trainer/Advantage Score Std 0.614187 +trainer/Advantage Score Max 0.931908 +trainer/Advantage Score Min -3.37359 +trainer/V1 Predictions Mean -74.1437 +trainer/V1 Predictions Std 17.9888 +trainer/V1 Predictions Max -0.400594 +trainer/V1 Predictions Min -87.5865 +trainer/VF Loss 0.0670736 +expl/num steps total 751000 +expl/num paths total 1009 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.187659 +expl/Actions Std 0.82508 +expl/Actions Max 2.52061 +expl/Actions Min -2.33396 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 686289 +eval/num paths total 755 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0425415 +eval/Actions Std 0.76073 +eval/Actions Max 0.999659 +eval/Actions Min -0.999742 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.00802e-05 +time/evaluation sampling (s) 4.80761 +time/exploration sampling (s) 6.99009 +time/logging (s) 0.00836278 +time/saving (s) 0.0233591 +time/training (s) 18.947 +time/epoch (s) 30.7764 +time/total (s) 19327.2 +Epoch -250 +------------------------------ ---------------- +2022-05-15 23:25:08.010731 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -249 finished +------------------------------ ---------------- +epoch -249 +replay_buffer/size 999047 +trainer/num train calls 752000 +trainer/QF1 Loss 1.3628 +trainer/QF2 Loss 1.25261 +trainer/Policy Loss 17.0381 +trainer/Q1 Predictions Mean -74.8341 +trainer/Q1 Predictions Std 17.4095 +trainer/Q1 Predictions Max -1.31292 +trainer/Q1 Predictions Min -87.8849 +trainer/Q2 Predictions Mean -74.791 +trainer/Q2 Predictions Std 17.4328 +trainer/Q2 Predictions Max -0.518681 +trainer/Q2 Predictions Min -87.7467 +trainer/Q Targets Mean -75.2186 +trainer/Q Targets Std 17.1676 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9804 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0363628 +trainer/policy/mean Std 0.73068 +trainer/policy/mean Max 0.99984 +trainer/policy/mean Min -0.99992 +trainer/policy/std Mean 0.409718 +trainer/policy/std Std 0.0185694 +trainer/policy/std Max 0.430057 +trainer/policy/std Min 0.381015 +trainer/Advantage Weights Mean 4.47887 +trainer/Advantage Weights Std 17.3984 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.34258e-17 +trainer/Advantage Score Mean -0.254377 +trainer/Advantage Score Std 0.571242 +trainer/Advantage Score Max 1.5151 +trainer/Advantage Score Min -3.82927 +trainer/V1 Predictions Mean -74.8824 +trainer/V1 Predictions Std 17.3418 +trainer/V1 Predictions Max -0.0727636 +trainer/V1 Predictions Min -87.7928 +trainer/VF Loss 0.065618 +expl/num steps total 752000 +expl/num paths total 1010 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0351333 +expl/Actions Std 0.87375 +expl/Actions Max 2.4371 +expl/Actions Min -2.36889 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 687146 +eval/num paths total 756 +eval/path length Mean 857 +eval/path length Std 0 +eval/path length Max 857 +eval/path length Min 857 +eval/Rewards Mean 0.00116686 +eval/Rewards Std 0.0341394 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00391931 +eval/Actions Std 0.758725 +eval/Actions Max 0.999799 +eval/Actions Min -0.999716 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.22208e-05 +time/evaluation sampling (s) 4.97313 +time/exploration sampling (s) 7.37062 +time/logging (s) 0.0072918 +time/saving (s) 0.0105121 +time/training (s) 18.5487 +time/epoch (s) 30.9103 +time/total (s) 19358.1 +Epoch -249 +------------------------------ ---------------- +2022-05-15 23:25:38.982373 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -248 finished +------------------------------ ---------------- +epoch -248 +replay_buffer/size 999047 +trainer/num train calls 753000 +trainer/QF1 Loss 0.362905 +trainer/QF2 Loss 0.294014 +trainer/Policy Loss 19.7827 +trainer/Q1 Predictions Mean -74.4038 +trainer/Q1 Predictions Std 18.4099 +trainer/Q1 Predictions Max -0.47686 +trainer/Q1 Predictions Min -87.6234 +trainer/Q2 Predictions Mean -74.4723 +trainer/Q2 Predictions Std 18.5397 +trainer/Q2 Predictions Max 0.478752 +trainer/Q2 Predictions Min -87.8896 +trainer/Q Targets Mean -74.517 +trainer/Q Targets Std 18.5371 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8119 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0109111 +trainer/policy/mean Std 0.72988 +trainer/policy/mean Max 0.999873 +trainer/policy/mean Min -0.999688 +trainer/policy/std Mean 0.410323 +trainer/policy/std Std 0.0198661 +trainer/policy/std Max 0.433013 +trainer/policy/std Min 0.379241 +trainer/Advantage Weights Mean 3.01349 +trainer/Advantage Weights Std 12.9565 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.45744e-12 +trainer/Advantage Score Mean -0.341411 +trainer/Advantage Score Std 0.516081 +trainer/Advantage Score Max 0.641964 +trainer/Advantage Score Min -2.72543 +trainer/V1 Predictions Mean -74.2589 +trainer/V1 Predictions Std 18.7347 +trainer/V1 Predictions Max 1.00007 +trainer/V1 Predictions Min -87.8535 +trainer/VF Loss 0.0465246 +expl/num steps total 753000 +expl/num paths total 1012 +expl/path length Mean 500 +expl/path length Std 375 +expl/path length Max 875 +expl/path length Min 125 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0301458 +expl/Actions Std 0.825699 +expl/Actions Max 2.35337 +expl/Actions Min -2.29171 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 688146 +eval/num paths total 757 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0981405 +eval/Actions Std 0.705926 +eval/Actions Max 0.99976 +eval/Actions Min -0.999065 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.29337e-05 +time/evaluation sampling (s) 5.19476 +time/exploration sampling (s) 6.68747 +time/logging (s) 0.0111631 +time/saving (s) 0.0154378 +time/training (s) 19.0547 +time/epoch (s) 30.9635 +time/total (s) 19389 +Epoch -248 +------------------------------ ---------------- +2022-05-15 23:26:09.244492 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -247 finished +------------------------------ ---------------- +epoch -247 +replay_buffer/size 999047 +trainer/num train calls 754000 +trainer/QF1 Loss 0.839335 +trainer/QF2 Loss 0.766415 +trainer/Policy Loss 18.0304 +trainer/Q1 Predictions Mean -74.7318 +trainer/Q1 Predictions Std 17.6912 +trainer/Q1 Predictions Max -0.0638241 +trainer/Q1 Predictions Min -87.4046 +trainer/Q2 Predictions Mean -74.7918 +trainer/Q2 Predictions Std 17.7637 +trainer/Q2 Predictions Max 1.0182 +trainer/Q2 Predictions Min -87.5994 +trainer/Q Targets Mean -74.6597 +trainer/Q Targets Std 17.8909 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5338 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00113046 +trainer/policy/mean Std 0.723462 +trainer/policy/mean Max 0.99911 +trainer/policy/mean Min -0.999805 +trainer/policy/std Mean 0.410823 +trainer/policy/std Std 0.0188528 +trainer/policy/std Max 0.434118 +trainer/policy/std Min 0.382364 +trainer/Advantage Weights Mean 4.40519 +trainer/Advantage Weights Std 16.8028 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.75838e-17 +trainer/Advantage Score Mean -0.383734 +trainer/Advantage Score Std 0.621287 +trainer/Advantage Score Max 1.13036 +trainer/Advantage Score Min -3.782 +trainer/V1 Predictions Mean -74.4172 +trainer/V1 Predictions Std 17.8771 +trainer/V1 Predictions Max 0.254275 +trainer/V1 Predictions Min -87.3895 +trainer/VF Loss 0.0728058 +expl/num steps total 754000 +expl/num paths total 1014 +expl/path length Mean 500 +expl/path length Std 107 +expl/path length Max 607 +expl/path length Min 393 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0154425 +expl/Actions Std 0.816982 +expl/Actions Max 2.48675 +expl/Actions Min -2.42747 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 689146 +eval/num paths total 758 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0519386 +eval/Actions Std 0.640024 +eval/Actions Max 0.999598 +eval/Actions Min -0.999424 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.13971e-05 +time/evaluation sampling (s) 4.97963 +time/exploration sampling (s) 6.34923 +time/logging (s) 0.0116255 +time/saving (s) 0.0151657 +time/training (s) 18.8878 +time/epoch (s) 30.2435 +time/total (s) 19419.3 +Epoch -247 +------------------------------ ---------------- +2022-05-15 23:26:39.795632 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -246 finished +------------------------------ ---------------- +epoch -246 +replay_buffer/size 999047 +trainer/num train calls 755000 +trainer/QF1 Loss 0.536604 +trainer/QF2 Loss 0.55776 +trainer/Policy Loss 7.99167 +trainer/Q1 Predictions Mean -75.4357 +trainer/Q1 Predictions Std 16.9591 +trainer/Q1 Predictions Max -0.662876 +trainer/Q1 Predictions Min -87.7191 +trainer/Q2 Predictions Mean -75.3705 +trainer/Q2 Predictions Std 16.936 +trainer/Q2 Predictions Max -0.116624 +trainer/Q2 Predictions Min -87.6331 +trainer/Q Targets Mean -75.1516 +trainer/Q Targets Std 16.8567 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.2656 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00577325 +trainer/policy/mean Std 0.73473 +trainer/policy/mean Max 0.999928 +trainer/policy/mean Min -0.999505 +trainer/policy/std Mean 0.413058 +trainer/policy/std Std 0.018675 +trainer/policy/std Max 0.434512 +trainer/policy/std Min 0.384313 +trainer/Advantage Weights Mean 1.23247 +trainer/Advantage Weights Std 9.23232 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.4247e-17 +trainer/Advantage Score Mean -0.578593 +trainer/Advantage Score Std 0.578773 +trainer/Advantage Score Max 0.602753 +trainer/Advantage Score Min -3.7453 +trainer/V1 Predictions Mean -74.8629 +trainer/V1 Predictions Std 17.1246 +trainer/V1 Predictions Max -0.476783 +trainer/V1 Predictions Min -87.1785 +trainer/VF Loss 0.0703008 +expl/num steps total 755000 +expl/num paths total 1015 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0333618 +expl/Actions Std 0.833564 +expl/Actions Max 2.45553 +expl/Actions Min -2.37646 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 690146 +eval/num paths total 759 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.109992 +eval/Actions Std 0.788227 +eval/Actions Max 0.998951 +eval/Actions Min -0.99648 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.29014e-06 +time/evaluation sampling (s) 5.14732 +time/exploration sampling (s) 6.60123 +time/logging (s) 0.0122349 +time/saving (s) 0.0174536 +time/training (s) 18.7553 +time/epoch (s) 30.5336 +time/total (s) 19449.8 +Epoch -246 +------------------------------ ---------------- +2022-05-15 23:27:10.745257 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -245 finished +------------------------------ ---------------- +epoch -245 +replay_buffer/size 999047 +trainer/num train calls 756000 +trainer/QF1 Loss 1.05653 +trainer/QF2 Loss 1.19722 +trainer/Policy Loss 1.02504 +trainer/Q1 Predictions Mean -73.9828 +trainer/Q1 Predictions Std 17.9975 +trainer/Q1 Predictions Max -1.84444 +trainer/Q1 Predictions Min -88.2897 +trainer/Q2 Predictions Mean -74.0752 +trainer/Q2 Predictions Std 18.0267 +trainer/Q2 Predictions Max -1.15187 +trainer/Q2 Predictions Min -88.3399 +trainer/Q Targets Mean -73.4172 +trainer/Q Targets Std 17.7054 +trainer/Q Targets Max -2.55154 +trainer/Q Targets Min -87.3499 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000578692 +trainer/policy/mean Std 0.724914 +trainer/policy/mean Max 0.999279 +trainer/policy/mean Min -0.999221 +trainer/policy/std Mean 0.413244 +trainer/policy/std Std 0.0188858 +trainer/policy/std Max 0.432625 +trainer/policy/std Min 0.382387 +trainer/Advantage Weights Mean 0.456678 +trainer/Advantage Weights Std 6.24063 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.1488e-24 +trainer/Advantage Score Mean -0.749983 +trainer/Advantage Score Std 0.664737 +trainer/Advantage Score Max 1.44296 +trainer/Advantage Score Min -5.32951 +trainer/V1 Predictions Mean -73.0863 +trainer/V1 Predictions Std 17.9548 +trainer/V1 Predictions Max -1.62121 +trainer/V1 Predictions Min -87.2176 +trainer/VF Loss 0.107017 +expl/num steps total 756000 +expl/num paths total 1017 +expl/path length Mean 500 +expl/path length Std 165 +expl/path length Max 665 +expl/path length Min 335 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0195379 +expl/Actions Std 0.822877 +expl/Actions Max 2.19403 +expl/Actions Min -2.38206 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 691146 +eval/num paths total 760 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.229914 +eval/Actions Std 0.69874 +eval/Actions Max 0.999652 +eval/Actions Min -0.999731 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.31689e-05 +time/evaluation sampling (s) 5.06864 +time/exploration sampling (s) 7.04912 +time/logging (s) 0.0135655 +time/saving (s) 0.0174357 +time/training (s) 18.783 +time/epoch (s) 30.9318 +time/total (s) 19480.8 +Epoch -245 +------------------------------ ---------------- +2022-05-15 23:27:41.115782 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -244 finished +------------------------------ ---------------- +epoch -244 +replay_buffer/size 999047 +trainer/num train calls 757000 +trainer/QF1 Loss 0.745637 +trainer/QF2 Loss 0.744542 +trainer/Policy Loss 17.6871 +trainer/Q1 Predictions Mean -74.9916 +trainer/Q1 Predictions Std 16.3642 +trainer/Q1 Predictions Max -1.36687 +trainer/Q1 Predictions Min -88.3498 +trainer/Q2 Predictions Mean -75.0137 +trainer/Q2 Predictions Std 16.3784 +trainer/Q2 Predictions Max -1.27099 +trainer/Q2 Predictions Min -88.426 +trainer/Q Targets Mean -74.5607 +trainer/Q Targets Std 16.2037 +trainer/Q Targets Max -3.4887 +trainer/Q Targets Min -87.4001 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.014211 +trainer/policy/mean Std 0.727228 +trainer/policy/mean Max 0.999288 +trainer/policy/mean Min -0.998169 +trainer/policy/std Mean 0.412724 +trainer/policy/std Std 0.0184328 +trainer/policy/std Max 0.434582 +trainer/policy/std Min 0.38159 +trainer/Advantage Weights Mean 2.44237 +trainer/Advantage Weights Std 13.303 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.36272e-22 +trainer/Advantage Score Mean -0.538507 +trainer/Advantage Score Std 0.609806 +trainer/Advantage Score Max 0.942368 +trainer/Advantage Score Min -4.94441 +trainer/V1 Predictions Mean -74.3148 +trainer/V1 Predictions Std 16.3531 +trainer/V1 Predictions Max -1.57017 +trainer/V1 Predictions Min -87.2757 +trainer/VF Loss 0.0753479 +expl/num steps total 757000 +expl/num paths total 1018 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0290736 +expl/Actions Std 0.82994 +expl/Actions Max 2.39765 +expl/Actions Min -2.18014 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 691660 +eval/num paths total 761 +eval/path length Mean 514 +eval/path length Std 0 +eval/path length Max 514 +eval/path length Min 514 +eval/Rewards Mean 0.00194553 +eval/Rewards Std 0.0440652 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0401373 +eval/Actions Std 0.764465 +eval/Actions Max 0.999478 +eval/Actions Min -0.998981 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.61027e-06 +time/evaluation sampling (s) 4.6469 +time/exploration sampling (s) 6.87462 +time/logging (s) 0.00958016 +time/saving (s) 0.0183655 +time/training (s) 18.7953 +time/epoch (s) 30.3447 +time/total (s) 19511.1 +Epoch -244 +------------------------------ ---------------- +2022-05-15 23:28:11.991986 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -243 finished +------------------------------ ---------------- +epoch -243 +replay_buffer/size 999047 +trainer/num train calls 758000 +trainer/QF1 Loss 0.778695 +trainer/QF2 Loss 0.809045 +trainer/Policy Loss 8.66959 +trainer/Q1 Predictions Mean -72.8545 +trainer/Q1 Predictions Std 20.6314 +trainer/Q1 Predictions Max -0.361265 +trainer/Q1 Predictions Min -88.1482 +trainer/Q2 Predictions Mean -72.8812 +trainer/Q2 Predictions Std 20.5033 +trainer/Q2 Predictions Max 0.0284796 +trainer/Q2 Predictions Min -88.1984 +trainer/Q Targets Mean -72.5733 +trainer/Q Targets Std 20.7869 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.725 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00584861 +trainer/policy/mean Std 0.725375 +trainer/policy/mean Max 0.999688 +trainer/policy/mean Min -0.998936 +trainer/policy/std Mean 0.411389 +trainer/policy/std Std 0.0207939 +trainer/policy/std Max 0.434426 +trainer/policy/std Min 0.377194 +trainer/Advantage Weights Mean 1.93284 +trainer/Advantage Weights Std 12.503 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.33632e-25 +trainer/Advantage Score Mean -0.595634 +trainer/Advantage Score Std 0.733393 +trainer/Advantage Score Max 1.01493 +trainer/Advantage Score Min -5.53307 +trainer/V1 Predictions Mean -72.3216 +trainer/V1 Predictions Std 20.8721 +trainer/V1 Predictions Max 0.569667 +trainer/V1 Predictions Min -87.6051 +trainer/VF Loss 0.0964501 +expl/num steps total 758000 +expl/num paths total 1019 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0594561 +expl/Actions Std 0.825665 +expl/Actions Max 2.54851 +expl/Actions Min -2.26043 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 692144 +eval/num paths total 762 +eval/path length Mean 484 +eval/path length Std 0 +eval/path length Max 484 +eval/path length Min 484 +eval/Rewards Mean 0.00206612 +eval/Rewards Std 0.0454076 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00724166 +eval/Actions Std 0.736598 +eval/Actions Max 0.999046 +eval/Actions Min -0.999877 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.19824e-05 +time/evaluation sampling (s) 5.13609 +time/exploration sampling (s) 6.68456 +time/logging (s) 0.0102638 +time/saving (s) 0.0186072 +time/training (s) 19.0113 +time/epoch (s) 30.8609 +time/total (s) 19542 +Epoch -243 +------------------------------ ---------------- +2022-05-15 23:28:41.813137 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -242 finished +------------------------------ ---------------- +epoch -242 +replay_buffer/size 999047 +trainer/num train calls 759000 +trainer/QF1 Loss 1.07074 +trainer/QF2 Loss 0.892089 +trainer/Policy Loss 20.7165 +trainer/Q1 Predictions Mean -74.7015 +trainer/Q1 Predictions Std 18.8952 +trainer/Q1 Predictions Max -0.345431 +trainer/Q1 Predictions Min -87.7022 +trainer/Q2 Predictions Mean -74.6316 +trainer/Q2 Predictions Std 18.8726 +trainer/Q2 Predictions Max -1.21938 +trainer/Q2 Predictions Min -87.5246 +trainer/Q Targets Mean -74.7088 +trainer/Q Targets Std 19.0655 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5856 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0446916 +trainer/policy/mean Std 0.724849 +trainer/policy/mean Max 0.999406 +trainer/policy/mean Min -0.999669 +trainer/policy/std Mean 0.411594 +trainer/policy/std Std 0.0185327 +trainer/policy/std Max 0.433424 +trainer/policy/std Min 0.381407 +trainer/Advantage Weights Mean 4.00648 +trainer/Advantage Weights Std 14.3394 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.14231e-21 +trainer/Advantage Score Mean -0.274325 +trainer/Advantage Score Std 0.668398 +trainer/Advantage Score Max 1.64022 +trainer/Advantage Score Min -4.67168 +trainer/V1 Predictions Mean -74.5081 +trainer/V1 Predictions Std 19.1608 +trainer/V1 Predictions Max 0.157417 +trainer/V1 Predictions Min -87.5159 +trainer/VF Loss 0.0704481 +expl/num steps total 759000 +expl/num paths total 1020 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0145032 +expl/Actions Std 0.830187 +expl/Actions Max 2.49668 +expl/Actions Min -2.18727 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 692724 +eval/num paths total 763 +eval/path length Mean 580 +eval/path length Std 0 +eval/path length Max 580 +eval/path length Min 580 +eval/Rewards Mean 0.00172414 +eval/Rewards Std 0.0414869 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00102197 +eval/Actions Std 0.728778 +eval/Actions Max 0.999525 +eval/Actions Min -0.999496 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.49479e-06 +time/evaluation sampling (s) 4.82742 +time/exploration sampling (s) 6.44614 +time/logging (s) 0.0108092 +time/saving (s) 0.0181831 +time/training (s) 18.4986 +time/epoch (s) 29.8011 +time/total (s) 19571.8 +Epoch -242 +------------------------------ ---------------- +2022-05-15 23:29:12.664283 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -241 finished +------------------------------ ---------------- +epoch -241 +replay_buffer/size 999047 +trainer/num train calls 760000 +trainer/QF1 Loss 2.05963 +trainer/QF2 Loss 2.22628 +trainer/Policy Loss 18.549 +trainer/Q1 Predictions Mean -75.9703 +trainer/Q1 Predictions Std 15.0496 +trainer/Q1 Predictions Max -4.31158 +trainer/Q1 Predictions Min -87.1502 +trainer/Q2 Predictions Mean -76.0306 +trainer/Q2 Predictions Std 15.0266 +trainer/Q2 Predictions Max -3.03873 +trainer/Q2 Predictions Min -87.2313 +trainer/Q Targets Mean -76.1474 +trainer/Q Targets Std 15.1442 +trainer/Q Targets Max -3.58703 +trainer/Q Targets Min -88.0174 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0276381 +trainer/policy/mean Std 0.744262 +trainer/policy/mean Max 0.999764 +trainer/policy/mean Min -0.999369 +trainer/policy/std Mean 0.410407 +trainer/policy/std Std 0.0204764 +trainer/policy/std Max 0.436565 +trainer/policy/std Min 0.377626 +trainer/Advantage Weights Mean 4.20586 +trainer/Advantage Weights Std 16.1397 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.26273e-13 +trainer/Advantage Score Mean -0.256644 +trainer/Advantage Score Std 0.515079 +trainer/Advantage Score Max 1.53397 +trainer/Advantage Score Min -2.84837 +trainer/V1 Predictions Mean -76.0387 +trainer/V1 Predictions Std 15.1801 +trainer/V1 Predictions Max -2.47699 +trainer/V1 Predictions Min -87.3713 +trainer/VF Loss 0.0519955 +expl/num steps total 760000 +expl/num paths total 1022 +expl/path length Mean 500 +expl/path length Std 94 +expl/path length Max 594 +expl/path length Min 406 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.0692451 +expl/Actions Std 0.847442 +expl/Actions Max 2.30298 +expl/Actions Min -2.72236 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 693724 +eval/num paths total 764 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00892193 +eval/Actions Std 0.746046 +eval/Actions Max 0.999615 +eval/Actions Min -0.9992 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.97697e-06 +time/evaluation sampling (s) 5.48089 +time/exploration sampling (s) 6.69176 +time/logging (s) 0.0118501 +time/saving (s) 0.0153023 +time/training (s) 18.6338 +time/epoch (s) 30.8336 +time/total (s) 19602.6 +Epoch -241 +------------------------------ ---------------- +2022-05-15 23:29:43.121169 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -240 finished +------------------------------ ---------------- +epoch -240 +replay_buffer/size 999047 +trainer/num train calls 761000 +trainer/QF1 Loss 0.804236 +trainer/QF2 Loss 0.85304 +trainer/Policy Loss 4.6032 +trainer/Q1 Predictions Mean -75.0974 +trainer/Q1 Predictions Std 16.9496 +trainer/Q1 Predictions Max -1.55897 +trainer/Q1 Predictions Min -87.8843 +trainer/Q2 Predictions Mean -75.1105 +trainer/Q2 Predictions Std 16.9245 +trainer/Q2 Predictions Max -1.80428 +trainer/Q2 Predictions Min -88.1266 +trainer/Q Targets Mean -74.9027 +trainer/Q Targets Std 16.7303 +trainer/Q Targets Max -2.71252 +trainer/Q Targets Min -87.6544 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00829899 +trainer/policy/mean Std 0.732631 +trainer/policy/mean Max 0.999655 +trainer/policy/mean Min -0.999795 +trainer/policy/std Mean 0.410023 +trainer/policy/std Std 0.0187587 +trainer/policy/std Max 0.430493 +trainer/policy/std Min 0.379867 +trainer/Advantage Weights Mean 1.19712 +trainer/Advantage Weights Std 8.94842 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.68975e-27 +trainer/Advantage Score Mean -0.5544 +trainer/Advantage Score Std 0.650482 +trainer/Advantage Score Max 0.607567 +trainer/Advantage Score Min -6.01299 +trainer/V1 Predictions Mean -74.5859 +trainer/V1 Predictions Std 16.9133 +trainer/V1 Predictions Max -0.674812 +trainer/V1 Predictions Min -87.3325 +trainer/VF Loss 0.0762934 +expl/num steps total 761000 +expl/num paths total 1024 +expl/path length Mean 500 +expl/path length Std 203 +expl/path length Max 703 +expl/path length Min 297 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0198933 +expl/Actions Std 0.848537 +expl/Actions Max 2.24916 +expl/Actions Min -2.24342 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 694360 +eval/num paths total 765 +eval/path length Mean 636 +eval/path length Std 0 +eval/path length Max 636 +eval/path length Min 636 +eval/Rewards Mean 0.00157233 +eval/Rewards Std 0.0396214 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0388939 +eval/Actions Std 0.741566 +eval/Actions Max 0.99997 +eval/Actions Min -0.999524 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.4096e-05 +time/evaluation sampling (s) 5.19413 +time/exploration sampling (s) 6.37305 +time/logging (s) 0.0109019 +time/saving (s) 0.0187293 +time/training (s) 18.8397 +time/epoch (s) 30.4366 +time/total (s) 19633.1 +Epoch -240 +------------------------------ ---------------- +2022-05-15 23:30:13.829730 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -239 finished +------------------------------ ---------------- +epoch -239 +replay_buffer/size 999047 +trainer/num train calls 762000 +trainer/QF1 Loss 0.5705 +trainer/QF2 Loss 0.645678 +trainer/Policy Loss 23.8388 +trainer/Q1 Predictions Mean -75.2603 +trainer/Q1 Predictions Std 15.7495 +trainer/Q1 Predictions Max -1.49425 +trainer/Q1 Predictions Min -87.6869 +trainer/Q2 Predictions Mean -75.1921 +trainer/Q2 Predictions Std 15.8358 +trainer/Q2 Predictions Max -0.759786 +trainer/Q2 Predictions Min -87.4711 +trainer/Q Targets Mean -75.3995 +trainer/Q Targets Std 15.9226 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0462 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0216302 +trainer/policy/mean Std 0.729383 +trainer/policy/mean Max 0.999587 +trainer/policy/mean Min -0.999247 +trainer/policy/std Mean 0.409746 +trainer/policy/std Std 0.0187269 +trainer/policy/std Max 0.430541 +trainer/policy/std Min 0.379735 +trainer/Advantage Weights Mean 6.66002 +trainer/Advantage Weights Std 20.506 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.48391e-20 +trainer/Advantage Score Mean -0.277185 +trainer/Advantage Score Std 0.686821 +trainer/Advantage Score Max 1.52038 +trainer/Advantage Score Min -4.51419 +trainer/V1 Predictions Mean -75.1482 +trainer/V1 Predictions Std 16.0921 +trainer/V1 Predictions Max -0.195663 +trainer/V1 Predictions Min -87.8581 +trainer/VF Loss 0.0816846 +expl/num steps total 762000 +expl/num paths total 1026 +expl/path length Mean 500 +expl/path length Std 466 +expl/path length Max 966 +expl/path length Min 34 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0427537 +expl/Actions Std 0.830049 +expl/Actions Max 2.49452 +expl/Actions Min -2.7034 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 695360 +eval/num paths total 766 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.047027 +eval/Actions Std 0.747664 +eval/Actions Max 0.999957 +eval/Actions Min -0.999895 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.20178e-05 +time/evaluation sampling (s) 5.23495 +time/exploration sampling (s) 6.55555 +time/logging (s) 0.0110495 +time/saving (s) 0.0160015 +time/training (s) 18.8709 +time/epoch (s) 30.6885 +time/total (s) 19663.8 +Epoch -239 +------------------------------ ---------------- +2022-05-15 23:30:45.191125 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -238 finished +------------------------------ ---------------- +epoch -238 +replay_buffer/size 999047 +trainer/num train calls 763000 +trainer/QF1 Loss 0.877081 +trainer/QF2 Loss 0.714781 +trainer/Policy Loss 13.8329 +trainer/Q1 Predictions Mean -74.2357 +trainer/Q1 Predictions Std 19.615 +trainer/Q1 Predictions Max -0.594551 +trainer/Q1 Predictions Min -88.0798 +trainer/Q2 Predictions Mean -74.1744 +trainer/Q2 Predictions Std 19.721 +trainer/Q2 Predictions Max -0.0701014 +trainer/Q2 Predictions Min -87.8866 +trainer/Q Targets Mean -74.2123 +trainer/Q Targets Std 19.8448 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8559 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.018292 +trainer/policy/mean Std 0.736699 +trainer/policy/mean Max 0.999157 +trainer/policy/mean Min -0.997784 +trainer/policy/std Mean 0.409543 +trainer/policy/std Std 0.0192568 +trainer/policy/std Max 0.429416 +trainer/policy/std Min 0.377953 +trainer/Advantage Weights Mean 3.50915 +trainer/Advantage Weights Std 13.6778 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.55272e-21 +trainer/Advantage Score Mean -0.361457 +trainer/Advantage Score Std 0.715142 +trainer/Advantage Score Max 1.19823 +trainer/Advantage Score Min -4.79143 +trainer/V1 Predictions Mean -73.9628 +trainer/V1 Predictions Std 20.0145 +trainer/V1 Predictions Max 0.18848 +trainer/V1 Predictions Min -87.7332 +trainer/VF Loss 0.0775299 +expl/num steps total 763000 +expl/num paths total 1027 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.224455 +expl/Actions Std 0.795436 +expl/Actions Max 2.50837 +expl/Actions Min -2.43308 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 696360 +eval/num paths total 767 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.259563 +eval/Actions Std 0.603426 +eval/Actions Max 0.99908 +eval/Actions Min -0.998273 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.25421e-05 +time/evaluation sampling (s) 5.29842 +time/exploration sampling (s) 6.68136 +time/logging (s) 0.0114066 +time/saving (s) 0.0161007 +time/training (s) 19.3368 +time/epoch (s) 31.3441 +time/total (s) 19695.1 +Epoch -238 +------------------------------ ---------------- +2022-05-15 23:31:15.489190 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -237 finished +------------------------------ ---------------- +epoch -237 +replay_buffer/size 999047 +trainer/num train calls 764000 +trainer/QF1 Loss 0.529486 +trainer/QF2 Loss 0.62002 +trainer/Policy Loss 33.044 +trainer/Q1 Predictions Mean -74.682 +trainer/Q1 Predictions Std 17.2708 +trainer/Q1 Predictions Max -0.633092 +trainer/Q1 Predictions Min -88.3204 +trainer/Q2 Predictions Mean -74.6475 +trainer/Q2 Predictions Std 17.3154 +trainer/Q2 Predictions Max -0.402649 +trainer/Q2 Predictions Min -88.0793 +trainer/Q Targets Mean -74.8259 +trainer/Q Targets Std 17.3666 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.3826 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00684643 +trainer/policy/mean Std 0.733752 +trainer/policy/mean Max 0.999177 +trainer/policy/mean Min -0.999548 +trainer/policy/std Mean 0.409226 +trainer/policy/std Std 0.017799 +trainer/policy/std Max 0.429348 +trainer/policy/std Min 0.37944 +trainer/Advantage Weights Mean 6.84152 +trainer/Advantage Weights Std 22.0819 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.60425e-14 +trainer/Advantage Score Mean -0.290707 +trainer/Advantage Score Std 0.636438 +trainer/Advantage Score Max 1.32577 +trainer/Advantage Score Min -3.17635 +trainer/V1 Predictions Mean -74.6109 +trainer/V1 Predictions Std 17.4764 +trainer/V1 Predictions Max 1.59388 +trainer/V1 Predictions Min -88.3622 +trainer/VF Loss 0.0774064 +expl/num steps total 764000 +expl/num paths total 1029 +expl/path length Mean 500 +expl/path length Std 299 +expl/path length Max 799 +expl/path length Min 201 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0323931 +expl/Actions Std 0.823565 +expl/Actions Max 2.18021 +expl/Actions Min -2.19585 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 697360 +eval/num paths total 768 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0429046 +eval/Actions Std 0.725901 +eval/Actions Max 0.999898 +eval/Actions Min -0.999818 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.43496e-06 +time/evaluation sampling (s) 4.79648 +time/exploration sampling (s) 6.29559 +time/logging (s) 0.0129363 +time/saving (s) 0.0185616 +time/training (s) 19.159 +time/epoch (s) 30.2825 +time/total (s) 19725.4 +Epoch -237 +------------------------------ ---------------- +2022-05-15 23:31:45.018248 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -236 finished +------------------------------ ---------------- +epoch -236 +replay_buffer/size 999047 +trainer/num train calls 765000 +trainer/QF1 Loss 0.418059 +trainer/QF2 Loss 0.461007 +trainer/Policy Loss 6.64767 +trainer/Q1 Predictions Mean -76.4769 +trainer/Q1 Predictions Std 14.7976 +trainer/Q1 Predictions Max -3.38933 +trainer/Q1 Predictions Min -87.5645 +trainer/Q2 Predictions Mean -76.4615 +trainer/Q2 Predictions Std 14.6895 +trainer/Q2 Predictions Max -3.21061 +trainer/Q2 Predictions Min -87.4915 +trainer/Q Targets Mean -76.4107 +trainer/Q Targets Std 14.8087 +trainer/Q Targets Max -4.20081 +trainer/Q Targets Min -87.8749 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00440762 +trainer/policy/mean Std 0.724267 +trainer/policy/mean Max 0.998229 +trainer/policy/mean Min -0.998153 +trainer/policy/std Mean 0.409347 +trainer/policy/std Std 0.0184138 +trainer/policy/std Max 0.428702 +trainer/policy/std Min 0.37988 +trainer/Advantage Weights Mean 1.89572 +trainer/Advantage Weights Std 8.38145 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.64894e-10 +trainer/Advantage Score Mean -0.358793 +trainer/Advantage Score Std 0.470927 +trainer/Advantage Score Max 0.685516 +trainer/Advantage Score Min -2.17314 +trainer/V1 Predictions Mean -76.1576 +trainer/V1 Predictions Std 14.9213 +trainer/V1 Predictions Max -2.97003 +trainer/V1 Predictions Min -87.772 +trainer/VF Loss 0.0408364 +expl/num steps total 765000 +expl/num paths total 1031 +expl/path length Mean 500 +expl/path length Std 239 +expl/path length Max 739 +expl/path length Min 261 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0269029 +expl/Actions Std 0.824731 +expl/Actions Max 2.20375 +expl/Actions Min -2.23629 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 698040 +eval/num paths total 769 +eval/path length Mean 680 +eval/path length Std 0 +eval/path length Max 680 +eval/path length Min 680 +eval/Rewards Mean 0.00147059 +eval/Rewards Std 0.03832 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0396026 +eval/Actions Std 0.71527 +eval/Actions Max 0.999958 +eval/Actions Min -0.999956 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.01307e-06 +time/evaluation sampling (s) 4.56408 +time/exploration sampling (s) 6.62398 +time/logging (s) 0.0113999 +time/saving (s) 0.0181437 +time/training (s) 18.2893 +time/epoch (s) 29.5069 +time/total (s) 19754.9 +Epoch -236 +------------------------------ ---------------- +2022-05-15 23:32:14.950051 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -235 finished +------------------------------ ---------------- +epoch -235 +replay_buffer/size 999047 +trainer/num train calls 766000 +trainer/QF1 Loss 1.01111 +trainer/QF2 Loss 0.922224 +trainer/Policy Loss 9.56486 +trainer/Q1 Predictions Mean -70.4925 +trainer/Q1 Predictions Std 23.0359 +trainer/Q1 Predictions Max -0.222121 +trainer/Q1 Predictions Min -87.8397 +trainer/Q2 Predictions Mean -70.4765 +trainer/Q2 Predictions Std 22.99 +trainer/Q2 Predictions Max -0.013217 +trainer/Q2 Predictions Min -87.9583 +trainer/Q Targets Mean -70.4108 +trainer/Q Targets Std 22.8338 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7383 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0163094 +trainer/policy/mean Std 0.736645 +trainer/policy/mean Max 0.999788 +trainer/policy/mean Min -0.999916 +trainer/policy/std Mean 0.410638 +trainer/policy/std Std 0.0205893 +trainer/policy/std Max 0.43306 +trainer/policy/std Min 0.377282 +trainer/Advantage Weights Mean 3.38009 +trainer/Advantage Weights Std 15.7272 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.53952e-16 +trainer/Advantage Score Mean -0.462551 +trainer/Advantage Score Std 0.642134 +trainer/Advantage Score Max 2.66187 +trainer/Advantage Score Min -3.59094 +trainer/V1 Predictions Mean -70.1716 +trainer/V1 Predictions Std 22.9326 +trainer/V1 Predictions Max -0.09128 +trainer/V1 Predictions Min -87.2924 +trainer/VF Loss 0.11505 +expl/num steps total 766000 +expl/num paths total 1032 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0548264 +expl/Actions Std 0.875019 +expl/Actions Max 2.25779 +expl/Actions Min -2.33412 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 698617 +eval/num paths total 770 +eval/path length Mean 577 +eval/path length Std 0 +eval/path length Max 577 +eval/path length Min 577 +eval/Rewards Mean 0.0017331 +eval/Rewards Std 0.0415945 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00211437 +eval/Actions Std 0.747145 +eval/Actions Max 0.999588 +eval/Actions Min -0.999367 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.23098e-05 +time/evaluation sampling (s) 4.33597 +time/exploration sampling (s) 6.73704 +time/logging (s) 0.0109693 +time/saving (s) 0.0188857 +time/training (s) 18.8078 +time/epoch (s) 29.9107 +time/total (s) 19784.8 +Epoch -235 +------------------------------ ---------------- +2022-05-15 23:32:45.162774 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -234 finished +------------------------------ ---------------- +epoch -234 +replay_buffer/size 999047 +trainer/num train calls 767000 +trainer/QF1 Loss 1.33724 +trainer/QF2 Loss 1.39934 +trainer/Policy Loss 58.9012 +trainer/Q1 Predictions Mean -75.5409 +trainer/Q1 Predictions Std 15.5533 +trainer/Q1 Predictions Max -0.221734 +trainer/Q1 Predictions Min -87.5124 +trainer/Q2 Predictions Mean -75.4859 +trainer/Q2 Predictions Std 15.5651 +trainer/Q2 Predictions Max 0.24204 +trainer/Q2 Predictions Min -87.4938 +trainer/Q Targets Mean -76.3228 +trainer/Q Targets Std 15.5382 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.3594 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0419097 +trainer/policy/mean Std 0.743277 +trainer/policy/mean Max 0.999746 +trainer/policy/mean Min -0.999818 +trainer/policy/std Mean 0.409672 +trainer/policy/std Std 0.0200186 +trainer/policy/std Max 0.429864 +trainer/policy/std Min 0.37829 +trainer/Advantage Weights Mean 12.3337 +trainer/Advantage Weights Std 26.3432 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.65189e-16 +trainer/Advantage Score Mean -0.153076 +trainer/Advantage Score Std 0.52548 +trainer/Advantage Score Max 1.06461 +trainer/Advantage Score Min -3.58661 +trainer/V1 Predictions Mean -76.1159 +trainer/V1 Predictions Std 15.5539 +trainer/V1 Predictions Max 0.0120137 +trainer/V1 Predictions Min -88.2299 +trainer/VF Loss 0.0666227 +expl/num steps total 767000 +expl/num paths total 1034 +expl/path length Mean 500 +expl/path length Std 44 +expl/path length Max 544 +expl/path length Min 456 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0127383 +expl/Actions Std 0.863506 +expl/Actions Max 2.24554 +expl/Actions Min -2.3727 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 699617 +eval/num paths total 771 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.127191 +eval/Actions Std 0.714564 +eval/Actions Max 0.999876 +eval/Actions Min -0.99976 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.36138e-06 +time/evaluation sampling (s) 4.90842 +time/exploration sampling (s) 6.27431 +time/logging (s) 0.012541 +time/saving (s) 0.0180062 +time/training (s) 18.9805 +time/epoch (s) 30.1938 +time/total (s) 19815 +Epoch -234 +------------------------------ ---------------- +2022-05-15 23:33:18.080571 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -233 finished +------------------------------ ---------------- +epoch -233 +replay_buffer/size 999047 +trainer/num train calls 768000 +trainer/QF1 Loss 0.663722 +trainer/QF2 Loss 0.588285 +trainer/Policy Loss 9.35325 +trainer/Q1 Predictions Mean -74.1666 +trainer/Q1 Predictions Std 18.3787 +trainer/Q1 Predictions Max -1.39235 +trainer/Q1 Predictions Min -87.8074 +trainer/Q2 Predictions Mean -74.1219 +trainer/Q2 Predictions Std 18.3812 +trainer/Q2 Predictions Max -1.34956 +trainer/Q2 Predictions Min -88.0073 +trainer/Q Targets Mean -73.8768 +trainer/Q Targets Std 18.482 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8023 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00856508 +trainer/policy/mean Std 0.731523 +trainer/policy/mean Max 0.999843 +trainer/policy/mean Min -0.999053 +trainer/policy/std Mean 0.410189 +trainer/policy/std Std 0.01948 +trainer/policy/std Max 0.430779 +trainer/policy/std Min 0.380851 +trainer/Advantage Weights Mean 2.46263 +trainer/Advantage Weights Std 14.1468 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.36481e-13 +trainer/Advantage Score Mean -0.539369 +trainer/Advantage Score Std 0.53086 +trainer/Advantage Score Max 1.07647 +trainer/Advantage Score Min -2.76966 +trainer/V1 Predictions Mean -73.7048 +trainer/V1 Predictions Std 18.5197 +trainer/V1 Predictions Max -0.420508 +trainer/V1 Predictions Min -87.7363 +trainer/VF Loss 0.069434 +expl/num steps total 768000 +expl/num paths total 1036 +expl/path length Mean 500 +expl/path length Std 468 +expl/path length Max 968 +expl/path length Min 32 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0187798 +expl/Actions Std 0.831103 +expl/Actions Max 2.28596 +expl/Actions Min -2.41771 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 700617 +eval/num paths total 772 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0786149 +eval/Actions Std 0.731264 +eval/Actions Max 0.999829 +eval/Actions Min -0.999424 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.12527e-05 +time/evaluation sampling (s) 5.46166 +time/exploration sampling (s) 7.89925 +time/logging (s) 0.00879772 +time/saving (s) 0.0120483 +time/training (s) 19.5069 +time/epoch (s) 32.8886 +time/total (s) 19847.9 +Epoch -233 +------------------------------ ---------------- +2022-05-15 23:33:51.577043 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -232 finished +------------------------------ ---------------- +epoch -232 +replay_buffer/size 999047 +trainer/num train calls 769000 +trainer/QF1 Loss 0.685302 +trainer/QF2 Loss 0.809985 +trainer/Policy Loss 18.6051 +trainer/Q1 Predictions Mean -74.0857 +trainer/Q1 Predictions Std 18.7673 +trainer/Q1 Predictions Max -1.11748 +trainer/Q1 Predictions Min -88.4515 +trainer/Q2 Predictions Mean -74.1271 +trainer/Q2 Predictions Std 18.7282 +trainer/Q2 Predictions Max -0.888383 +trainer/Q2 Predictions Min -88.0156 +trainer/Q Targets Mean -74.1274 +trainer/Q Targets Std 18.8194 +trainer/Q Targets Max -2.52513 +trainer/Q Targets Min -88.0346 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0297489 +trainer/policy/mean Std 0.728618 +trainer/policy/mean Max 0.998887 +trainer/policy/mean Min -0.999313 +trainer/policy/std Mean 0.409807 +trainer/policy/std Std 0.0179564 +trainer/policy/std Max 0.430049 +trainer/policy/std Min 0.380584 +trainer/Advantage Weights Mean 5.80948 +trainer/Advantage Weights Std 21.8543 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.87401e-16 +trainer/Advantage Score Mean -0.370609 +trainer/Advantage Score Std 0.638786 +trainer/Advantage Score Max 1.35188 +trainer/Advantage Score Min -3.47778 +trainer/V1 Predictions Mean -73.8741 +trainer/V1 Predictions Std 19.0138 +trainer/V1 Predictions Max 0.616375 +trainer/V1 Predictions Min -87.9098 +trainer/VF Loss 0.0760006 +expl/num steps total 769000 +expl/num paths total 1037 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00819463 +expl/Actions Std 0.866997 +expl/Actions Max 2.26117 +expl/Actions Min -2.27203 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 701376 +eval/num paths total 773 +eval/path length Mean 759 +eval/path length Std 0 +eval/path length Max 759 +eval/path length Min 759 +eval/Rewards Mean 0.00131752 +eval/Rewards Std 0.0362738 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.010545 +eval/Actions Std 0.735174 +eval/Actions Max 0.999714 +eval/Actions Min -0.999016 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.3296e-05 +time/evaluation sampling (s) 5.31869 +time/exploration sampling (s) 8.22397 +time/logging (s) 0.00969358 +time/saving (s) 0.0136758 +time/training (s) 19.9164 +time/epoch (s) 33.4825 +time/total (s) 19881.4 +Epoch -232 +------------------------------ ---------------- +2022-05-15 23:34:23.834216 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -231 finished +------------------------------ ---------------- +epoch -231 +replay_buffer/size 999047 +trainer/num train calls 770000 +trainer/QF1 Loss 0.64785 +trainer/QF2 Loss 0.586959 +trainer/Policy Loss 7.44505 +trainer/Q1 Predictions Mean -74.4637 +trainer/Q1 Predictions Std 18.5148 +trainer/Q1 Predictions Max -0.517744 +trainer/Q1 Predictions Min -88.3276 +trainer/Q2 Predictions Mean -74.4866 +trainer/Q2 Predictions Std 18.5236 +trainer/Q2 Predictions Max -0.329974 +trainer/Q2 Predictions Min -88.2201 +trainer/Q Targets Mean -74.224 +trainer/Q Targets Std 18.5108 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7749 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000173761 +trainer/policy/mean Std 0.732665 +trainer/policy/mean Max 0.998941 +trainer/policy/mean Min -0.999109 +trainer/policy/std Mean 0.410183 +trainer/policy/std Std 0.0199988 +trainer/policy/std Max 0.4344 +trainer/policy/std Min 0.376781 +trainer/Advantage Weights Mean 2.46914 +trainer/Advantage Weights Std 13.8501 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.12178e-18 +trainer/Advantage Score Mean -0.608677 +trainer/Advantage Score Std 0.626773 +trainer/Advantage Score Max 1.4954 +trainer/Advantage Score Min -4.06943 +trainer/V1 Predictions Mean -73.9519 +trainer/V1 Predictions Std 18.6734 +trainer/V1 Predictions Max -0.279687 +trainer/V1 Predictions Min -87.6138 +trainer/VF Loss 0.0879236 +expl/num steps total 770000 +expl/num paths total 1039 +expl/path length Mean 500 +expl/path length Std 133 +expl/path length Max 633 +expl/path length Min 367 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0252292 +expl/Actions Std 0.836978 +expl/Actions Max 2.24562 +expl/Actions Min -2.44599 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 702183 +eval/num paths total 774 +eval/path length Mean 807 +eval/path length Std 0 +eval/path length Max 807 +eval/path length Min 807 +eval/Rewards Mean 0.00123916 +eval/Rewards Std 0.0351799 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.043223 +eval/Actions Std 0.736415 +eval/Actions Max 0.999462 +eval/Actions Min -0.998221 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.1337e-05 +time/evaluation sampling (s) 5.29411 +time/exploration sampling (s) 6.91189 +time/logging (s) 0.00788879 +time/saving (s) 0.0113204 +time/training (s) 20.0117 +time/epoch (s) 32.2369 +time/total (s) 19913.7 +Epoch -231 +------------------------------ ---------------- +2022-05-15 23:34:55.108510 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -230 finished +------------------------------ ---------------- +epoch -230 +replay_buffer/size 999047 +trainer/num train calls 771000 +trainer/QF1 Loss 0.652081 +trainer/QF2 Loss 0.559805 +trainer/Policy Loss 21.6392 +trainer/Q1 Predictions Mean -73.2492 +trainer/Q1 Predictions Std 18.8959 +trainer/Q1 Predictions Max -0.224874 +trainer/Q1 Predictions Min -87.7015 +trainer/Q2 Predictions Mean -73.2585 +trainer/Q2 Predictions Std 18.8564 +trainer/Q2 Predictions Max -0.0188478 +trainer/Q2 Predictions Min -87.9655 +trainer/Q Targets Mean -73.5195 +trainer/Q Targets Std 19.0172 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.3089 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00900431 +trainer/policy/mean Std 0.733364 +trainer/policy/mean Max 0.999822 +trainer/policy/mean Min -0.999158 +trainer/policy/std Mean 0.408839 +trainer/policy/std Std 0.018634 +trainer/policy/std Max 0.428798 +trainer/policy/std Min 0.377747 +trainer/Advantage Weights Mean 5.95549 +trainer/Advantage Weights Std 19.7789 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.27157e-19 +trainer/Advantage Score Mean -0.354345 +trainer/Advantage Score Std 0.699111 +trainer/Advantage Score Max 0.970383 +trainer/Advantage Score Min -4.19131 +trainer/V1 Predictions Mean -73.2458 +trainer/V1 Predictions Std 19.1792 +trainer/V1 Predictions Max 0.253104 +trainer/V1 Predictions Min -88.1899 +trainer/VF Loss 0.0834946 +expl/num steps total 771000 +expl/num paths total 1041 +expl/path length Mean 500 +expl/path length Std 449 +expl/path length Max 949 +expl/path length Min 51 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0383791 +expl/Actions Std 0.830863 +expl/Actions Max 2.38442 +expl/Actions Min -2.36809 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 703183 +eval/num paths total 775 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00894253 +eval/Actions Std 0.704877 +eval/Actions Max 0.998757 +eval/Actions Min -0.999415 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.38953e-05 +time/evaluation sampling (s) 5.41828 +time/exploration sampling (s) 6.44842 +time/logging (s) 0.0129059 +time/saving (s) 0.0152703 +time/training (s) 19.3715 +time/epoch (s) 31.2664 +time/total (s) 19944.9 +Epoch -230 +------------------------------ ---------------- +2022-05-15 23:35:25.392635 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -229 finished +------------------------------ ---------------- +epoch -229 +replay_buffer/size 999047 +trainer/num train calls 772000 +trainer/QF1 Loss 0.750967 +trainer/QF2 Loss 0.777123 +trainer/Policy Loss 22.2604 +trainer/Q1 Predictions Mean -71.9194 +trainer/Q1 Predictions Std 20.11 +trainer/Q1 Predictions Max -0.855926 +trainer/Q1 Predictions Min -87.9887 +trainer/Q2 Predictions Mean -71.9096 +trainer/Q2 Predictions Std 20.1039 +trainer/Q2 Predictions Max -0.195729 +trainer/Q2 Predictions Min -88.0078 +trainer/Q Targets Mean -71.9554 +trainer/Q Targets Std 20.0091 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1208 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0264918 +trainer/policy/mean Std 0.735489 +trainer/policy/mean Max 0.999143 +trainer/policy/mean Min -0.999072 +trainer/policy/std Mean 0.41066 +trainer/policy/std Std 0.0193744 +trainer/policy/std Max 0.433584 +trainer/policy/std Min 0.378822 +trainer/Advantage Weights Mean 5.39175 +trainer/Advantage Weights Std 19.1329 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.23665e-22 +trainer/Advantage Score Mean -0.395763 +trainer/Advantage Score Std 0.678574 +trainer/Advantage Score Max 1.05575 +trainer/Advantage Score Min -4.98519 +trainer/V1 Predictions Mean -71.6586 +trainer/V1 Predictions Std 20.3413 +trainer/V1 Predictions Max -1.05253 +trainer/V1 Predictions Min -88.0022 +trainer/VF Loss 0.0799691 +expl/num steps total 772000 +expl/num paths total 1043 +expl/path length Mean 500 +expl/path length Std 112 +expl/path length Max 612 +expl/path length Min 388 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0308424 +expl/Actions Std 0.839641 +expl/Actions Max 2.22184 +expl/Actions Min -2.48483 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 703918 +eval/num paths total 776 +eval/path length Mean 735 +eval/path length Std 0 +eval/path length Max 735 +eval/path length Min 735 +eval/Rewards Mean 0.00136054 +eval/Rewards Std 0.0368605 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0139964 +eval/Actions Std 0.760024 +eval/Actions Max 0.999753 +eval/Actions Min -0.99975 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.0937e-05 +time/evaluation sampling (s) 4.82816 +time/exploration sampling (s) 6.57837 +time/logging (s) 0.0117506 +time/saving (s) 0.0197054 +time/training (s) 18.8269 +time/epoch (s) 30.2649 +time/total (s) 19975.2 +Epoch -229 +------------------------------ ---------------- +2022-05-15 23:35:58.089994 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -228 finished +------------------------------ ---------------- +epoch -228 +replay_buffer/size 999047 +trainer/num train calls 773000 +trainer/QF1 Loss 0.751658 +trainer/QF2 Loss 0.848475 +trainer/Policy Loss 12.3334 +trainer/Q1 Predictions Mean -73.6551 +trainer/Q1 Predictions Std 18.8949 +trainer/Q1 Predictions Max -2.88651 +trainer/Q1 Predictions Min -88.0741 +trainer/Q2 Predictions Mean -73.6421 +trainer/Q2 Predictions Std 18.8546 +trainer/Q2 Predictions Max -2.36339 +trainer/Q2 Predictions Min -88.0247 +trainer/Q Targets Mean -73.8237 +trainer/Q Targets Std 18.8128 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0232 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0110032 +trainer/policy/mean Std 0.723188 +trainer/policy/mean Max 0.998897 +trainer/policy/mean Min -0.999599 +trainer/policy/std Mean 0.409455 +trainer/policy/std Std 0.0193494 +trainer/policy/std Max 0.431531 +trainer/policy/std Min 0.377476 +trainer/Advantage Weights Mean 3.00444 +trainer/Advantage Weights Std 14.573 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.67782e-13 +trainer/Advantage Score Mean -0.388316 +trainer/Advantage Score Std 0.551061 +trainer/Advantage Score Max 2.87787 +trainer/Advantage Score Min -2.86313 +trainer/V1 Predictions Mean -73.5752 +trainer/V1 Predictions Std 18.8721 +trainer/V1 Predictions Max -1.8603 +trainer/V1 Predictions Min -87.9029 +trainer/VF Loss 0.0863708 +expl/num steps total 773000 +expl/num paths total 1045 +expl/path length Mean 500 +expl/path length Std 116 +expl/path length Max 616 +expl/path length Min 384 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0157985 +expl/Actions Std 0.837274 +expl/Actions Max 2.40152 +expl/Actions Min -2.14244 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 704918 +eval/num paths total 777 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.158592 +eval/Actions Std 0.582732 +eval/Actions Max 0.9999 +eval/Actions Min -0.99913 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.07782e-05 +time/evaluation sampling (s) 5.46235 +time/exploration sampling (s) 7.72313 +time/logging (s) 0.00793429 +time/saving (s) 0.0162428 +time/training (s) 19.4631 +time/epoch (s) 32.6728 +time/total (s) 20007.9 +Epoch -228 +------------------------------ ---------------- +2022-05-15 23:36:31.154145 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -227 finished +------------------------------ ---------------- +epoch -227 +replay_buffer/size 999047 +trainer/num train calls 774000 +trainer/QF1 Loss 0.406673 +trainer/QF2 Loss 0.485485 +trainer/Policy Loss 27.44 +trainer/Q1 Predictions Mean -73.8902 +trainer/Q1 Predictions Std 18.9282 +trainer/Q1 Predictions Max -0.222453 +trainer/Q1 Predictions Min -88.2801 +trainer/Q2 Predictions Mean -73.918 +trainer/Q2 Predictions Std 18.8115 +trainer/Q2 Predictions Max 0.161559 +trainer/Q2 Predictions Min -88.0406 +trainer/Q Targets Mean -74.0309 +trainer/Q Targets Std 18.8194 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0062 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00481569 +trainer/policy/mean Std 0.732116 +trainer/policy/mean Max 0.999522 +trainer/policy/mean Min -0.99938 +trainer/policy/std Mean 0.409849 +trainer/policy/std Std 0.0196173 +trainer/policy/std Max 0.431749 +trainer/policy/std Min 0.378101 +trainer/Advantage Weights Mean 5.46363 +trainer/Advantage Weights Std 19.8138 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.6801e-17 +trainer/Advantage Score Mean -0.285944 +trainer/Advantage Score Std 0.559504 +trainer/Advantage Score Max 0.813194 +trainer/Advantage Score Min -3.68739 +trainer/V1 Predictions Mean -73.7536 +trainer/V1 Predictions Std 19.0039 +trainer/V1 Predictions Max 0.394899 +trainer/V1 Predictions Min -87.9904 +trainer/VF Loss 0.0570791 +expl/num steps total 774000 +expl/num paths total 1047 +expl/path length Mean 500 +expl/path length Std 436 +expl/path length Max 936 +expl/path length Min 64 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0485458 +expl/Actions Std 0.81246 +expl/Actions Max 2.22882 +expl/Actions Min -2.25576 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 705436 +eval/num paths total 778 +eval/path length Mean 518 +eval/path length Std 0 +eval/path length Max 518 +eval/path length Min 518 +eval/Rewards Mean 0.0019305 +eval/Rewards Std 0.043895 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0023789 +eval/Actions Std 0.748715 +eval/Actions Max 0.999885 +eval/Actions Min -0.999517 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.27489e-05 +time/evaluation sampling (s) 6.20686 +time/exploration sampling (s) 7.5344 +time/logging (s) 0.010221 +time/saving (s) 0.0183719 +time/training (s) 19.2774 +time/epoch (s) 33.0473 +time/total (s) 20041 +Epoch -227 +------------------------------ ---------------- +2022-05-15 23:37:03.817989 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -226 finished +------------------------------ --------------- +epoch -226 +replay_buffer/size 999047 +trainer/num train calls 775000 +trainer/QF1 Loss 1.30324 +trainer/QF2 Loss 1.2881 +trainer/Policy Loss 33.6406 +trainer/Q1 Predictions Mean -74.3737 +trainer/Q1 Predictions Std 16.9811 +trainer/Q1 Predictions Max -1.28651 +trainer/Q1 Predictions Min -87.4019 +trainer/Q2 Predictions Mean -74.3757 +trainer/Q2 Predictions Std 16.9761 +trainer/Q2 Predictions Max -2.04719 +trainer/Q2 Predictions Min -87.4277 +trainer/Q Targets Mean -74.8195 +trainer/Q Targets Std 16.7272 +trainer/Q Targets Max -3.61742 +trainer/Q Targets Min -87.9239 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0104249 +trainer/policy/mean Std 0.726453 +trainer/policy/mean Max 0.99972 +trainer/policy/mean Min -0.997912 +trainer/policy/std Mean 0.410954 +trainer/policy/std Std 0.0201521 +trainer/policy/std Max 0.432616 +trainer/policy/std Min 0.381884 +trainer/Advantage Weights Mean 8.04607 +trainer/Advantage Weights Std 24.4477 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.5597e-09 +trainer/Advantage Score Mean -0.195104 +trainer/Advantage Score Std 0.458386 +trainer/Advantage Score Max 1.18813 +trainer/Advantage Score Min -1.90077 +trainer/V1 Predictions Mean -74.5771 +trainer/V1 Predictions Std 16.8792 +trainer/V1 Predictions Max -1.81729 +trainer/V1 Predictions Min -87.8022 +trainer/VF Loss 0.0579501 +expl/num steps total 775000 +expl/num paths total 1049 +expl/path length Mean 500 +expl/path length Std 120 +expl/path length Max 620 +expl/path length Min 380 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0859101 +expl/Actions Std 0.816098 +expl/Actions Max 2.23824 +expl/Actions Min -2.32722 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 706436 +eval/num paths total 779 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.167528 +eval/Actions Std 0.790668 +eval/Actions Max 0.999383 +eval/Actions Min -0.999692 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.8261e-06 +time/evaluation sampling (s) 5.26883 +time/exploration sampling (s) 7.65483 +time/logging (s) 0.00925648 +time/saving (s) 0.0166132 +time/training (s) 19.6931 +time/epoch (s) 32.6426 +time/total (s) 20073.6 +Epoch -226 +------------------------------ --------------- +2022-05-15 23:37:35.363799 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -225 finished +------------------------------ ---------------- +epoch -225 +replay_buffer/size 999047 +trainer/num train calls 776000 +trainer/QF1 Loss 0.975669 +trainer/QF2 Loss 0.854886 +trainer/Policy Loss 11.8862 +trainer/Q1 Predictions Mean -75.8488 +trainer/Q1 Predictions Std 15.9034 +trainer/Q1 Predictions Max -0.709056 +trainer/Q1 Predictions Min -86.9632 +trainer/Q2 Predictions Mean -75.8152 +trainer/Q2 Predictions Std 15.8926 +trainer/Q2 Predictions Max -1.41824 +trainer/Q2 Predictions Min -86.9301 +trainer/Q Targets Mean -75.8057 +trainer/Q Targets Std 15.8363 +trainer/Q Targets Max -4.36322 +trainer/Q Targets Min -87.0327 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00561767 +trainer/policy/mean Std 0.736721 +trainer/policy/mean Max 0.999493 +trainer/policy/mean Min -0.998539 +trainer/policy/std Mean 0.410723 +trainer/policy/std Std 0.0205592 +trainer/policy/std Max 0.434862 +trainer/policy/std Min 0.379734 +trainer/Advantage Weights Mean 4.09759 +trainer/Advantage Weights Std 16.5082 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.33741e-14 +trainer/Advantage Score Mean -0.348438 +trainer/Advantage Score Std 0.539206 +trainer/Advantage Score Max 0.845985 +trainer/Advantage Score Min -3.19455 +trainer/V1 Predictions Mean -75.5211 +trainer/V1 Predictions Std 16.046 +trainer/V1 Predictions Max -1.16398 +trainer/V1 Predictions Min -86.9351 +trainer/VF Loss 0.0534035 +expl/num steps total 776000 +expl/num paths total 1050 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0194539 +expl/Actions Std 0.827225 +expl/Actions Max 2.56243 +expl/Actions Min -2.24773 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 706909 +eval/num paths total 780 +eval/path length Mean 473 +eval/path length Std 0 +eval/path length Max 473 +eval/path length Min 473 +eval/Rewards Mean 0.00211416 +eval/Rewards Std 0.0459314 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0198102 +eval/Actions Std 0.759983 +eval/Actions Max 0.999685 +eval/Actions Min -0.999602 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.25831e-05 +time/evaluation sampling (s) 5.60523 +time/exploration sampling (s) 6.30849 +time/logging (s) 0.0100529 +time/saving (s) 0.0188583 +time/training (s) 19.5881 +time/epoch (s) 31.5307 +time/total (s) 20105.1 +Epoch -225 +------------------------------ ---------------- +2022-05-15 23:38:05.703583 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -224 finished +------------------------------ --------------- +epoch -224 +replay_buffer/size 999047 +trainer/num train calls 777000 +trainer/QF1 Loss 0.923209 +trainer/QF2 Loss 0.978119 +trainer/Policy Loss 27.7438 +trainer/Q1 Predictions Mean -74.5985 +trainer/Q1 Predictions Std 16.9841 +trainer/Q1 Predictions Max -0.817251 +trainer/Q1 Predictions Min -87.6633 +trainer/Q2 Predictions Mean -74.5936 +trainer/Q2 Predictions Std 16.9807 +trainer/Q2 Predictions Max -0.392285 +trainer/Q2 Predictions Min -87.4571 +trainer/Q Targets Mean -74.903 +trainer/Q Targets Std 16.7588 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7623 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0322294 +trainer/policy/mean Std 0.730038 +trainer/policy/mean Max 0.999619 +trainer/policy/mean Min -0.999689 +trainer/policy/std Mean 0.409764 +trainer/policy/std Std 0.0199378 +trainer/policy/std Max 0.43054 +trainer/policy/std Min 0.378089 +trainer/Advantage Weights Mean 6.31194 +trainer/Advantage Weights Std 20.8821 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.6569e-15 +trainer/Advantage Score Mean -0.318051 +trainer/Advantage Score Std 0.5457 +trainer/Advantage Score Max 1.58484 +trainer/Advantage Score Min -3.28059 +trainer/V1 Predictions Mean -74.6504 +trainer/V1 Predictions Std 16.9112 +trainer/V1 Predictions Max -0.730642 +trainer/V1 Predictions Min -87.5908 +trainer/VF Loss 0.0775 +expl/num steps total 777000 +expl/num paths total 1052 +expl/path length Mean 500 +expl/path length Std 362 +expl/path length Max 862 +expl/path length Min 138 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0544168 +expl/Actions Std 0.825285 +expl/Actions Max 2.42107 +expl/Actions Min -2.22414 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 707909 +eval/num paths total 781 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.2592 +eval/Actions Std 0.615563 +eval/Actions Max 0.999689 +eval/Actions Min -0.999866 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.074e-05 +time/evaluation sampling (s) 4.78647 +time/exploration sampling (s) 6.54204 +time/logging (s) 0.0108577 +time/saving (s) 0.0154015 +time/training (s) 18.9652 +time/epoch (s) 30.3199 +time/total (s) 20135.5 +Epoch -224 +------------------------------ --------------- +2022-05-15 23:38:36.463024 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -223 finished +------------------------------ ---------------- +epoch -223 +replay_buffer/size 999047 +trainer/num train calls 778000 +trainer/QF1 Loss 1.40876 +trainer/QF2 Loss 1.44621 +trainer/Policy Loss 16.7579 +trainer/Q1 Predictions Mean -72.9513 +trainer/Q1 Predictions Std 18.7017 +trainer/Q1 Predictions Max -0.96853 +trainer/Q1 Predictions Min -88.0018 +trainer/Q2 Predictions Mean -73.0061 +trainer/Q2 Predictions Std 18.6875 +trainer/Q2 Predictions Max -1.53852 +trainer/Q2 Predictions Min -88.4566 +trainer/Q Targets Mean -72.7287 +trainer/Q Targets Std 19.3327 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8144 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0125802 +trainer/policy/mean Std 0.729253 +trainer/policy/mean Max 0.999548 +trainer/policy/mean Min -0.999102 +trainer/policy/std Mean 0.410642 +trainer/policy/std Std 0.0195099 +trainer/policy/std Max 0.431167 +trainer/policy/std Min 0.379329 +trainer/Advantage Weights Mean 4.34797 +trainer/Advantage Weights Std 17.7036 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.30668e-17 +trainer/Advantage Score Mean -0.300524 +trainer/Advantage Score Std 0.56535 +trainer/Advantage Score Max 1.17757 +trainer/Advantage Score Min -3.88765 +trainer/V1 Predictions Mean -72.665 +trainer/V1 Predictions Std 19.033 +trainer/V1 Predictions Max -0.459849 +trainer/V1 Predictions Min -87.6918 +trainer/VF Loss 0.0581564 +expl/num steps total 778000 +expl/num paths total 1054 +expl/path length Mean 500 +expl/path length Std 299 +expl/path length Max 799 +expl/path length Min 201 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0109294 +expl/Actions Std 0.840637 +expl/Actions Max 2.2226 +expl/Actions Min -2.19293 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 708909 +eval/num paths total 782 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.108292 +eval/Actions Std 0.711299 +eval/Actions Max 0.999963 +eval/Actions Min -0.999794 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.234e-05 +time/evaluation sampling (s) 4.68165 +time/exploration sampling (s) 6.77649 +time/logging (s) 0.0125658 +time/saving (s) 0.0187812 +time/training (s) 19.2578 +time/epoch (s) 30.7473 +time/total (s) 20166.2 +Epoch -223 +------------------------------ ---------------- +2022-05-15 23:39:07.392674 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -222 finished +------------------------------ ---------------- +epoch -222 +replay_buffer/size 999047 +trainer/num train calls 779000 +trainer/QF1 Loss 0.590373 +trainer/QF2 Loss 0.614111 +trainer/Policy Loss 19.5368 +trainer/Q1 Predictions Mean -73.6573 +trainer/Q1 Predictions Std 19.1111 +trainer/Q1 Predictions Max -0.270869 +trainer/Q1 Predictions Min -88.2716 +trainer/Q2 Predictions Mean -73.5985 +trainer/Q2 Predictions Std 19.0671 +trainer/Q2 Predictions Max 0.516393 +trainer/Q2 Predictions Min -88.1084 +trainer/Q Targets Mean -73.6858 +trainer/Q Targets Std 19.2192 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8098 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0270736 +trainer/policy/mean Std 0.724406 +trainer/policy/mean Max 0.999425 +trainer/policy/mean Min -0.999933 +trainer/policy/std Mean 0.409391 +trainer/policy/std Std 0.0202997 +trainer/policy/std Max 0.429855 +trainer/policy/std Min 0.375286 +trainer/Advantage Weights Mean 4.0195 +trainer/Advantage Weights Std 16.8615 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.18947e-15 +trainer/Advantage Score Mean -0.389984 +trainer/Advantage Score Std 0.567855 +trainer/Advantage Score Max 1.4417 +trainer/Advantage Score Min -3.43653 +trainer/V1 Predictions Mean -73.5172 +trainer/V1 Predictions Std 19.1028 +trainer/V1 Predictions Max 0.422167 +trainer/V1 Predictions Min -87.8773 +trainer/VF Loss 0.0634142 +expl/num steps total 779000 +expl/num paths total 1056 +expl/path length Mean 500 +expl/path length Std 252 +expl/path length Max 752 +expl/path length Min 248 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0293163 +expl/Actions Std 0.827882 +expl/Actions Max 2.13984 +expl/Actions Min -2.58157 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 709588 +eval/num paths total 783 +eval/path length Mean 679 +eval/path length Std 0 +eval/path length Max 679 +eval/path length Min 679 +eval/Rewards Mean 0.00147275 +eval/Rewards Std 0.0383482 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0765852 +eval/Actions Std 0.72849 +eval/Actions Max 0.999993 +eval/Actions Min -0.999498 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.05542e-05 +time/evaluation sampling (s) 4.71914 +time/exploration sampling (s) 7.37635 +time/logging (s) 0.0111123 +time/saving (s) 0.016411 +time/training (s) 18.7843 +time/epoch (s) 30.9073 +time/total (s) 20197.1 +Epoch -222 +------------------------------ ---------------- +2022-05-15 23:39:38.680971 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -221 finished +------------------------------ ---------------- +epoch -221 +replay_buffer/size 999047 +trainer/num train calls 780000 +trainer/QF1 Loss 0.597947 +trainer/QF2 Loss 0.723361 +trainer/Policy Loss 9.52077 +trainer/Q1 Predictions Mean -75.5234 +trainer/Q1 Predictions Std 16.1195 +trainer/Q1 Predictions Max -2.44844 +trainer/Q1 Predictions Min -88.0795 +trainer/Q2 Predictions Mean -75.6013 +trainer/Q2 Predictions Std 16.1366 +trainer/Q2 Predictions Max -3.04375 +trainer/Q2 Predictions Min -88.3583 +trainer/Q Targets Mean -75.4344 +trainer/Q Targets Std 16.272 +trainer/Q Targets Max -4.32818 +trainer/Q Targets Min -87.9989 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00181051 +trainer/policy/mean Std 0.729008 +trainer/policy/mean Max 0.998886 +trainer/policy/mean Min -0.999761 +trainer/policy/std Mean 0.409785 +trainer/policy/std Std 0.0195649 +trainer/policy/std Max 0.429579 +trainer/policy/std Min 0.377081 +trainer/Advantage Weights Mean 2.84678 +trainer/Advantage Weights Std 13.1616 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.57698e-21 +trainer/Advantage Score Mean -0.354126 +trainer/Advantage Score Std 0.566128 +trainer/Advantage Score Max 0.705204 +trainer/Advantage Score Min -4.68332 +trainer/V1 Predictions Mean -75.1951 +trainer/V1 Predictions Std 16.3414 +trainer/V1 Predictions Max -1.87199 +trainer/V1 Predictions Min -87.8209 +trainer/VF Loss 0.0525754 +expl/num steps total 780000 +expl/num paths total 1057 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0935535 +expl/Actions Std 0.88541 +expl/Actions Max 2.37567 +expl/Actions Min -2.41678 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 710535 +eval/num paths total 784 +eval/path length Mean 947 +eval/path length Std 0 +eval/path length Max 947 +eval/path length Min 947 +eval/Rewards Mean 0.00105597 +eval/Rewards Std 0.0324785 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0409308 +eval/Actions Std 0.708219 +eval/Actions Max 0.99992 +eval/Actions Min -0.999233 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.13398e-05 +time/evaluation sampling (s) 5.48608 +time/exploration sampling (s) 6.90708 +time/logging (s) 0.0109668 +time/saving (s) 0.0152415 +time/training (s) 18.8488 +time/epoch (s) 31.2681 +time/total (s) 20228.4 +Epoch -221 +------------------------------ ---------------- +2022-05-15 23:40:09.268008 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -220 finished +------------------------------ ---------------- +epoch -220 +replay_buffer/size 999047 +trainer/num train calls 781000 +trainer/QF1 Loss 0.570574 +trainer/QF2 Loss 0.458948 +trainer/Policy Loss 11.0459 +trainer/Q1 Predictions Mean -73.5094 +trainer/Q1 Predictions Std 18.8005 +trainer/Q1 Predictions Max -0.219495 +trainer/Q1 Predictions Min -87.9985 +trainer/Q2 Predictions Mean -73.4426 +trainer/Q2 Predictions Std 18.7826 +trainer/Q2 Predictions Max -0.317822 +trainer/Q2 Predictions Min -87.7939 +trainer/Q Targets Mean -73.4809 +trainer/Q Targets Std 18.6332 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7324 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00757814 +trainer/policy/mean Std 0.736217 +trainer/policy/mean Max 0.999186 +trainer/policy/mean Min -0.99869 +trainer/policy/std Mean 0.411328 +trainer/policy/std Std 0.0190137 +trainer/policy/std Max 0.430203 +trainer/policy/std Min 0.383107 +trainer/Advantage Weights Mean 3.02991 +trainer/Advantage Weights Std 14.5326 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.47476e-18 +trainer/Advantage Score Mean -0.398131 +trainer/Advantage Score Std 0.589794 +trainer/Advantage Score Max 1.22852 +trainer/Advantage Score Min -4.05404 +trainer/V1 Predictions Mean -73.2537 +trainer/V1 Predictions Std 18.7484 +trainer/V1 Predictions Max 0.0712103 +trainer/V1 Predictions Min -87.6084 +trainer/VF Loss 0.064609 +expl/num steps total 781000 +expl/num paths total 1058 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.401034 +expl/Actions Std 0.798661 +expl/Actions Max 2.15478 +expl/Actions Min -2.47506 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 711535 +eval/num paths total 785 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.206514 +eval/Actions Std 0.550683 +eval/Actions Max 0.999648 +eval/Actions Min -0.999557 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.34106e-05 +time/evaluation sampling (s) 4.74007 +time/exploration sampling (s) 6.71453 +time/logging (s) 0.00760617 +time/saving (s) 0.0162507 +time/training (s) 19.0886 +time/epoch (s) 30.5671 +time/total (s) 20259 +Epoch -220 +------------------------------ ---------------- +2022-05-15 23:40:39.521159 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -219 finished +------------------------------ ---------------- +epoch -219 +replay_buffer/size 999047 +trainer/num train calls 782000 +trainer/QF1 Loss 0.710858 +trainer/QF2 Loss 0.787665 +trainer/Policy Loss 24.8796 +trainer/Q1 Predictions Mean -74.7193 +trainer/Q1 Predictions Std 17.8104 +trainer/Q1 Predictions Max -2.70104 +trainer/Q1 Predictions Min -87.0775 +trainer/Q2 Predictions Mean -74.7137 +trainer/Q2 Predictions Std 17.8143 +trainer/Q2 Predictions Max -1.63562 +trainer/Q2 Predictions Min -87.0513 +trainer/Q Targets Mean -74.6432 +trainer/Q Targets Std 17.7227 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6353 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0141066 +trainer/policy/mean Std 0.724307 +trainer/policy/mean Max 0.999555 +trainer/policy/mean Min -0.997985 +trainer/policy/std Mean 0.409939 +trainer/policy/std Std 0.0198465 +trainer/policy/std Max 0.432627 +trainer/policy/std Min 0.380612 +trainer/Advantage Weights Mean 4.50134 +trainer/Advantage Weights Std 18.5176 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.42619e-11 +trainer/Advantage Score Mean -0.348286 +trainer/Advantage Score Std 0.526708 +trainer/Advantage Score Max 1.82739 +trainer/Advantage Score Min -2.36372 +trainer/V1 Predictions Mean -74.4487 +trainer/V1 Predictions Std 17.6778 +trainer/V1 Predictions Max -1.99512 +trainer/V1 Predictions Min -86.5149 +trainer/VF Loss 0.0710421 +expl/num steps total 782000 +expl/num paths total 1060 +expl/path length Mean 500 +expl/path length Std 453 +expl/path length Max 953 +expl/path length Min 47 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.032935 +expl/Actions Std 0.818781 +expl/Actions Max 2.21961 +expl/Actions Min -2.3731 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 712535 +eval/num paths total 786 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.105801 +eval/Actions Std 0.765353 +eval/Actions Max 0.999027 +eval/Actions Min -0.997357 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 6.51693e-06 +time/evaluation sampling (s) 5.20232 +time/exploration sampling (s) 6.46589 +time/logging (s) 0.0127385 +time/saving (s) 0.0181469 +time/training (s) 18.5457 +time/epoch (s) 30.2448 +time/total (s) 20289.2 +Epoch -219 +------------------------------ ---------------- +2022-05-15 23:41:11.047732 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -218 finished +------------------------------ ---------------- +epoch -218 +replay_buffer/size 999047 +trainer/num train calls 783000 +trainer/QF1 Loss 0.579201 +trainer/QF2 Loss 0.549373 +trainer/Policy Loss 5.66173 +trainer/Q1 Predictions Mean -76.5833 +trainer/Q1 Predictions Std 15.2858 +trainer/Q1 Predictions Max -2.91709 +trainer/Q1 Predictions Min -88.1014 +trainer/Q2 Predictions Mean -76.5971 +trainer/Q2 Predictions Std 15.3086 +trainer/Q2 Predictions Max -1.89753 +trainer/Q2 Predictions Min -88.1784 +trainer/Q Targets Mean -76.165 +trainer/Q Targets Std 15.3811 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7048 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0180354 +trainer/policy/mean Std 0.72917 +trainer/policy/mean Max 0.999964 +trainer/policy/mean Min -0.999295 +trainer/policy/std Mean 0.411048 +trainer/policy/std Std 0.0193936 +trainer/policy/std Max 0.432813 +trainer/policy/std Min 0.379278 +trainer/Advantage Weights Mean 1.51614 +trainer/Advantage Weights Std 9.64674 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.92549e-18 +trainer/Advantage Score Mean -0.432582 +trainer/Advantage Score Std 0.539587 +trainer/Advantage Score Max 1.75105 +trainer/Advantage Score Min -4.0079 +trainer/V1 Predictions Mean -75.9595 +trainer/V1 Predictions Std 15.357 +trainer/V1 Predictions Max -2.73909 +trainer/V1 Predictions Min -87.5769 +trainer/VF Loss 0.0598837 +expl/num steps total 783000 +expl/num paths total 1061 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.000762007 +expl/Actions Std 0.812694 +expl/Actions Max 2.20153 +expl/Actions Min -2.49815 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 713535 +eval/num paths total 787 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.170155 +eval/Actions Std 0.70049 +eval/Actions Max 0.999501 +eval/Actions Min -0.996733 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.06103e-06 +time/evaluation sampling (s) 4.74453 +time/exploration sampling (s) 7.53662 +time/logging (s) 0.0122736 +time/saving (s) 0.0139891 +time/training (s) 19.1975 +time/epoch (s) 31.5049 +time/total (s) 20320.8 +Epoch -218 +------------------------------ ---------------- +2022-05-15 23:41:42.321524 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -217 finished +------------------------------ ---------------- +epoch -217 +replay_buffer/size 999047 +trainer/num train calls 784000 +trainer/QF1 Loss 1.07495 +trainer/QF2 Loss 1.07973 +trainer/Policy Loss 10.9811 +trainer/Q1 Predictions Mean -73.2122 +trainer/Q1 Predictions Std 20.1542 +trainer/Q1 Predictions Max -0.688984 +trainer/Q1 Predictions Min -87.8263 +trainer/Q2 Predictions Mean -73.1177 +trainer/Q2 Predictions Std 20.1051 +trainer/Q2 Predictions Max -1.05968 +trainer/Q2 Predictions Min -88.0163 +trainer/Q Targets Mean -72.9082 +trainer/Q Targets Std 19.7233 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.6769 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0244241 +trainer/policy/mean Std 0.73375 +trainer/policy/mean Max 0.997014 +trainer/policy/mean Min -0.998136 +trainer/policy/std Mean 0.411552 +trainer/policy/std Std 0.0195597 +trainer/policy/std Max 0.43116 +trainer/policy/std Min 0.377621 +trainer/Advantage Weights Mean 3.87177 +trainer/Advantage Weights Std 18.4262 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.16961e-15 +trainer/Advantage Score Mean -0.481805 +trainer/Advantage Score Std 0.579898 +trainer/Advantage Score Max 3.21212 +trainer/Advantage Score Min -3.37642 +trainer/V1 Predictions Mean -72.6491 +trainer/V1 Predictions Std 19.807 +trainer/V1 Predictions Max -0.248414 +trainer/V1 Predictions Min -86.6925 +trainer/VF Loss 0.117094 +expl/num steps total 784000 +expl/num paths total 1063 +expl/path length Mean 500 +expl/path length Std 259 +expl/path length Max 759 +expl/path length Min 241 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00963171 +expl/Actions Std 0.833455 +expl/Actions Max 2.36422 +expl/Actions Min -2.27927 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 714535 +eval/num paths total 788 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0249278 +eval/Actions Std 0.743381 +eval/Actions Max 0.999934 +eval/Actions Min -0.999874 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.11973e-05 +time/evaluation sampling (s) 5.17125 +time/exploration sampling (s) 7.52903 +time/logging (s) 0.00768356 +time/saving (s) 0.0146129 +time/training (s) 18.5286 +time/epoch (s) 31.2512 +time/total (s) 20352 +Epoch -217 +------------------------------ ---------------- +2022-05-15 23:42:12.765689 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -216 finished +------------------------------ ---------------- +epoch -216 +replay_buffer/size 999047 +trainer/num train calls 785000 +trainer/QF1 Loss 0.827812 +trainer/QF2 Loss 0.851108 +trainer/Policy Loss 18.228 +trainer/Q1 Predictions Mean -74.8595 +trainer/Q1 Predictions Std 15.9171 +trainer/Q1 Predictions Max -0.895247 +trainer/Q1 Predictions Min -88.1067 +trainer/Q2 Predictions Mean -74.9239 +trainer/Q2 Predictions Std 15.9405 +trainer/Q2 Predictions Max -1.41866 +trainer/Q2 Predictions Min -88.2186 +trainer/Q Targets Mean -74.787 +trainer/Q Targets Std 16.3338 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1756 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00109509 +trainer/policy/mean Std 0.731191 +trainer/policy/mean Max 0.998935 +trainer/policy/mean Min -0.999951 +trainer/policy/std Mean 0.411441 +trainer/policy/std Std 0.018873 +trainer/policy/std Max 0.432281 +trainer/policy/std Min 0.37978 +trainer/Advantage Weights Mean 2.6784 +trainer/Advantage Weights Std 12.3304 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.16581e-16 +trainer/Advantage Score Mean -0.421076 +trainer/Advantage Score Std 0.685705 +trainer/Advantage Score Max 0.636879 +trainer/Advantage Score Min -3.5689 +trainer/V1 Predictions Mean -74.4982 +trainer/V1 Predictions Std 16.4339 +trainer/V1 Predictions Max -0.213145 +trainer/V1 Predictions Min -87.9115 +trainer/VF Loss 0.071845 +expl/num steps total 785000 +expl/num paths total 1065 +expl/path length Mean 500 +expl/path length Std 173 +expl/path length Max 673 +expl/path length Min 327 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0217148 +expl/Actions Std 0.845053 +expl/Actions Max 2.26627 +expl/Actions Min -2.31553 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 715349 +eval/num paths total 789 +eval/path length Mean 814 +eval/path length Std 0 +eval/path length Max 814 +eval/path length Min 814 +eval/Rewards Mean 0.0012285 +eval/Rewards Std 0.0350284 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0364097 +eval/Actions Std 0.735075 +eval/Actions Max 0.999729 +eval/Actions Min -0.999586 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.03088e-05 +time/evaluation sampling (s) 5.31923 +time/exploration sampling (s) 6.27906 +time/logging (s) 0.0120177 +time/saving (s) 0.0185405 +time/training (s) 18.8069 +time/epoch (s) 30.4357 +time/total (s) 20382.4 +Epoch -216 +------------------------------ ---------------- +2022-05-15 23:42:42.474998 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -215 finished +------------------------------ ---------------- +epoch -215 +replay_buffer/size 999047 +trainer/num train calls 786000 +trainer/QF1 Loss 3.79905 +trainer/QF2 Loss 3.66847 +trainer/Policy Loss 13.9073 +trainer/Q1 Predictions Mean -75.5087 +trainer/Q1 Predictions Std 16.2686 +trainer/Q1 Predictions Max -0.94544 +trainer/Q1 Predictions Min -88.0062 +trainer/Q2 Predictions Mean -75.4805 +trainer/Q2 Predictions Std 16.2655 +trainer/Q2 Predictions Max -1.34339 +trainer/Q2 Predictions Min -87.9248 +trainer/Q Targets Mean -75.331 +trainer/Q Targets Std 16.1208 +trainer/Q Targets Max -2.12543 +trainer/Q Targets Min -88.2563 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0125016 +trainer/policy/mean Std 0.740989 +trainer/policy/mean Max 0.999016 +trainer/policy/mean Min -0.999246 +trainer/policy/std Mean 0.411788 +trainer/policy/std Std 0.0183735 +trainer/policy/std Max 0.43145 +trainer/policy/std Min 0.381673 +trainer/Advantage Weights Mean 2.7425 +trainer/Advantage Weights Std 13.1811 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.2715e-19 +trainer/Advantage Score Mean -0.366707 +trainer/Advantage Score Std 0.557629 +trainer/Advantage Score Max 1.65187 +trainer/Advantage Score Min -4.19131 +trainer/V1 Predictions Mean -75.1977 +trainer/V1 Predictions Std 16.2841 +trainer/V1 Predictions Max -0.804796 +trainer/V1 Predictions Min -88.1671 +trainer/VF Loss 0.0613323 +expl/num steps total 786000 +expl/num paths total 1067 +expl/path length Mean 500 +expl/path length Std 268 +expl/path length Max 768 +expl/path length Min 232 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0200794 +expl/Actions Std 0.828133 +expl/Actions Max 2.44711 +expl/Actions Min -2.60458 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 715866 +eval/num paths total 790 +eval/path length Mean 517 +eval/path length Std 0 +eval/path length Max 517 +eval/path length Min 517 +eval/Rewards Mean 0.00193424 +eval/Rewards Std 0.0439374 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.005131 +eval/Actions Std 0.743723 +eval/Actions Max 0.999846 +eval/Actions Min -0.999002 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.94905e-06 +time/evaluation sampling (s) 4.58619 +time/exploration sampling (s) 5.93159 +time/logging (s) 0.0105294 +time/saving (s) 0.0183232 +time/training (s) 19.1401 +time/epoch (s) 29.6868 +time/total (s) 20412.1 +Epoch -215 +------------------------------ ---------------- +2022-05-15 23:43:13.166317 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -214 finished +------------------------------ ---------------- +epoch -214 +replay_buffer/size 999047 +trainer/num train calls 787000 +trainer/QF1 Loss 4.61709 +trainer/QF2 Loss 4.89933 +trainer/Policy Loss 21.2333 +trainer/Q1 Predictions Mean -74.8941 +trainer/Q1 Predictions Std 16.8212 +trainer/Q1 Predictions Max -2.13772 +trainer/Q1 Predictions Min -87.7097 +trainer/Q2 Predictions Mean -74.8551 +trainer/Q2 Predictions Std 16.8775 +trainer/Q2 Predictions Max -2.21187 +trainer/Q2 Predictions Min -87.4594 +trainer/Q Targets Mean -75.0431 +trainer/Q Targets Std 16.7978 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7115 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00314483 +trainer/policy/mean Std 0.72254 +trainer/policy/mean Max 0.999969 +trainer/policy/mean Min -0.997957 +trainer/policy/std Mean 0.409653 +trainer/policy/std Std 0.0179324 +trainer/policy/std Max 0.429614 +trainer/policy/std Min 0.379776 +trainer/Advantage Weights Mean 5.1634 +trainer/Advantage Weights Std 20.0215 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.92067e-15 +trainer/Advantage Score Mean -0.329163 +trainer/Advantage Score Std 0.490932 +trainer/Advantage Score Max 1.11199 +trainer/Advantage Score Min -3.38861 +trainer/V1 Predictions Mean -74.9037 +trainer/V1 Predictions Std 16.8841 +trainer/V1 Predictions Max -1.8251 +trainer/V1 Predictions Min -87.6957 +trainer/VF Loss 0.0539539 +expl/num steps total 787000 +expl/num paths total 1069 +expl/path length Mean 500 +expl/path length Std 363 +expl/path length Max 863 +expl/path length Min 137 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0298782 +expl/Actions Std 0.826564 +expl/Actions Max 2.4529 +expl/Actions Min -2.35499 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 716815 +eval/num paths total 792 +eval/path length Mean 474.5 +eval/path length Std 30.5 +eval/path length Max 505 +eval/path length Min 444 +eval/Rewards Mean 0.00210748 +eval/Rewards Std 0.0458589 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0155751 +eval/Actions Std 0.740979 +eval/Actions Max 0.9999 +eval/Actions Min -0.999645 +eval/Num Paths 2 +eval/Average Returns 1 +time/data storing (s) 3.03332e-06 +time/evaluation sampling (s) 4.52799 +time/exploration sampling (s) 7.73524 +time/logging (s) 0.00747052 +time/saving (s) 0.0107548 +time/training (s) 18.3859 +time/epoch (s) 30.6673 +time/total (s) 20442.8 +Epoch -214 +------------------------------ ---------------- +2022-05-15 23:43:44.254467 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -213 finished +------------------------------ ---------------- +epoch -213 +replay_buffer/size 999047 +trainer/num train calls 788000 +trainer/QF1 Loss 0.723395 +trainer/QF2 Loss 0.597071 +trainer/Policy Loss 12.2744 +trainer/Q1 Predictions Mean -72.5368 +trainer/Q1 Predictions Std 19.5564 +trainer/Q1 Predictions Max -0.972685 +trainer/Q1 Predictions Min -87.2477 +trainer/Q2 Predictions Mean -72.4817 +trainer/Q2 Predictions Std 19.6847 +trainer/Q2 Predictions Max -0.619781 +trainer/Q2 Predictions Min -87.1356 +trainer/Q Targets Mean -72.3031 +trainer/Q Targets Std 19.7065 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3997 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00683937 +trainer/policy/mean Std 0.730346 +trainer/policy/mean Max 0.999452 +trainer/policy/mean Min -0.997788 +trainer/policy/std Mean 0.410065 +trainer/policy/std Std 0.0203302 +trainer/policy/std Max 0.432559 +trainer/policy/std Min 0.378536 +trainer/Advantage Weights Mean 3.45452 +trainer/Advantage Weights Std 15.5552 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.16463e-17 +trainer/Advantage Score Mean -0.417375 +trainer/Advantage Score Std 0.64965 +trainer/Advantage Score Max 0.933067 +trainer/Advantage Score Min -3.89915 +trainer/V1 Predictions Mean -72.0968 +trainer/V1 Predictions Std 19.7056 +trainer/V1 Predictions Max -0.60899 +trainer/V1 Predictions Min -87.2729 +trainer/VF Loss 0.0739515 +expl/num steps total 788000 +expl/num paths total 1071 +expl/path length Mean 500 +expl/path length Std 6 +expl/path length Max 506 +expl/path length Min 494 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0442245 +expl/Actions Std 0.824834 +expl/Actions Max 2.63394 +expl/Actions Min -2.39512 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 717815 +eval/num paths total 793 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0187395 +eval/Actions Std 0.714014 +eval/Actions Max 0.999906 +eval/Actions Min -0.999935 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.99003e-06 +time/evaluation sampling (s) 5.6906 +time/exploration sampling (s) 6.95822 +time/logging (s) 0.0127381 +time/saving (s) 0.0193197 +time/training (s) 18.4003 +time/epoch (s) 31.0812 +time/total (s) 20473.9 +Epoch -213 +------------------------------ ---------------- +2022-05-15 23:44:14.120779 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -212 finished +------------------------------ ---------------- +epoch -212 +replay_buffer/size 999047 +trainer/num train calls 789000 +trainer/QF1 Loss 0.762302 +trainer/QF2 Loss 0.711364 +trainer/Policy Loss 30.3119 +trainer/Q1 Predictions Mean -77.1864 +trainer/Q1 Predictions Std 13.8597 +trainer/Q1 Predictions Max -1.40579 +trainer/Q1 Predictions Min -87.6876 +trainer/Q2 Predictions Mean -77.194 +trainer/Q2 Predictions Std 13.7844 +trainer/Q2 Predictions Max -1.95576 +trainer/Q2 Predictions Min -87.7224 +trainer/Q Targets Mean -77.4188 +trainer/Q Targets Std 13.8386 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2298 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0022164 +trainer/policy/mean Std 0.742483 +trainer/policy/mean Max 0.999616 +trainer/policy/mean Min -0.999609 +trainer/policy/std Mean 0.408849 +trainer/policy/std Std 0.0188054 +trainer/policy/std Max 0.430993 +trainer/policy/std Min 0.381074 +trainer/Advantage Weights Mean 7.43013 +trainer/Advantage Weights Std 21.6451 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.20635e-13 +trainer/Advantage Score Mean -0.224197 +trainer/Advantage Score Std 0.533107 +trainer/Advantage Score Max 0.946816 +trainer/Advantage Score Min -2.91423 +trainer/V1 Predictions Mean -77.1371 +trainer/V1 Predictions Std 14.1599 +trainer/V1 Predictions Max -0.702672 +trainer/V1 Predictions Min -88.0783 +trainer/VF Loss 0.0550124 +expl/num steps total 789000 +expl/num paths total 1073 +expl/path length Mean 500 +expl/path length Std 332 +expl/path length Max 832 +expl/path length Min 168 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0443406 +expl/Actions Std 0.829743 +expl/Actions Max 2.48718 +expl/Actions Min -2.27612 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 718512 +eval/num paths total 794 +eval/path length Mean 697 +eval/path length Std 0 +eval/path length Max 697 +eval/path length Min 697 +eval/Rewards Mean 0.00143472 +eval/Rewards Std 0.0378505 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0371453 +eval/Actions Std 0.749497 +eval/Actions Max 0.99991 +eval/Actions Min -0.999816 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.84007e-06 +time/evaluation sampling (s) 4.7232 +time/exploration sampling (s) 6.25875 +time/logging (s) 0.00706754 +time/saving (s) 0.0143988 +time/training (s) 18.8353 +time/epoch (s) 29.8387 +time/total (s) 20503.7 +Epoch -212 +------------------------------ ---------------- +2022-05-15 23:44:44.632240 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -211 finished +------------------------------ ---------------- +epoch -211 +replay_buffer/size 999047 +trainer/num train calls 790000 +trainer/QF1 Loss 7.81302 +trainer/QF2 Loss 7.90334 +trainer/Policy Loss 8.8581 +trainer/Q1 Predictions Mean -74.7939 +trainer/Q1 Predictions Std 16.9826 +trainer/Q1 Predictions Max -1.22609 +trainer/Q1 Predictions Min -87.8999 +trainer/Q2 Predictions Mean -74.8818 +trainer/Q2 Predictions Std 17.0215 +trainer/Q2 Predictions Max -1.09545 +trainer/Q2 Predictions Min -88.218 +trainer/Q Targets Mean -74.8979 +trainer/Q Targets Std 16.7017 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7593 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0436038 +trainer/policy/mean Std 0.726821 +trainer/policy/mean Max 0.999735 +trainer/policy/mean Min -0.998304 +trainer/policy/std Mean 0.408836 +trainer/policy/std Std 0.0194103 +trainer/policy/std Max 0.43084 +trainer/policy/std Min 0.377462 +trainer/Advantage Weights Mean 2.77092 +trainer/Advantage Weights Std 14.1882 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.14194e-15 +trainer/Advantage Score Mean -0.451913 +trainer/Advantage Score Std 0.563069 +trainer/Advantage Score Max 1.55205 +trainer/Advantage Score Min -3.44061 +trainer/V1 Predictions Mean -74.4058 +trainer/V1 Predictions Std 17.162 +trainer/V1 Predictions Max -1.26667 +trainer/V1 Predictions Min -87.6674 +trainer/VF Loss 0.0671309 +expl/num steps total 790000 +expl/num paths total 1075 +expl/path length Mean 500 +expl/path length Std 182 +expl/path length Max 682 +expl/path length Min 318 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00296971 +expl/Actions Std 0.801594 +expl/Actions Max 2.29617 +expl/Actions Min -2.32801 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 719512 +eval/num paths total 795 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.110787 +eval/Actions Std 0.618329 +eval/Actions Max 0.999922 +eval/Actions Min -0.999368 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.09177e-06 +time/evaluation sampling (s) 4.47981 +time/exploration sampling (s) 6.85909 +time/logging (s) 0.0123673 +time/saving (s) 0.0171588 +time/training (s) 19.1358 +time/epoch (s) 30.5042 +time/total (s) 20534.3 +Epoch -211 +------------------------------ ---------------- +2022-05-15 23:45:15.378209 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -210 finished +------------------------------ ---------------- +epoch -210 +replay_buffer/size 999047 +trainer/num train calls 791000 +trainer/QF1 Loss 0.7244 +trainer/QF2 Loss 0.890293 +trainer/Policy Loss 28.2678 +trainer/Q1 Predictions Mean -74.0825 +trainer/Q1 Predictions Std 16.8613 +trainer/Q1 Predictions Max -1.7166 +trainer/Q1 Predictions Min -87.6051 +trainer/Q2 Predictions Mean -74.1002 +trainer/Q2 Predictions Std 16.8094 +trainer/Q2 Predictions Max -2.1334 +trainer/Q2 Predictions Min -87.4589 +trainer/Q Targets Mean -74.115 +trainer/Q Targets Std 16.7612 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9651 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00168855 +trainer/policy/mean Std 0.725045 +trainer/policy/mean Max 0.999099 +trainer/policy/mean Min -0.999358 +trainer/policy/std Mean 0.408955 +trainer/policy/std Std 0.0199971 +trainer/policy/std Max 0.434196 +trainer/policy/std Min 0.382302 +trainer/Advantage Weights Mean 5.9662 +trainer/Advantage Weights Std 20.9488 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.18839e-28 +trainer/Advantage Score Mean -0.457519 +trainer/Advantage Score Std 0.725799 +trainer/Advantage Score Max 0.974007 +trainer/Advantage Score Min -6.26497 +trainer/V1 Predictions Mean -73.8948 +trainer/V1 Predictions Std 16.9619 +trainer/V1 Predictions Max -2.29674 +trainer/V1 Predictions Min -87.2676 +trainer/VF Loss 0.0975559 +expl/num steps total 791000 +expl/num paths total 1076 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00157196 +expl/Actions Std 0.865171 +expl/Actions Max 2.21156 +expl/Actions Min -2.3685 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 720088 +eval/num paths total 796 +eval/path length Mean 576 +eval/path length Std 0 +eval/path length Max 576 +eval/path length Min 576 +eval/Rewards Mean 0.00173611 +eval/Rewards Std 0.0416305 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0468946 +eval/Actions Std 0.737311 +eval/Actions Max 0.999976 +eval/Actions Min -0.999823 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.1072e-05 +time/evaluation sampling (s) 5.01993 +time/exploration sampling (s) 6.75433 +time/logging (s) 0.00617312 +time/saving (s) 0.0120211 +time/training (s) 18.9265 +time/epoch (s) 30.719 +time/total (s) 20565 +Epoch -210 +------------------------------ ---------------- +2022-05-15 23:45:46.319741 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -209 finished +------------------------------ ---------------- +epoch -209 +replay_buffer/size 999047 +trainer/num train calls 792000 +trainer/QF1 Loss 1.69523 +trainer/QF2 Loss 1.83134 +trainer/Policy Loss 9.56796 +trainer/Q1 Predictions Mean -71.0808 +trainer/Q1 Predictions Std 20.3218 +trainer/Q1 Predictions Max -0.225778 +trainer/Q1 Predictions Min -88.4731 +trainer/Q2 Predictions Mean -71.1818 +trainer/Q2 Predictions Std 20.2818 +trainer/Q2 Predictions Max -0.171827 +trainer/Q2 Predictions Min -88.7102 +trainer/Q Targets Mean -70.5038 +trainer/Q Targets Std 20.0034 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8153 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00414211 +trainer/policy/mean Std 0.724555 +trainer/policy/mean Max 0.999181 +trainer/policy/mean Min -0.998335 +trainer/policy/std Mean 0.409358 +trainer/policy/std Std 0.0194458 +trainer/policy/std Max 0.432047 +trainer/policy/std Min 0.379402 +trainer/Advantage Weights Mean 2.7639 +trainer/Advantage Weights Std 15.4964 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.1824e-14 +trainer/Advantage Score Mean -0.706143 +trainer/Advantage Score Std 0.632042 +trainer/Advantage Score Max 1.74254 +trainer/Advantage Score Min -3.14558 +trainer/V1 Predictions Mean -70.2059 +trainer/V1 Predictions Std 20.2877 +trainer/V1 Predictions Max 0.308608 +trainer/V1 Predictions Min -87.7178 +trainer/VF Loss 0.113106 +expl/num steps total 792000 +expl/num paths total 1077 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.000726295 +expl/Actions Std 0.82688 +expl/Actions Max 2.12317 +expl/Actions Min -2.17664 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 721088 +eval/num paths total 797 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0570034 +eval/Actions Std 0.745756 +eval/Actions Max 0.999939 +eval/Actions Min -0.999825 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.8932e-06 +time/evaluation sampling (s) 5.14383 +time/exploration sampling (s) 7.06373 +time/logging (s) 0.00780537 +time/saving (s) 0.0108554 +time/training (s) 18.7065 +time/epoch (s) 30.9327 +time/total (s) 20595.9 +Epoch -209 +------------------------------ ---------------- +2022-05-15 23:46:16.385352 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -208 finished +------------------------------ ---------------- +epoch -208 +replay_buffer/size 999047 +trainer/num train calls 793000 +trainer/QF1 Loss 0.593887 +trainer/QF2 Loss 0.57403 +trainer/Policy Loss 25.8844 +trainer/Q1 Predictions Mean -73.9195 +trainer/Q1 Predictions Std 17.3916 +trainer/Q1 Predictions Max -0.571003 +trainer/Q1 Predictions Min -87.8343 +trainer/Q2 Predictions Mean -73.8687 +trainer/Q2 Predictions Std 17.4533 +trainer/Q2 Predictions Max -1.88066 +trainer/Q2 Predictions Min -87.7632 +trainer/Q Targets Mean -73.9354 +trainer/Q Targets Std 17.5411 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0359 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.007174 +trainer/policy/mean Std 0.741452 +trainer/policy/mean Max 0.999863 +trainer/policy/mean Min -0.999493 +trainer/policy/std Mean 0.409789 +trainer/policy/std Std 0.0178881 +trainer/policy/std Max 0.43131 +trainer/policy/std Min 0.383408 +trainer/Advantage Weights Mean 4.58013 +trainer/Advantage Weights Std 17.6888 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.53196e-28 +trainer/Advantage Score Mean -0.423038 +trainer/Advantage Score Std 0.718778 +trainer/Advantage Score Max 0.985363 +trainer/Advantage Score Min -6.27618 +trainer/V1 Predictions Mean -73.6307 +trainer/V1 Predictions Std 17.7759 +trainer/V1 Predictions Max -0.200214 +trainer/V1 Predictions Min -87.9118 +trainer/VF Loss 0.0825739 +expl/num steps total 793000 +expl/num paths total 1078 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.153498 +expl/Actions Std 0.871333 +expl/Actions Max 2.12423 +expl/Actions Min -2.38622 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 721548 +eval/num paths total 798 +eval/path length Mean 460 +eval/path length Std 0 +eval/path length Max 460 +eval/path length Min 460 +eval/Rewards Mean 0.00217391 +eval/Rewards Std 0.0465745 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0390386 +eval/Actions Std 0.743626 +eval/Actions Max 0.999869 +eval/Actions Min -0.999421 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.39886e-06 +time/evaluation sampling (s) 5.04071 +time/exploration sampling (s) 6.14013 +time/logging (s) 0.00823452 +time/saving (s) 0.0146541 +time/training (s) 18.8501 +time/epoch (s) 30.0539 +time/total (s) 20626 +Epoch -208 +------------------------------ ---------------- +2022-05-15 23:46:47.463830 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -207 finished +------------------------------ ---------------- +epoch -207 +replay_buffer/size 999047 +trainer/num train calls 794000 +trainer/QF1 Loss 0.787512 +trainer/QF2 Loss 0.751644 +trainer/Policy Loss 16.769 +trainer/Q1 Predictions Mean -72.5204 +trainer/Q1 Predictions Std 20.5155 +trainer/Q1 Predictions Max -1.54333 +trainer/Q1 Predictions Min -87.9502 +trainer/Q2 Predictions Mean -72.5352 +trainer/Q2 Predictions Std 20.4675 +trainer/Q2 Predictions Max -1.22418 +trainer/Q2 Predictions Min -87.8929 +trainer/Q Targets Mean -72.5034 +trainer/Q Targets Std 20.4852 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8265 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0174577 +trainer/policy/mean Std 0.727266 +trainer/policy/mean Max 0.999425 +trainer/policy/mean Min -0.998964 +trainer/policy/std Mean 0.409058 +trainer/policy/std Std 0.0182249 +trainer/policy/std Max 0.431489 +trainer/policy/std Min 0.381476 +trainer/Advantage Weights Mean 4.27419 +trainer/Advantage Weights Std 16.8334 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.57581e-24 +trainer/Advantage Score Mean -0.463597 +trainer/Advantage Score Std 0.737414 +trainer/Advantage Score Max 0.74353 +trainer/Advantage Score Min -5.48073 +trainer/V1 Predictions Mean -72.1815 +trainer/V1 Predictions Std 20.7659 +trainer/V1 Predictions Max -0.219967 +trainer/V1 Predictions Min -87.6619 +trainer/VF Loss 0.087677 +expl/num steps total 794000 +expl/num paths total 1080 +expl/path length Mean 500 +expl/path length Std 361 +expl/path length Max 861 +expl/path length Min 139 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0230438 +expl/Actions Std 0.820126 +expl/Actions Max 2.4049 +expl/Actions Min -2.53335 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 722548 +eval/num paths total 799 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.204008 +eval/Actions Std 0.819737 +eval/Actions Max 0.999389 +eval/Actions Min -0.998335 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.86685e-06 +time/evaluation sampling (s) 5.1496 +time/exploration sampling (s) 6.25786 +time/logging (s) 0.0120206 +time/saving (s) 0.0171708 +time/training (s) 19.6287 +time/epoch (s) 31.0654 +time/total (s) 20657.1 +Epoch -207 +------------------------------ ---------------- +2022-05-15 23:47:19.255652 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -206 finished +------------------------------ ---------------- +epoch -206 +replay_buffer/size 999047 +trainer/num train calls 795000 +trainer/QF1 Loss 0.69211 +trainer/QF2 Loss 0.57236 +trainer/Policy Loss 8.50444 +trainer/Q1 Predictions Mean -73.918 +trainer/Q1 Predictions Std 17.9905 +trainer/Q1 Predictions Max -1.10135 +trainer/Q1 Predictions Min -87.8098 +trainer/Q2 Predictions Mean -73.9085 +trainer/Q2 Predictions Std 17.9682 +trainer/Q2 Predictions Max -0.552322 +trainer/Q2 Predictions Min -87.7009 +trainer/Q Targets Mean -73.9765 +trainer/Q Targets Std 17.683 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5904 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0259991 +trainer/policy/mean Std 0.741116 +trainer/policy/mean Max 0.998959 +trainer/policy/mean Min -0.999615 +trainer/policy/std Mean 0.407324 +trainer/policy/std Std 0.0193409 +trainer/policy/std Max 0.430042 +trainer/policy/std Min 0.378602 +trainer/Advantage Weights Mean 2.14265 +trainer/Advantage Weights Std 12.7472 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.24505e-18 +trainer/Advantage Score Mean -0.597963 +trainer/Advantage Score Std 0.618883 +trainer/Advantage Score Max 1.57853 +trainer/Advantage Score Min -4.12274 +trainer/V1 Predictions Mean -73.7031 +trainer/V1 Predictions Std 17.8082 +trainer/V1 Predictions Max -0.273441 +trainer/V1 Predictions Min -87.5355 +trainer/VF Loss 0.0924031 +expl/num steps total 795000 +expl/num paths total 1081 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0947561 +expl/Actions Std 0.929465 +expl/Actions Max 2.24039 +expl/Actions Min -2.34322 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 723548 +eval/num paths total 800 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0577696 +eval/Actions Std 0.728712 +eval/Actions Max 0.999641 +eval/Actions Min -0.999768 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.02608e-05 +time/evaluation sampling (s) 5.1089 +time/exploration sampling (s) 7.68241 +time/logging (s) 0.0105904 +time/saving (s) 0.0136526 +time/training (s) 18.9544 +time/epoch (s) 31.77 +time/total (s) 20688.8 +Epoch -206 +------------------------------ ---------------- +2022-05-15 23:47:50.055736 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -205 finished +------------------------------ ---------------- +epoch -205 +replay_buffer/size 999047 +trainer/num train calls 796000 +trainer/QF1 Loss 0.827045 +trainer/QF2 Loss 0.855226 +trainer/Policy Loss 7.72498 +trainer/Q1 Predictions Mean -75.4596 +trainer/Q1 Predictions Std 17.1985 +trainer/Q1 Predictions Max -7.37051 +trainer/Q1 Predictions Min -87.8506 +trainer/Q2 Predictions Mean -75.4613 +trainer/Q2 Predictions Std 17.2693 +trainer/Q2 Predictions Max -7.39073 +trainer/Q2 Predictions Min -88.3993 +trainer/Q Targets Mean -75.2478 +trainer/Q Targets Std 17.4386 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0724 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0345032 +trainer/policy/mean Std 0.741537 +trainer/policy/mean Max 0.999108 +trainer/policy/mean Min -0.999691 +trainer/policy/std Mean 0.40788 +trainer/policy/std Std 0.0200024 +trainer/policy/std Max 0.427644 +trainer/policy/std Min 0.374489 +trainer/Advantage Weights Mean 2.18316 +trainer/Advantage Weights Std 12.5178 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.31078e-17 +trainer/Advantage Score Mean -0.421918 +trainer/Advantage Score Std 0.637212 +trainer/Advantage Score Max 2.79689 +trainer/Advantage Score Min -3.88733 +trainer/V1 Predictions Mean -75.0378 +trainer/V1 Predictions Std 17.4114 +trainer/V1 Predictions Max -4.88533 +trainer/V1 Predictions Min -87.7367 +trainer/VF Loss 0.0886238 +expl/num steps total 796000 +expl/num paths total 1083 +expl/path length Mean 500 +expl/path length Std 154 +expl/path length Max 654 +expl/path length Min 346 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0410242 +expl/Actions Std 0.85621 +expl/Actions Max 2.2596 +expl/Actions Min -2.49659 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 724548 +eval/num paths total 801 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.152998 +eval/Actions Std 0.751343 +eval/Actions Max 0.999975 +eval/Actions Min -0.99996 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.90805e-06 +time/evaluation sampling (s) 5.43208 +time/exploration sampling (s) 6.5897 +time/logging (s) 0.00976999 +time/saving (s) 0.0146645 +time/training (s) 18.7338 +time/epoch (s) 30.7801 +time/total (s) 20719.6 +Epoch -205 +------------------------------ ---------------- +2022-05-15 23:48:20.496178 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -204 finished +------------------------------ ---------------- +epoch -204 +replay_buffer/size 999047 +trainer/num train calls 797000 +trainer/QF1 Loss 0.490528 +trainer/QF2 Loss 0.513487 +trainer/Policy Loss 21.1862 +trainer/Q1 Predictions Mean -75.7716 +trainer/Q1 Predictions Std 14.5683 +trainer/Q1 Predictions Max -2.35308 +trainer/Q1 Predictions Min -88.5689 +trainer/Q2 Predictions Mean -75.8056 +trainer/Q2 Predictions Std 14.6226 +trainer/Q2 Predictions Max -3.28029 +trainer/Q2 Predictions Min -88.4516 +trainer/Q Targets Mean -76.0532 +trainer/Q Targets Std 14.6714 +trainer/Q Targets Max -1.61121 +trainer/Q Targets Min -88.512 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0296772 +trainer/policy/mean Std 0.729453 +trainer/policy/mean Max 0.999345 +trainer/policy/mean Min -0.999357 +trainer/policy/std Mean 0.409174 +trainer/policy/std Std 0.0193763 +trainer/policy/std Max 0.428981 +trainer/policy/std Min 0.376435 +trainer/Advantage Weights Mean 6.5309 +trainer/Advantage Weights Std 21.956 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.23253e-18 +trainer/Advantage Score Mean -0.303384 +trainer/Advantage Score Std 0.604468 +trainer/Advantage Score Max 0.908501 +trainer/Advantage Score Min -4.02733 +trainer/V1 Predictions Mean -75.7849 +trainer/V1 Predictions Std 14.8612 +trainer/V1 Predictions Max 0.32493 +trainer/V1 Predictions Min -88.474 +trainer/VF Loss 0.0656506 +expl/num steps total 797000 +expl/num paths total 1085 +expl/path length Mean 500 +expl/path length Std 477 +expl/path length Max 977 +expl/path length Min 23 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0344163 +expl/Actions Std 0.816763 +expl/Actions Max 2.2664 +expl/Actions Min -2.27063 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 725548 +eval/num paths total 802 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.144411 +eval/Actions Std 0.731207 +eval/Actions Max 0.999805 +eval/Actions Min -0.999795 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.30418e-05 +time/evaluation sampling (s) 4.89837 +time/exploration sampling (s) 6.36435 +time/logging (s) 0.0122296 +time/saving (s) 0.0169667 +time/training (s) 19.136 +time/epoch (s) 30.4279 +time/total (s) 20750 +Epoch -204 +------------------------------ ---------------- +2022-05-15 23:48:51.800199 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -203 finished +------------------------------ ---------------- +epoch -203 +replay_buffer/size 999047 +trainer/num train calls 798000 +trainer/QF1 Loss 0.446726 +trainer/QF2 Loss 0.456707 +trainer/Policy Loss 12.8324 +trainer/Q1 Predictions Mean -73.4905 +trainer/Q1 Predictions Std 19.3395 +trainer/Q1 Predictions Max -0.571182 +trainer/Q1 Predictions Min -87.2137 +trainer/Q2 Predictions Mean -73.5354 +trainer/Q2 Predictions Std 19.3104 +trainer/Q2 Predictions Max -0.301603 +trainer/Q2 Predictions Min -86.9141 +trainer/Q Targets Mean -73.718 +trainer/Q Targets Std 19.4849 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4681 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00761371 +trainer/policy/mean Std 0.735997 +trainer/policy/mean Max 0.99863 +trainer/policy/mean Min -0.998797 +trainer/policy/std Mean 0.407331 +trainer/policy/std Std 0.0194973 +trainer/policy/std Max 0.426904 +trainer/policy/std Min 0.376839 +trainer/Advantage Weights Mean 2.90859 +trainer/Advantage Weights Std 14.8101 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.30725e-14 +trainer/Advantage Score Mean -0.466593 +trainer/Advantage Score Std 0.514999 +trainer/Advantage Score Max 0.889206 +trainer/Advantage Score Min -3.19683 +trainer/V1 Predictions Mean -73.5032 +trainer/V1 Predictions Std 19.5191 +trainer/V1 Predictions Max -0.568511 +trainer/V1 Predictions Min -87.34 +trainer/VF Loss 0.0573644 +expl/num steps total 798000 +expl/num paths total 1086 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.313109 +expl/Actions Std 0.856734 +expl/Actions Max 2.23893 +expl/Actions Min -2.26313 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 726548 +eval/num paths total 803 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0303072 +eval/Actions Std 0.726173 +eval/Actions Max 0.999976 +eval/Actions Min -0.999855 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78186e-06 +time/evaluation sampling (s) 4.87855 +time/exploration sampling (s) 7.67425 +time/logging (s) 0.0124614 +time/saving (s) 0.0185038 +time/training (s) 18.7001 +time/epoch (s) 31.2838 +time/total (s) 20781.3 +Epoch -203 +------------------------------ ---------------- +2022-05-15 23:49:20.480406 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -202 finished +------------------------------ ---------------- +epoch -202 +replay_buffer/size 999047 +trainer/num train calls 799000 +trainer/QF1 Loss 0.636945 +trainer/QF2 Loss 0.740158 +trainer/Policy Loss 34.7016 +trainer/Q1 Predictions Mean -75.3657 +trainer/Q1 Predictions Std 17.2197 +trainer/Q1 Predictions Max 0.0323869 +trainer/Q1 Predictions Min -88.5849 +trainer/Q2 Predictions Mean -75.3293 +trainer/Q2 Predictions Std 17.1534 +trainer/Q2 Predictions Max -0.205823 +trainer/Q2 Predictions Min -88.5075 +trainer/Q Targets Mean -75.5166 +trainer/Q Targets Std 17.0074 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.3229 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0272607 +trainer/policy/mean Std 0.734103 +trainer/policy/mean Max 0.999275 +trainer/policy/mean Min -0.999802 +trainer/policy/std Mean 0.409708 +trainer/policy/std Std 0.020897 +trainer/policy/std Max 0.430189 +trainer/policy/std Min 0.376731 +trainer/Advantage Weights Mean 8.5319 +trainer/Advantage Weights Std 23.8478 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.93104e-10 +trainer/Advantage Score Mean -0.240071 +trainer/Advantage Score Std 0.538335 +trainer/Advantage Score Max 2.05272 +trainer/Advantage Score Min -2.19505 +trainer/V1 Predictions Mean -75.277 +trainer/V1 Predictions Std 16.9788 +trainer/V1 Predictions Max -0.225633 +trainer/V1 Predictions Min -88.2201 +trainer/VF Loss 0.0859248 +expl/num steps total 799000 +expl/num paths total 1088 +expl/path length Mean 500 +expl/path length Std 134 +expl/path length Max 634 +expl/path length Min 366 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00356583 +expl/Actions Std 0.821499 +expl/Actions Max 2.62399 +expl/Actions Min -2.48481 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 727136 +eval/num paths total 804 +eval/path length Mean 588 +eval/path length Std 0 +eval/path length Max 588 +eval/path length Min 588 +eval/Rewards Mean 0.00170068 +eval/Rewards Std 0.0412042 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00928166 +eval/Actions Std 0.745528 +eval/Actions Max 0.999952 +eval/Actions Min -0.999864 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.88803e-06 +time/evaluation sampling (s) 4.90053 +time/exploration sampling (s) 5.52498 +time/logging (s) 0.00982236 +time/saving (s) 0.0172832 +time/training (s) 18.2042 +time/epoch (s) 28.6568 +time/total (s) 20810 +Epoch -202 +------------------------------ ---------------- +2022-05-15 23:49:47.911541 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -201 finished +------------------------------ ---------------- +epoch -201 +replay_buffer/size 999047 +trainer/num train calls 800000 +trainer/QF1 Loss 2.24654 +trainer/QF2 Loss 1.95762 +trainer/Policy Loss 8.41507 +trainer/Q1 Predictions Mean -73.1143 +trainer/Q1 Predictions Std 19.9431 +trainer/Q1 Predictions Max -0.941404 +trainer/Q1 Predictions Min -87.3362 +trainer/Q2 Predictions Mean -73.2032 +trainer/Q2 Predictions Std 19.9564 +trainer/Q2 Predictions Max -1.35307 +trainer/Q2 Predictions Min -87.2793 +trainer/Q Targets Mean -72.9305 +trainer/Q Targets Std 20.1886 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1264 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00314263 +trainer/policy/mean Std 0.72911 +trainer/policy/mean Max 0.999599 +trainer/policy/mean Min -0.998235 +trainer/policy/std Mean 0.409814 +trainer/policy/std Std 0.0197077 +trainer/policy/std Max 0.432274 +trainer/policy/std Min 0.378751 +trainer/Advantage Weights Mean 2.95299 +trainer/Advantage Weights Std 14.5377 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.64465e-19 +trainer/Advantage Score Mean -0.513661 +trainer/Advantage Score Std 0.723711 +trainer/Advantage Score Max 1.13572 +trainer/Advantage Score Min -4.22134 +trainer/V1 Predictions Mean -72.7058 +trainer/V1 Predictions Std 20.147 +trainer/V1 Predictions Max -0.301064 +trainer/V1 Predictions Min -87.0396 +trainer/VF Loss 0.092259 +expl/num steps total 800000 +expl/num paths total 1090 +expl/path length Mean 500 +expl/path length Std 295 +expl/path length Max 795 +expl/path length Min 205 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0639081 +expl/Actions Std 0.820164 +expl/Actions Max 2.37085 +expl/Actions Min -2.47207 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 728136 +eval/num paths total 805 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0269073 +eval/Actions Std 0.727209 +eval/Actions Max 0.999874 +eval/Actions Min -0.999826 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 6.4848e-06 +time/evaluation sampling (s) 3.9789 +time/exploration sampling (s) 5.32654 +time/logging (s) 0.00878726 +time/saving (s) 0.0104568 +time/training (s) 18.0849 +time/epoch (s) 27.4096 +time/total (s) 20837.4 +Epoch -201 +------------------------------ ---------------- +2022-05-15 23:50:17.880607 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -200 finished +------------------------------ ---------------- +epoch -200 +replay_buffer/size 999047 +trainer/num train calls 801000 +trainer/QF1 Loss 0.477078 +trainer/QF2 Loss 0.546102 +trainer/Policy Loss 16.2305 +trainer/Q1 Predictions Mean -73.576 +trainer/Q1 Predictions Std 19.8647 +trainer/Q1 Predictions Max -0.338347 +trainer/Q1 Predictions Min -88.3027 +trainer/Q2 Predictions Mean -73.6002 +trainer/Q2 Predictions Std 19.8222 +trainer/Q2 Predictions Max -0.60351 +trainer/Q2 Predictions Min -88.0302 +trainer/Q Targets Mean -73.5401 +trainer/Q Targets Std 19.8217 +trainer/Q Targets Max -1.42303 +trainer/Q Targets Min -87.9043 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00728739 +trainer/policy/mean Std 0.72981 +trainer/policy/mean Max 0.999726 +trainer/policy/mean Min -0.998165 +trainer/policy/std Mean 0.410014 +trainer/policy/std Std 0.0186532 +trainer/policy/std Max 0.430193 +trainer/policy/std Min 0.378813 +trainer/Advantage Weights Mean 5.10431 +trainer/Advantage Weights Std 19.744 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.02239e-14 +trainer/Advantage Score Mean -0.30255 +trainer/Advantage Score Std 0.499936 +trainer/Advantage Score Max 2.33535 +trainer/Advantage Score Min -3.11301 +trainer/V1 Predictions Mean -73.3766 +trainer/V1 Predictions Std 19.8694 +trainer/V1 Predictions Max 0.549803 +trainer/V1 Predictions Min -87.7842 +trainer/VF Loss 0.0760974 +expl/num steps total 801000 +expl/num paths total 1091 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0215768 +expl/Actions Std 0.829175 +expl/Actions Max 2.37923 +expl/Actions Min -2.37159 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 728784 +eval/num paths total 806 +eval/path length Mean 648 +eval/path length Std 0 +eval/path length Max 648 +eval/path length Min 648 +eval/Rewards Mean 0.00154321 +eval/Rewards Std 0.0392534 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0260253 +eval/Actions Std 0.739697 +eval/Actions Max 0.999718 +eval/Actions Min -0.999683 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.92016e-06 +time/evaluation sampling (s) 4.50146 +time/exploration sampling (s) 7.3036 +time/logging (s) 0.00904233 +time/saving (s) 0.0266654 +time/training (s) 18.1179 +time/epoch (s) 29.9587 +time/total (s) 20867.4 +Epoch -200 +------------------------------ ---------------- +2022-05-15 23:50:45.965694 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -199 finished +------------------------------ ---------------- +epoch -199 +replay_buffer/size 999047 +trainer/num train calls 802000 +trainer/QF1 Loss 1.15858 +trainer/QF2 Loss 1.28471 +trainer/Policy Loss 8.65101 +trainer/Q1 Predictions Mean -74.8063 +trainer/Q1 Predictions Std 18.7048 +trainer/Q1 Predictions Max -0.212557 +trainer/Q1 Predictions Min -87.5358 +trainer/Q2 Predictions Mean -74.6829 +trainer/Q2 Predictions Std 18.7226 +trainer/Q2 Predictions Max -0.139808 +trainer/Q2 Predictions Min -87.4982 +trainer/Q Targets Mean -74.7349 +trainer/Q Targets Std 18.5426 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8045 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00929561 +trainer/policy/mean Std 0.729712 +trainer/policy/mean Max 0.999275 +trainer/policy/mean Min -0.999341 +trainer/policy/std Mean 0.410396 +trainer/policy/std Std 0.0184218 +trainer/policy/std Max 0.431341 +trainer/policy/std Min 0.380572 +trainer/Advantage Weights Mean 1.81543 +trainer/Advantage Weights Std 11.3353 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.69813e-15 +trainer/Advantage Score Mean -0.488627 +trainer/Advantage Score Std 0.541941 +trainer/Advantage Score Max 2.59322 +trainer/Advantage Score Min -3.35462 +trainer/V1 Predictions Mean -74.5092 +trainer/V1 Predictions Std 18.6665 +trainer/V1 Predictions Max 0.52081 +trainer/V1 Predictions Min -87.6457 +trainer/VF Loss 0.0863182 +expl/num steps total 802000 +expl/num paths total 1093 +expl/path length Mean 500 +expl/path length Std 342 +expl/path length Max 842 +expl/path length Min 158 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0124393 +expl/Actions Std 0.852159 +expl/Actions Max 2.30687 +expl/Actions Min -2.23436 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 729233 +eval/num paths total 807 +eval/path length Mean 449 +eval/path length Std 0 +eval/path length Max 449 +eval/path length Min 449 +eval/Rewards Mean 0.00222717 +eval/Rewards Std 0.0471403 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0353651 +eval/Actions Std 0.742161 +eval/Actions Max 0.999846 +eval/Actions Min -0.999251 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 5.63031e-06 +time/evaluation sampling (s) 4.29741 +time/exploration sampling (s) 5.56903 +time/logging (s) 0.00829734 +time/saving (s) 0.0149647 +time/training (s) 18.1767 +time/epoch (s) 28.0664 +time/total (s) 20895.5 +Epoch -199 +------------------------------ ---------------- +2022-05-15 23:51:14.773421 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -198 finished +------------------------------ ---------------- +epoch -198 +replay_buffer/size 999047 +trainer/num train calls 803000 +trainer/QF1 Loss 0.655802 +trainer/QF2 Loss 0.74636 +trainer/Policy Loss 22.0482 +trainer/Q1 Predictions Mean -75.388 +trainer/Q1 Predictions Std 17.3061 +trainer/Q1 Predictions Max -1.03859 +trainer/Q1 Predictions Min -87.9061 +trainer/Q2 Predictions Mean -75.4893 +trainer/Q2 Predictions Std 17.244 +trainer/Q2 Predictions Max -0.279269 +trainer/Q2 Predictions Min -88.1799 +trainer/Q Targets Mean -75.2068 +trainer/Q Targets Std 17.3726 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7882 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0145448 +trainer/policy/mean Std 0.73208 +trainer/policy/mean Max 0.99971 +trainer/policy/mean Min -0.99974 +trainer/policy/std Mean 0.40709 +trainer/policy/std Std 0.0191914 +trainer/policy/std Max 0.427746 +trainer/policy/std Min 0.37684 +trainer/Advantage Weights Mean 2.70282 +trainer/Advantage Weights Std 14.1565 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.58061e-20 +trainer/Advantage Score Mean -0.386669 +trainer/Advantage Score Std 0.575771 +trainer/Advantage Score Max 1.02064 +trainer/Advantage Score Min -4.55939 +trainer/V1 Predictions Mean -74.9867 +trainer/V1 Predictions Std 17.5141 +trainer/V1 Predictions Max -0.493267 +trainer/V1 Predictions Min -87.8443 +trainer/VF Loss 0.0568054 +expl/num steps total 803000 +expl/num paths total 1094 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0393507 +expl/Actions Std 0.811546 +expl/Actions Max 2.24979 +expl/Actions Min -2.27875 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 729661 +eval/num paths total 808 +eval/path length Mean 428 +eval/path length Std 0 +eval/path length Max 428 +eval/path length Min 428 +eval/Rewards Mean 0.00233645 +eval/Rewards Std 0.0482803 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0132799 +eval/Actions Std 0.741492 +eval/Actions Max 0.999616 +eval/Actions Min -0.999911 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.9509e-06 +time/evaluation sampling (s) 4.28535 +time/exploration sampling (s) 6.16228 +time/logging (s) 0.00603863 +time/saving (s) 0.0129841 +time/training (s) 18.3236 +time/epoch (s) 28.7903 +time/total (s) 20924.2 +Epoch -198 +------------------------------ ---------------- +2022-05-15 23:51:43.462178 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -197 finished +------------------------------ ---------------- +epoch -197 +replay_buffer/size 999047 +trainer/num train calls 804000 +trainer/QF1 Loss 0.667181 +trainer/QF2 Loss 0.607561 +trainer/Policy Loss 22.9561 +trainer/Q1 Predictions Mean -72.181 +trainer/Q1 Predictions Std 21.4069 +trainer/Q1 Predictions Max -0.723821 +trainer/Q1 Predictions Min -88.4947 +trainer/Q2 Predictions Mean -72.2002 +trainer/Q2 Predictions Std 21.4415 +trainer/Q2 Predictions Max 0.171072 +trainer/Q2 Predictions Min -88.3279 +trainer/Q Targets Mean -71.9477 +trainer/Q Targets Std 21.4011 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8782 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0334593 +trainer/policy/mean Std 0.733335 +trainer/policy/mean Max 0.998321 +trainer/policy/mean Min -0.999657 +trainer/policy/std Mean 0.408995 +trainer/policy/std Std 0.0198954 +trainer/policy/std Max 0.430195 +trainer/policy/std Min 0.375844 +trainer/Advantage Weights Mean 3.88565 +trainer/Advantage Weights Std 17.5786 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.6223e-18 +trainer/Advantage Score Mean -0.481814 +trainer/Advantage Score Std 0.573389 +trainer/Advantage Score Max 1.46546 +trainer/Advantage Score Min -4.09627 +trainer/V1 Predictions Mean -71.7142 +trainer/V1 Predictions Std 21.4598 +trainer/V1 Predictions Max 0.31725 +trainer/V1 Predictions Min -87.5585 +trainer/VF Loss 0.0789625 +expl/num steps total 804000 +expl/num paths total 1095 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0502798 +expl/Actions Std 0.837512 +expl/Actions Max 2.40977 +expl/Actions Min -2.26822 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 730661 +eval/num paths total 809 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0516196 +eval/Actions Std 0.789815 +eval/Actions Max 0.999848 +eval/Actions Min -0.999377 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.17698e-06 +time/evaluation sampling (s) 4.80392 +time/exploration sampling (s) 5.93627 +time/logging (s) 0.0107575 +time/saving (s) 0.0155505 +time/training (s) 17.9144 +time/epoch (s) 28.6809 +time/total (s) 20952.9 +Epoch -197 +------------------------------ ---------------- +2022-05-15 23:52:12.021820 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -196 finished +------------------------------ ---------------- +epoch -196 +replay_buffer/size 999047 +trainer/num train calls 805000 +trainer/QF1 Loss 0.944031 +trainer/QF2 Loss 1.11293 +trainer/Policy Loss 26.0622 +trainer/Q1 Predictions Mean -74.5632 +trainer/Q1 Predictions Std 17.1702 +trainer/Q1 Predictions Max -0.609932 +trainer/Q1 Predictions Min -87.6104 +trainer/Q2 Predictions Mean -74.546 +trainer/Q2 Predictions Std 17.0958 +trainer/Q2 Predictions Max -1.32905 +trainer/Q2 Predictions Min -87.5073 +trainer/Q Targets Mean -74.934 +trainer/Q Targets Std 17.3565 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5248 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0034692 +trainer/policy/mean Std 0.727575 +trainer/policy/mean Max 0.999261 +trainer/policy/mean Min -0.9998 +trainer/policy/std Mean 0.409063 +trainer/policy/std Std 0.0191426 +trainer/policy/std Max 0.42851 +trainer/policy/std Min 0.378568 +trainer/Advantage Weights Mean 7.17873 +trainer/Advantage Weights Std 21.5867 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.46643e-15 +trainer/Advantage Score Mean -0.304834 +trainer/Advantage Score Std 0.610981 +trainer/Advantage Score Max 1.89403 +trainer/Advantage Score Min -3.24027 +trainer/V1 Predictions Mean -74.682 +trainer/V1 Predictions Std 17.4275 +trainer/V1 Predictions Max 0.288938 +trainer/V1 Predictions Min -87.4272 +trainer/VF Loss 0.0781524 +expl/num steps total 805000 +expl/num paths total 1097 +expl/path length Mean 500 +expl/path length Std 201 +expl/path length Max 701 +expl/path length Min 299 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0189163 +expl/Actions Std 0.835479 +expl/Actions Max 2.30427 +expl/Actions Min -2.61128 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 731199 +eval/num paths total 810 +eval/path length Mean 538 +eval/path length Std 0 +eval/path length Max 538 +eval/path length Min 538 +eval/Rewards Mean 0.00185874 +eval/Rewards Std 0.043073 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0221277 +eval/Actions Std 0.734052 +eval/Actions Max 0.999522 +eval/Actions Min -0.999403 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.74509e-06 +time/evaluation sampling (s) 4.27327 +time/exploration sampling (s) 6.08246 +time/logging (s) 0.00595638 +time/saving (s) 0.0101167 +time/training (s) 18.1659 +time/epoch (s) 28.5377 +time/total (s) 20981.5 +Epoch -196 +------------------------------ ---------------- +2022-05-15 23:52:39.895203 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -195 finished +------------------------------ ---------------- +epoch -195 +replay_buffer/size 999047 +trainer/num train calls 806000 +trainer/QF1 Loss 2.86971 +trainer/QF2 Loss 2.72188 +trainer/Policy Loss 20.1255 +trainer/Q1 Predictions Mean -74.4042 +trainer/Q1 Predictions Std 17.5217 +trainer/Q1 Predictions Max -5.70654 +trainer/Q1 Predictions Min -88.5627 +trainer/Q2 Predictions Mean -74.4241 +trainer/Q2 Predictions Std 17.4856 +trainer/Q2 Predictions Max -5.146 +trainer/Q2 Predictions Min -88.3982 +trainer/Q Targets Mean -73.9418 +trainer/Q Targets Std 17.3073 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8327 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00313922 +trainer/policy/mean Std 0.731241 +trainer/policy/mean Max 0.99975 +trainer/policy/mean Min -0.999614 +trainer/policy/std Mean 0.410834 +trainer/policy/std Std 0.0182224 +trainer/policy/std Max 0.430357 +trainer/policy/std Min 0.384351 +trainer/Advantage Weights Mean 3.43172 +trainer/Advantage Weights Std 16.2602 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.69851e-15 +trainer/Advantage Score Mean -0.575056 +trainer/Advantage Score Std 0.617207 +trainer/Advantage Score Max 2.09936 +trainer/Advantage Score Min -3.24978 +trainer/V1 Predictions Mean -73.7919 +trainer/V1 Predictions Std 17.3244 +trainer/V1 Predictions Max -4.11418 +trainer/V1 Predictions Min -87.7439 +trainer/VF Loss 0.111361 +expl/num steps total 806000 +expl/num paths total 1098 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.24998 +expl/Actions Std 0.858218 +expl/Actions Max 2.29036 +expl/Actions Min -2.61822 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 731797 +eval/num paths total 811 +eval/path length Mean 598 +eval/path length Std 0 +eval/path length Max 598 +eval/path length Min 598 +eval/Rewards Mean 0.00167224 +eval/Rewards Std 0.0408588 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00402185 +eval/Actions Std 0.7551 +eval/Actions Max 0.999816 +eval/Actions Min -0.998679 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.52226e-06 +time/evaluation sampling (s) 4.38776 +time/exploration sampling (s) 6.64426 +time/logging (s) 0.00639767 +time/saving (s) 0.0105031 +time/training (s) 16.8146 +time/epoch (s) 27.8635 +time/total (s) 21009.3 +Epoch -195 +------------------------------ ---------------- +2022-05-15 23:53:06.049293 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -194 finished +------------------------------ ---------------- +epoch -194 +replay_buffer/size 999047 +trainer/num train calls 807000 +trainer/QF1 Loss 0.573257 +trainer/QF2 Loss 0.604888 +trainer/Policy Loss 34.4382 +trainer/Q1 Predictions Mean -73.4284 +trainer/Q1 Predictions Std 19.3731 +trainer/Q1 Predictions Max -0.473325 +trainer/Q1 Predictions Min -87.553 +trainer/Q2 Predictions Mean -73.3967 +trainer/Q2 Predictions Std 19.3397 +trainer/Q2 Predictions Max -0.814106 +trainer/Q2 Predictions Min -87.9384 +trainer/Q Targets Mean -73.7389 +trainer/Q Targets Std 19.4013 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.4876 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0166442 +trainer/policy/mean Std 0.73019 +trainer/policy/mean Max 0.999429 +trainer/policy/mean Min -0.998925 +trainer/policy/std Mean 0.410098 +trainer/policy/std Std 0.0182974 +trainer/policy/std Max 0.432226 +trainer/policy/std Min 0.381727 +trainer/Advantage Weights Mean 7.18673 +trainer/Advantage Weights Std 22.4235 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.00792e-26 +trainer/Advantage Score Mean -0.250449 +trainer/Advantage Score Std 0.634284 +trainer/Advantage Score Max 0.924816 +trainer/Advantage Score Min -5.98593 +trainer/V1 Predictions Mean -73.5097 +trainer/V1 Predictions Std 19.5654 +trainer/V1 Predictions Max -1.08237 +trainer/V1 Predictions Min -88.3955 +trainer/VF Loss 0.0704896 +expl/num steps total 807000 +expl/num paths total 1099 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0285337 +expl/Actions Std 0.841526 +expl/Actions Max 2.29033 +expl/Actions Min -2.29912 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 732797 +eval/num paths total 812 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0294412 +eval/Actions Std 0.75152 +eval/Actions Max 0.99986 +eval/Actions Min -0.999756 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.91388e-06 +time/evaluation sampling (s) 3.84953 +time/exploration sampling (s) 4.92898 +time/logging (s) 0.00828834 +time/saving (s) 0.0111808 +time/training (s) 17.3478 +time/epoch (s) 26.1458 +time/total (s) 21035.5 +Epoch -194 +------------------------------ ---------------- +2022-05-15 23:53:32.064005 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -193 finished +------------------------------ ---------------- +epoch -193 +replay_buffer/size 999047 +trainer/num train calls 808000 +trainer/QF1 Loss 1.34822 +trainer/QF2 Loss 1.22224 +trainer/Policy Loss 37.0191 +trainer/Q1 Predictions Mean -74.063 +trainer/Q1 Predictions Std 18.5364 +trainer/Q1 Predictions Max -0.631311 +trainer/Q1 Predictions Min -87.7728 +trainer/Q2 Predictions Mean -74.105 +trainer/Q2 Predictions Std 18.5555 +trainer/Q2 Predictions Max -0.155872 +trainer/Q2 Predictions Min -87.754 +trainer/Q Targets Mean -74.6384 +trainer/Q Targets Std 18.371 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.7156 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00822594 +trainer/policy/mean Std 0.7242 +trainer/policy/mean Max 0.999533 +trainer/policy/mean Min -0.998223 +trainer/policy/std Mean 0.408164 +trainer/policy/std Std 0.0180394 +trainer/policy/std Max 0.431381 +trainer/policy/std Min 0.378237 +trainer/Advantage Weights Mean 9.23721 +trainer/Advantage Weights Std 23.6013 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.20534e-22 +trainer/Advantage Score Mean -0.152306 +trainer/Advantage Score Std 0.627455 +trainer/Advantage Score Max 2.04148 +trainer/Advantage Score Min -4.84371 +trainer/V1 Predictions Mean -74.4483 +trainer/V1 Predictions Std 18.2803 +trainer/V1 Predictions Max -2.59489 +trainer/V1 Predictions Min -88.4026 +trainer/VF Loss 0.103036 +expl/num steps total 808000 +expl/num paths total 1101 +expl/path length Mean 500 +expl/path length Std 280 +expl/path length Max 780 +expl/path length Min 220 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.000805946 +expl/Actions Std 0.852565 +expl/Actions Max 2.14305 +expl/Actions Min -2.35244 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 733296 +eval/num paths total 813 +eval/path length Mean 499 +eval/path length Std 0 +eval/path length Max 499 +eval/path length Min 499 +eval/Rewards Mean 0.00200401 +eval/Rewards Std 0.0447213 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00635105 +eval/Actions Std 0.735033 +eval/Actions Max 0.999225 +eval/Actions Min -0.999438 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.42703e-06 +time/evaluation sampling (s) 3.72011 +time/exploration sampling (s) 5.65766 +time/logging (s) 0.00621882 +time/saving (s) 0.0112502 +time/training (s) 16.6004 +time/epoch (s) 25.9957 +time/total (s) 21061.5 +Epoch -193 +------------------------------ ---------------- +2022-05-15 23:53:57.733325 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -192 finished +------------------------------ ---------------- +epoch -192 +replay_buffer/size 999047 +trainer/num train calls 809000 +trainer/QF1 Loss 0.614959 +trainer/QF2 Loss 0.916096 +trainer/Policy Loss 20.9167 +trainer/Q1 Predictions Mean -74.7512 +trainer/Q1 Predictions Std 18.3375 +trainer/Q1 Predictions Max -0.574342 +trainer/Q1 Predictions Min -88.5544 +trainer/Q2 Predictions Mean -74.7232 +trainer/Q2 Predictions Std 18.3339 +trainer/Q2 Predictions Max -0.989554 +trainer/Q2 Predictions Min -88.4709 +trainer/Q Targets Mean -74.4858 +trainer/Q Targets Std 18.3656 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9755 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0255382 +trainer/policy/mean Std 0.731039 +trainer/policy/mean Max 0.99852 +trainer/policy/mean Min -0.999931 +trainer/policy/std Mean 0.408117 +trainer/policy/std Std 0.0198087 +trainer/policy/std Max 0.432433 +trainer/policy/std Min 0.376876 +trainer/Advantage Weights Mean 4.57601 +trainer/Advantage Weights Std 17.3284 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.13813e-16 +trainer/Advantage Score Mean -0.333766 +trainer/Advantage Score Std 0.529825 +trainer/Advantage Score Max 0.907616 +trainer/Advantage Score Min -3.46289 +trainer/V1 Predictions Mean -74.3238 +trainer/V1 Predictions Std 18.3717 +trainer/V1 Predictions Max -0.328661 +trainer/V1 Predictions Min -87.8535 +trainer/VF Loss 0.0539269 +expl/num steps total 809000 +expl/num paths total 1103 +expl/path length Mean 500 +expl/path length Std 364 +expl/path length Max 864 +expl/path length Min 136 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00454123 +expl/Actions Std 0.809634 +expl/Actions Max 2.19642 +expl/Actions Min -2.22079 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 734296 +eval/num paths total 814 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.115018 +eval/Actions Std 0.731394 +eval/Actions Max 0.999914 +eval/Actions Min -0.999729 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.74392e-06 +time/evaluation sampling (s) 3.8364 +time/exploration sampling (s) 4.67434 +time/logging (s) 0.00953353 +time/saving (s) 0.0109105 +time/training (s) 17.1311 +time/epoch (s) 25.6622 +time/total (s) 21087.2 +Epoch -192 +------------------------------ ---------------- +2022-05-15 23:54:22.978931 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -191 finished +------------------------------ ---------------- +epoch -191 +replay_buffer/size 999047 +trainer/num train calls 810000 +trainer/QF1 Loss 0.681284 +trainer/QF2 Loss 0.863573 +trainer/Policy Loss 19.24 +trainer/Q1 Predictions Mean -76.1841 +trainer/Q1 Predictions Std 16.2523 +trainer/Q1 Predictions Max -4.18062 +trainer/Q1 Predictions Min -88.3045 +trainer/Q2 Predictions Mean -76.1212 +trainer/Q2 Predictions Std 16.1449 +trainer/Q2 Predictions Max -5.10468 +trainer/Q2 Predictions Min -88.1519 +trainer/Q Targets Mean -76.1625 +trainer/Q Targets Std 16.3352 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2986 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00112837 +trainer/policy/mean Std 0.732636 +trainer/policy/mean Max 0.998587 +trainer/policy/mean Min -0.999027 +trainer/policy/std Mean 0.407014 +trainer/policy/std Std 0.0197024 +trainer/policy/std Max 0.432797 +trainer/policy/std Min 0.374629 +trainer/Advantage Weights Mean 4.46279 +trainer/Advantage Weights Std 17.6573 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.88081e-18 +trainer/Advantage Score Mean -0.272131 +trainer/Advantage Score Std 0.588069 +trainer/Advantage Score Max 1.36669 +trainer/Advantage Score Min -3.98612 +trainer/V1 Predictions Mean -75.8321 +trainer/V1 Predictions Std 16.5229 +trainer/V1 Predictions Max -1.80663 +trainer/V1 Predictions Min -87.9582 +trainer/VF Loss 0.0615684 +expl/num steps total 810000 +expl/num paths total 1105 +expl/path length Mean 500 +expl/path length Std 436 +expl/path length Max 936 +expl/path length Min 64 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0662719 +expl/Actions Std 0.822524 +expl/Actions Max 2.28292 +expl/Actions Min -2.34499 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 735296 +eval/num paths total 815 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0579314 +eval/Actions Std 0.743923 +eval/Actions Max 0.999957 +eval/Actions Min -0.999909 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 5.18374e-06 +time/evaluation sampling (s) 4.02183 +time/exploration sampling (s) 4.86022 +time/logging (s) 0.0135155 +time/saving (s) 0.0192276 +time/training (s) 16.3208 +time/epoch (s) 25.2356 +time/total (s) 21112.4 +Epoch -191 +------------------------------ ---------------- +2022-05-15 23:54:46.919796 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -190 finished +------------------------------ ---------------- +epoch -190 +replay_buffer/size 999047 +trainer/num train calls 811000 +trainer/QF1 Loss 0.698093 +trainer/QF2 Loss 0.619819 +trainer/Policy Loss 7.76232 +trainer/Q1 Predictions Mean -75.5745 +trainer/Q1 Predictions Std 16.9684 +trainer/Q1 Predictions Max -0.282769 +trainer/Q1 Predictions Min -87.9982 +trainer/Q2 Predictions Mean -75.5724 +trainer/Q2 Predictions Std 16.9833 +trainer/Q2 Predictions Max -0.300025 +trainer/Q2 Predictions Min -87.892 +trainer/Q Targets Mean -75.2918 +trainer/Q Targets Std 16.8486 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.637 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0110236 +trainer/policy/mean Std 0.73092 +trainer/policy/mean Max 0.999598 +trainer/policy/mean Min -0.999612 +trainer/policy/std Mean 0.40808 +trainer/policy/std Std 0.0203461 +trainer/policy/std Max 0.43387 +trainer/policy/std Min 0.375789 +trainer/Advantage Weights Mean 2.44226 +trainer/Advantage Weights Std 14.0013 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.735e-23 +trainer/Advantage Score Mean -0.538489 +trainer/Advantage Score Std 0.665737 +trainer/Advantage Score Max 1.45089 +trainer/Advantage Score Min -5.24084 +trainer/V1 Predictions Mean -74.9738 +trainer/V1 Predictions Std 17.1099 +trainer/V1 Predictions Max -0.320541 +trainer/V1 Predictions Min -87.503 +trainer/VF Loss 0.0893056 +expl/num steps total 811000 +expl/num paths total 1107 +expl/path length Mean 500 +expl/path length Std 24 +expl/path length Max 524 +expl/path length Min 476 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00839987 +expl/Actions Std 0.833885 +expl/Actions Max 2.51709 +expl/Actions Min -2.17399 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 735918 +eval/num paths total 816 +eval/path length Mean 622 +eval/path length Std 0 +eval/path length Max 622 +eval/path length Min 622 +eval/Rewards Mean 0.00160772 +eval/Rewards Std 0.0400641 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0129187 +eval/Actions Std 0.730524 +eval/Actions Max 0.999945 +eval/Actions Min -0.999945 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.43192e-06 +time/evaluation sampling (s) 2.8019 +time/exploration sampling (s) 4.88531 +time/logging (s) 0.0111485 +time/saving (s) 0.0138226 +time/training (s) 16.2135 +time/epoch (s) 23.9257 +time/total (s) 21136.3 +Epoch -190 +------------------------------ ---------------- +2022-05-15 23:55:10.414685 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -189 finished +------------------------------ ---------------- +epoch -189 +replay_buffer/size 999047 +trainer/num train calls 812000 +trainer/QF1 Loss 6.29315 +trainer/QF2 Loss 6.36121 +trainer/Policy Loss 16.235 +trainer/Q1 Predictions Mean -73.3933 +trainer/Q1 Predictions Std 18.5206 +trainer/Q1 Predictions Max -0.570607 +trainer/Q1 Predictions Min -87.8334 +trainer/Q2 Predictions Mean -73.3481 +trainer/Q2 Predictions Std 18.4501 +trainer/Q2 Predictions Max -0.221245 +trainer/Q2 Predictions Min -87.7215 +trainer/Q Targets Mean -73.5657 +trainer/Q Targets Std 18.2296 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7165 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0083939 +trainer/policy/mean Std 0.742887 +trainer/policy/mean Max 0.999598 +trainer/policy/mean Min -0.999481 +trainer/policy/std Mean 0.408355 +trainer/policy/std Std 0.0201316 +trainer/policy/std Max 0.431471 +trainer/policy/std Min 0.377981 +trainer/Advantage Weights Mean 4.94038 +trainer/Advantage Weights Std 19.5153 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.51498e-13 +trainer/Advantage Score Mean -0.359988 +trainer/Advantage Score Std 0.60779 +trainer/Advantage Score Max 2.30942 +trainer/Advantage Score Min -2.95182 +trainer/V1 Predictions Mean -73.2207 +trainer/V1 Predictions Std 18.3561 +trainer/V1 Predictions Max 0.342755 +trainer/V1 Predictions Min -87.5928 +trainer/VF Loss 0.0876642 +expl/num steps total 812000 +expl/num paths total 1109 +expl/path length Mean 500 +expl/path length Std 393 +expl/path length Max 893 +expl/path length Min 107 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0267615 +expl/Actions Std 0.821351 +expl/Actions Max 2.28754 +expl/Actions Min -2.27483 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 736918 +eval/num paths total 817 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0224429 +eval/Actions Std 0.714478 +eval/Actions Max 0.99997 +eval/Actions Min -0.999706 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.08478e-06 +time/evaluation sampling (s) 3.8943 +time/exploration sampling (s) 3.87589 +time/logging (s) 0.0100969 +time/saving (s) 0.0140806 +time/training (s) 15.6786 +time/epoch (s) 23.4729 +time/total (s) 21159.8 +Epoch -189 +------------------------------ ---------------- +2022-05-15 23:55:34.805190 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -188 finished +------------------------------ ---------------- +epoch -188 +replay_buffer/size 999047 +trainer/num train calls 813000 +trainer/QF1 Loss 1.243 +trainer/QF2 Loss 1.29107 +trainer/Policy Loss 18.1067 +trainer/Q1 Predictions Mean -73.5599 +trainer/Q1 Predictions Std 19.39 +trainer/Q1 Predictions Max -1.12423 +trainer/Q1 Predictions Min -87.1343 +trainer/Q2 Predictions Mean -73.621 +trainer/Q2 Predictions Std 19.3646 +trainer/Q2 Predictions Max -3.10367 +trainer/Q2 Predictions Min -87.0871 +trainer/Q Targets Mean -73.6174 +trainer/Q Targets Std 19.6025 +trainer/Q Targets Max -1.90311 +trainer/Q Targets Min -87.184 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0249904 +trainer/policy/mean Std 0.737059 +trainer/policy/mean Max 0.999463 +trainer/policy/mean Min -0.999883 +trainer/policy/std Mean 0.40978 +trainer/policy/std Std 0.0210078 +trainer/policy/std Max 0.433897 +trainer/policy/std Min 0.377746 +trainer/Advantage Weights Mean 4.96458 +trainer/Advantage Weights Std 17.6636 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.27228e-19 +trainer/Advantage Score Mean -0.306249 +trainer/Advantage Score Std 0.624708 +trainer/Advantage Score Max 1.13496 +trainer/Advantage Score Min -4.25636 +trainer/V1 Predictions Mean -73.3466 +trainer/V1 Predictions Std 19.8235 +trainer/V1 Predictions Max -1.08424 +trainer/V1 Predictions Min -87.0583 +trainer/VF Loss 0.0670898 +expl/num steps total 813000 +expl/num paths total 1110 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.125895 +expl/Actions Std 0.825185 +expl/Actions Max 2.37238 +expl/Actions Min -2.28985 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 737918 +eval/num paths total 818 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0771279 +eval/Actions Std 0.711449 +eval/Actions Max 0.999882 +eval/Actions Min -0.99972 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.16091e-06 +time/evaluation sampling (s) 3.46184 +time/exploration sampling (s) 4.93563 +time/logging (s) 0.00920519 +time/saving (s) 0.01417 +time/training (s) 15.9553 +time/epoch (s) 24.3762 +time/total (s) 21184.2 +Epoch -188 +------------------------------ ---------------- +2022-05-15 23:55:58.942859 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -187 finished +------------------------------ ---------------- +epoch -187 +replay_buffer/size 999047 +trainer/num train calls 814000 +trainer/QF1 Loss 0.828645 +trainer/QF2 Loss 0.709417 +trainer/Policy Loss 11.8792 +trainer/Q1 Predictions Mean -74.5769 +trainer/Q1 Predictions Std 18.1994 +trainer/Q1 Predictions Max -0.423804 +trainer/Q1 Predictions Min -87.7877 +trainer/Q2 Predictions Mean -74.5926 +trainer/Q2 Predictions Std 18.2262 +trainer/Q2 Predictions Max 0.0692497 +trainer/Q2 Predictions Min -87.7534 +trainer/Q Targets Mean -74.5029 +trainer/Q Targets Std 17.9928 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7195 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00778938 +trainer/policy/mean Std 0.739232 +trainer/policy/mean Max 0.999478 +trainer/policy/mean Min -0.99969 +trainer/policy/std Mean 0.410258 +trainer/policy/std Std 0.021007 +trainer/policy/std Max 0.431656 +trainer/policy/std Min 0.377649 +trainer/Advantage Weights Mean 2.26447 +trainer/Advantage Weights Std 11.7629 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.67705e-11 +trainer/Advantage Score Mean -0.467467 +trainer/Advantage Score Std 0.501696 +trainer/Advantage Score Max 0.99121 +trainer/Advantage Score Min -2.48114 +trainer/V1 Predictions Mean -74.2558 +trainer/V1 Predictions Std 18.155 +trainer/V1 Predictions Max -0.359971 +trainer/V1 Predictions Min -87.5611 +trainer/VF Loss 0.0561418 +expl/num steps total 814000 +expl/num paths total 1112 +expl/path length Mean 500 +expl/path length Std 398 +expl/path length Max 898 +expl/path length Min 102 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0184034 +expl/Actions Std 0.816681 +expl/Actions Max 2.26884 +expl/Actions Min -2.33444 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 738918 +eval/num paths total 819 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.142394 +eval/Actions Std 0.572364 +eval/Actions Max 0.999584 +eval/Actions Min -0.999972 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.52623e-06 +time/evaluation sampling (s) 3.1908 +time/exploration sampling (s) 4.82954 +time/logging (s) 0.00772334 +time/saving (s) 0.0137019 +time/training (s) 16.0799 +time/epoch (s) 24.1216 +time/total (s) 21208.3 +Epoch -187 +------------------------------ ---------------- +2022-05-15 23:56:22.770812 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -186 finished +------------------------------ ---------------- +epoch -186 +replay_buffer/size 999047 +trainer/num train calls 815000 +trainer/QF1 Loss 3.72754 +trainer/QF2 Loss 3.75765 +trainer/Policy Loss 22.1615 +trainer/Q1 Predictions Mean -73.2822 +trainer/Q1 Predictions Std 19.9287 +trainer/Q1 Predictions Max -0.208522 +trainer/Q1 Predictions Min -87.8649 +trainer/Q2 Predictions Mean -73.2256 +trainer/Q2 Predictions Std 19.9866 +trainer/Q2 Predictions Max 0.57027 +trainer/Q2 Predictions Min -87.9101 +trainer/Q Targets Mean -73.5153 +trainer/Q Targets Std 20.0033 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.3636 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00216248 +trainer/policy/mean Std 0.742235 +trainer/policy/mean Max 0.998592 +trainer/policy/mean Min -0.999033 +trainer/policy/std Mean 0.4094 +trainer/policy/std Std 0.0197612 +trainer/policy/std Max 0.428145 +trainer/policy/std Min 0.378063 +trainer/Advantage Weights Mean 6.03374 +trainer/Advantage Weights Std 19.5664 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.83696e-12 +trainer/Advantage Score Mean -0.271287 +trainer/Advantage Score Std 0.553997 +trainer/Advantage Score Max 1.22207 +trainer/Advantage Score Min -2.65883 +trainer/V1 Predictions Mean -73.3347 +trainer/V1 Predictions Std 20.1128 +trainer/V1 Predictions Max 0.318452 +trainer/V1 Predictions Min -88.4327 +trainer/VF Loss 0.0611992 +expl/num steps total 815000 +expl/num paths total 1113 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0389521 +expl/Actions Std 0.811466 +expl/Actions Max 2.13829 +expl/Actions Min -2.25361 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 739918 +eval/num paths total 820 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.523567 +eval/Actions Std 0.700615 +eval/Actions Max 0.999031 +eval/Actions Min -0.999517 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.93786e-06 +time/evaluation sampling (s) 3.74354 +time/exploration sampling (s) 3.74907 +time/logging (s) 0.00741913 +time/saving (s) 0.0133907 +time/training (s) 16.299 +time/epoch (s) 23.8125 +time/total (s) 21232.1 +Epoch -186 +------------------------------ ---------------- +2022-05-15 23:56:47.262840 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -185 finished +------------------------------ ---------------- +epoch -185 +replay_buffer/size 999047 +trainer/num train calls 816000 +trainer/QF1 Loss 0.760898 +trainer/QF2 Loss 0.651751 +trainer/Policy Loss 12.0708 +trainer/Q1 Predictions Mean -74.4442 +trainer/Q1 Predictions Std 16.9725 +trainer/Q1 Predictions Max -3.08167 +trainer/Q1 Predictions Min -88.0463 +trainer/Q2 Predictions Mean -74.355 +trainer/Q2 Predictions Std 16.9249 +trainer/Q2 Predictions Max -3.48142 +trainer/Q2 Predictions Min -87.8173 +trainer/Q Targets Mean -74.3801 +trainer/Q Targets Std 17.0966 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1458 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0203685 +trainer/policy/mean Std 0.734708 +trainer/policy/mean Max 0.999744 +trainer/policy/mean Min -0.999754 +trainer/policy/std Mean 0.409314 +trainer/policy/std Std 0.0195548 +trainer/policy/std Max 0.430372 +trainer/policy/std Min 0.376872 +trainer/Advantage Weights Mean 3.53702 +trainer/Advantage Weights Std 14.0867 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.10441e-15 +trainer/Advantage Score Mean -0.362079 +trainer/Advantage Score Std 0.576245 +trainer/Advantage Score Max 1.02949 +trainer/Advantage Score Min -3.3406 +trainer/V1 Predictions Mean -74.1745 +trainer/V1 Predictions Std 17.2066 +trainer/V1 Predictions Max -2.42521 +trainer/V1 Predictions Min -87.9987 +trainer/VF Loss 0.059954 +expl/num steps total 816000 +expl/num paths total 1114 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.117248 +expl/Actions Std 0.87611 +expl/Actions Max 2.31975 +expl/Actions Min -2.30335 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 740918 +eval/num paths total 821 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.35641 +eval/Actions Std 0.726613 +eval/Actions Max 0.999363 +eval/Actions Min -0.999636 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.15209e-05 +time/evaluation sampling (s) 3.73018 +time/exploration sampling (s) 5.00527 +time/logging (s) 0.00786529 +time/saving (s) 0.0108889 +time/training (s) 15.7262 +time/epoch (s) 24.4804 +time/total (s) 21256.6 +Epoch -185 +------------------------------ ---------------- +2022-05-15 23:57:10.143447 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -184 finished +------------------------------ ---------------- +epoch -184 +replay_buffer/size 999047 +trainer/num train calls 817000 +trainer/QF1 Loss 0.738623 +trainer/QF2 Loss 0.737377 +trainer/Policy Loss 6.30748 +trainer/Q1 Predictions Mean -73.8306 +trainer/Q1 Predictions Std 19.0362 +trainer/Q1 Predictions Max -0.707044 +trainer/Q1 Predictions Min -88.5212 +trainer/Q2 Predictions Mean -73.8089 +trainer/Q2 Predictions Std 19.0679 +trainer/Q2 Predictions Max 0.189289 +trainer/Q2 Predictions Min -87.4288 +trainer/Q Targets Mean -73.8131 +trainer/Q Targets Std 19.1762 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.147 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0214615 +trainer/policy/mean Std 0.735556 +trainer/policy/mean Max 0.9994 +trainer/policy/mean Min -0.999822 +trainer/policy/std Mean 0.409647 +trainer/policy/std Std 0.0191177 +trainer/policy/std Max 0.432845 +trainer/policy/std Min 0.376747 +trainer/Advantage Weights Mean 2.24699 +trainer/Advantage Weights Std 12.6354 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.92001e-23 +trainer/Advantage Score Mean -0.461755 +trainer/Advantage Score Std 0.649145 +trainer/Advantage Score Max 0.730446 +trainer/Advantage Score Min -5.08901 +trainer/V1 Predictions Mean -73.5032 +trainer/V1 Predictions Std 19.4278 +trainer/V1 Predictions Max -0.00311077 +trainer/V1 Predictions Min -86.9478 +trainer/VF Loss 0.0713352 +expl/num steps total 817000 +expl/num paths total 1115 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0392352 +expl/Actions Std 0.820079 +expl/Actions Max 2.3835 +expl/Actions Min -2.38965 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 741634 +eval/num paths total 822 +eval/path length Mean 716 +eval/path length Std 0 +eval/path length Max 716 +eval/path length Min 716 +eval/Rewards Mean 0.00139665 +eval/Rewards Std 0.0373456 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.012875 +eval/Actions Std 0.752474 +eval/Actions Max 0.999831 +eval/Actions Min -0.999738 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.9779e-06 +time/evaluation sampling (s) 2.97761 +time/exploration sampling (s) 4.00264 +time/logging (s) 0.0100922 +time/saving (s) 0.0154725 +time/training (s) 15.8636 +time/epoch (s) 22.8694 +time/total (s) 21279.5 +Epoch -184 +------------------------------ ---------------- +2022-05-15 23:57:33.773369 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -183 finished +------------------------------ ---------------- +epoch -183 +replay_buffer/size 999047 +trainer/num train calls 818000 +trainer/QF1 Loss 0.741869 +trainer/QF2 Loss 0.573769 +trainer/Policy Loss 12.1175 +trainer/Q1 Predictions Mean -73.7785 +trainer/Q1 Predictions Std 19.2723 +trainer/Q1 Predictions Max -0.720439 +trainer/Q1 Predictions Min -88.3974 +trainer/Q2 Predictions Mean -73.7496 +trainer/Q2 Predictions Std 19.1452 +trainer/Q2 Predictions Max -1.52202 +trainer/Q2 Predictions Min -87.9255 +trainer/Q Targets Mean -73.8997 +trainer/Q Targets Std 19.1959 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0312 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0229956 +trainer/policy/mean Std 0.735042 +trainer/policy/mean Max 0.999698 +trainer/policy/mean Min -0.999369 +trainer/policy/std Mean 0.408211 +trainer/policy/std Std 0.0193195 +trainer/policy/std Max 0.430229 +trainer/policy/std Min 0.375199 +trainer/Advantage Weights Mean 3.65796 +trainer/Advantage Weights Std 16.9567 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.49381e-18 +trainer/Advantage Score Mean -0.417915 +trainer/Advantage Score Std 0.611903 +trainer/Advantage Score Max 1.60302 +trainer/Advantage Score Min -3.93072 +trainer/V1 Predictions Mean -73.7213 +trainer/V1 Predictions Std 19.2618 +trainer/V1 Predictions Max -0.540489 +trainer/V1 Predictions Min -87.9046 +trainer/VF Loss 0.0804919 +expl/num steps total 818000 +expl/num paths total 1116 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.118488 +expl/Actions Std 0.863828 +expl/Actions Max 2.37126 +expl/Actions Min -2.29168 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 742335 +eval/num paths total 823 +eval/path length Mean 701 +eval/path length Std 0 +eval/path length Max 701 +eval/path length Min 701 +eval/Rewards Mean 0.00142653 +eval/Rewards Std 0.0377425 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00627628 +eval/Actions Std 0.745401 +eval/Actions Max 0.999524 +eval/Actions Min -0.99895 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.24706e-06 +time/evaluation sampling (s) 3.41644 +time/exploration sampling (s) 4.32708 +time/logging (s) 0.00913554 +time/saving (s) 0.0161301 +time/training (s) 15.8424 +time/epoch (s) 23.6112 +time/total (s) 21303.1 +Epoch -183 +------------------------------ ---------------- +2022-05-15 23:57:58.155552 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -182 finished +------------------------------ ---------------- +epoch -182 +replay_buffer/size 999047 +trainer/num train calls 819000 +trainer/QF1 Loss 1.27046 +trainer/QF2 Loss 1.25042 +trainer/Policy Loss 3.37325 +trainer/Q1 Predictions Mean -75.3972 +trainer/Q1 Predictions Std 16.7427 +trainer/Q1 Predictions Max -1.16872 +trainer/Q1 Predictions Min -87.6601 +trainer/Q2 Predictions Mean -75.3902 +trainer/Q2 Predictions Std 16.7773 +trainer/Q2 Predictions Max -0.206253 +trainer/Q2 Predictions Min -87.5809 +trainer/Q Targets Mean -75.3052 +trainer/Q Targets Std 16.8216 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6966 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00825147 +trainer/policy/mean Std 0.730523 +trainer/policy/mean Max 0.999099 +trainer/policy/mean Min -0.998542 +trainer/policy/std Mean 0.408981 +trainer/policy/std Std 0.0196143 +trainer/policy/std Max 0.429012 +trainer/policy/std Min 0.377656 +trainer/Advantage Weights Mean 0.818151 +trainer/Advantage Weights Std 6.63725 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.62844e-16 +trainer/Advantage Score Mean -0.533094 +trainer/Advantage Score Std 0.57872 +trainer/Advantage Score Max 0.660372 +trainer/Advantage Score Min -3.5875 +trainer/V1 Predictions Mean -75.1347 +trainer/V1 Predictions Std 16.9711 +trainer/V1 Predictions Max -0.0476283 +trainer/V1 Predictions Min -87.5729 +trainer/VF Loss 0.0645412 +expl/num steps total 819000 +expl/num paths total 1117 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0186973 +expl/Actions Std 0.844297 +expl/Actions Max 2.25951 +expl/Actions Min -2.761 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 743335 +eval/num paths total 824 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.119086 +eval/Actions Std 0.741932 +eval/Actions Max 0.999977 +eval/Actions Min -0.999705 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.01423e-06 +time/evaluation sampling (s) 3.14783 +time/exploration sampling (s) 4.98744 +time/logging (s) 0.00683613 +time/saving (s) 0.0096552 +time/training (s) 16.2132 +time/epoch (s) 24.3649 +time/total (s) 21327.5 +Epoch -182 +------------------------------ ---------------- +2022-05-15 23:58:21.890178 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -181 finished +------------------------------ ---------------- +epoch -181 +replay_buffer/size 999047 +trainer/num train calls 820000 +trainer/QF1 Loss 1.12627 +trainer/QF2 Loss 0.937302 +trainer/Policy Loss 15.2536 +trainer/Q1 Predictions Mean -74.0036 +trainer/Q1 Predictions Std 16.38 +trainer/Q1 Predictions Max -3.32352 +trainer/Q1 Predictions Min -89.2487 +trainer/Q2 Predictions Mean -74.1461 +trainer/Q2 Predictions Std 16.427 +trainer/Q2 Predictions Max -2.35286 +trainer/Q2 Predictions Min -88.9762 +trainer/Q Targets Mean -73.9921 +trainer/Q Targets Std 16.8536 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.4887 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0102478 +trainer/policy/mean Std 0.739025 +trainer/policy/mean Max 0.999908 +trainer/policy/mean Min -0.999572 +trainer/policy/std Mean 0.409423 +trainer/policy/std Std 0.0183822 +trainer/policy/std Max 0.430607 +trainer/policy/std Min 0.381224 +trainer/Advantage Weights Mean 5.60114 +trainer/Advantage Weights Std 20.7529 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.06279e-23 +trainer/Advantage Score Mean -0.391625 +trainer/Advantage Score Std 0.695565 +trainer/Advantage Score Max 1.20183 +trainer/Advantage Score Min -5.28986 +trainer/V1 Predictions Mean -73.7654 +trainer/V1 Predictions Std 16.8885 +trainer/V1 Predictions Max -0.47333 +trainer/V1 Predictions Min -88.9192 +trainer/VF Loss 0.0912833 +expl/num steps total 820000 +expl/num paths total 1119 +expl/path length Mean 500 +expl/path length Std 255 +expl/path length Max 755 +expl/path length Min 245 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0286537 +expl/Actions Std 0.833145 +expl/Actions Max 2.30423 +expl/Actions Min -2.36242 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 743792 +eval/num paths total 825 +eval/path length Mean 457 +eval/path length Std 0 +eval/path length Max 457 +eval/path length Min 457 +eval/Rewards Mean 0.00218818 +eval/Rewards Std 0.0467268 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00678584 +eval/Actions Std 0.741143 +eval/Actions Max 0.999843 +eval/Actions Min -0.999772 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.6389e-06 +time/evaluation sampling (s) 3.4832 +time/exploration sampling (s) 3.87115 +time/logging (s) 0.00692422 +time/saving (s) 0.0120954 +time/training (s) 16.3494 +time/epoch (s) 23.7227 +time/total (s) 21351.2 +Epoch -181 +------------------------------ ---------------- +2022-05-15 23:58:47.069347 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -180 finished +------------------------------ ---------------- +epoch -180 +replay_buffer/size 999047 +trainer/num train calls 821000 +trainer/QF1 Loss 0.78168 +trainer/QF2 Loss 0.720586 +trainer/Policy Loss 62.633 +trainer/Q1 Predictions Mean -75.0062 +trainer/Q1 Predictions Std 15.6435 +trainer/Q1 Predictions Max -1.86122 +trainer/Q1 Predictions Min -87.4024 +trainer/Q2 Predictions Mean -75.0087 +trainer/Q2 Predictions Std 15.6433 +trainer/Q2 Predictions Max -1.70332 +trainer/Q2 Predictions Min -87.3137 +trainer/Q Targets Mean -75.494 +trainer/Q Targets Std 15.9144 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0512 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00641674 +trainer/policy/mean Std 0.732409 +trainer/policy/mean Max 0.99876 +trainer/policy/mean Min -0.998878 +trainer/policy/std Mean 0.4087 +trainer/policy/std Std 0.018934 +trainer/policy/std Max 0.429156 +trainer/policy/std Min 0.381368 +trainer/Advantage Weights Mean 12.9973 +trainer/Advantage Weights Std 27.5964 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.11904e-17 +trainer/Advantage Score Mean -0.173013 +trainer/Advantage Score Std 0.688809 +trainer/Advantage Score Max 1.01351 +trainer/Advantage Score Min -3.80064 +trainer/V1 Predictions Mean -75.2397 +trainer/V1 Predictions Std 16.0667 +trainer/V1 Predictions Max -0.658292 +trainer/V1 Predictions Min -87.9052 +trainer/VF Loss 0.0922455 +expl/num steps total 821000 +expl/num paths total 1121 +expl/path length Mean 500 +expl/path length Std 225 +expl/path length Max 725 +expl/path length Min 275 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00694327 +expl/Actions Std 0.818861 +expl/Actions Max 2.40868 +expl/Actions Min -2.28858 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 744792 +eval/num paths total 826 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.130536 +eval/Actions Std 0.685374 +eval/Actions Max 0.999578 +eval/Actions Min -0.999797 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.37702e-06 +time/evaluation sampling (s) 4.2703 +time/exploration sampling (s) 4.42692 +time/logging (s) 0.00747433 +time/saving (s) 0.0111596 +time/training (s) 16.45 +time/epoch (s) 25.1659 +time/total (s) 21376.4 +Epoch -180 +------------------------------ ---------------- +2022-05-15 23:59:11.098854 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -179 finished +------------------------------ ---------------- +epoch -179 +replay_buffer/size 999047 +trainer/num train calls 822000 +trainer/QF1 Loss 0.682383 +trainer/QF2 Loss 0.691189 +trainer/Policy Loss 23.4093 +trainer/Q1 Predictions Mean -71.6651 +trainer/Q1 Predictions Std 20.0967 +trainer/Q1 Predictions Max -0.539781 +trainer/Q1 Predictions Min -87.8766 +trainer/Q2 Predictions Mean -71.6698 +trainer/Q2 Predictions Std 20.0645 +trainer/Q2 Predictions Max -0.760744 +trainer/Q2 Predictions Min -87.9544 +trainer/Q Targets Mean -71.4664 +trainer/Q Targets Std 19.9425 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5907 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0186243 +trainer/policy/mean Std 0.712998 +trainer/policy/mean Max 0.998862 +trainer/policy/mean Min -0.999008 +trainer/policy/std Mean 0.409787 +trainer/policy/std Std 0.0187459 +trainer/policy/std Max 0.432018 +trainer/policy/std Min 0.380995 +trainer/Advantage Weights Mean 4.05469 +trainer/Advantage Weights Std 17.4681 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.11092e-13 +trainer/Advantage Score Mean -0.417798 +trainer/Advantage Score Std 0.601389 +trainer/Advantage Score Max 3.28644 +trainer/Advantage Score Min -2.98284 +trainer/V1 Predictions Mean -71.2364 +trainer/V1 Predictions Std 20.0988 +trainer/V1 Predictions Max 0.614525 +trainer/V1 Predictions Min -87.1394 +trainer/VF Loss 0.102601 +expl/num steps total 822000 +expl/num paths total 1122 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.136779 +expl/Actions Std 0.855254 +expl/Actions Max 2.3296 +expl/Actions Min -2.39251 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 745792 +eval/num paths total 827 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.333271 +eval/Actions Std 0.747353 +eval/Actions Max 0.999901 +eval/Actions Min -0.999212 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.95602e-06 +time/evaluation sampling (s) 3.39617 +time/exploration sampling (s) 4.6029 +time/logging (s) 0.00764581 +time/saving (s) 0.013573 +time/training (s) 15.9973 +time/epoch (s) 24.0176 +time/total (s) 21400.4 +Epoch -179 +------------------------------ ---------------- +2022-05-15 23:59:35.372903 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -178 finished +------------------------------ ---------------- +epoch -178 +replay_buffer/size 999047 +trainer/num train calls 823000 +trainer/QF1 Loss 0.809149 +trainer/QF2 Loss 0.867877 +trainer/Policy Loss 14.9367 +trainer/Q1 Predictions Mean -73.8062 +trainer/Q1 Predictions Std 19.5242 +trainer/Q1 Predictions Max -0.803162 +trainer/Q1 Predictions Min -87.4466 +trainer/Q2 Predictions Mean -73.7866 +trainer/Q2 Predictions Std 19.5705 +trainer/Q2 Predictions Max -1.24025 +trainer/Q2 Predictions Min -87.5091 +trainer/Q Targets Mean -73.813 +trainer/Q Targets Std 19.5354 +trainer/Q Targets Max 0.632678 +trainer/Q Targets Min -87.8371 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00860137 +trainer/policy/mean Std 0.729656 +trainer/policy/mean Max 0.99933 +trainer/policy/mean Min -0.999424 +trainer/policy/std Mean 0.41128 +trainer/policy/std Std 0.0189447 +trainer/policy/std Max 0.431001 +trainer/policy/std Min 0.380708 +trainer/Advantage Weights Mean 4.7827 +trainer/Advantage Weights Std 18.717 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.28342e-14 +trainer/Advantage Score Mean -0.327782 +trainer/Advantage Score Std 0.552578 +trainer/Advantage Score Max 1.58791 +trainer/Advantage Score Min -3.10473 +trainer/V1 Predictions Mean -73.5739 +trainer/V1 Predictions Std 19.6721 +trainer/V1 Predictions Max 0.641334 +trainer/V1 Predictions Min -87.9572 +trainer/VF Loss 0.0705087 +expl/num steps total 823000 +expl/num paths total 1123 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00879279 +expl/Actions Std 0.813959 +expl/Actions Max 2.12045 +expl/Actions Min -2.56011 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 746792 +eval/num paths total 828 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0608031 +eval/Actions Std 0.719987 +eval/Actions Max 0.999886 +eval/Actions Min -0.999787 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.61493e-06 +time/evaluation sampling (s) 3.94361 +time/exploration sampling (s) 4.2777 +time/logging (s) 0.00860166 +time/saving (s) 0.011208 +time/training (s) 16.0179 +time/epoch (s) 24.259 +time/total (s) 21424.7 +Epoch -178 +------------------------------ ---------------- +2022-05-15 23:59:59.097181 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -177 finished +------------------------------ ---------------- +epoch -177 +replay_buffer/size 999047 +trainer/num train calls 824000 +trainer/QF1 Loss 0.673761 +trainer/QF2 Loss 0.675622 +trainer/Policy Loss 24.6169 +trainer/Q1 Predictions Mean -73.5497 +trainer/Q1 Predictions Std 18.4791 +trainer/Q1 Predictions Max -0.205149 +trainer/Q1 Predictions Min -87.7803 +trainer/Q2 Predictions Mean -73.6251 +trainer/Q2 Predictions Std 18.4962 +trainer/Q2 Predictions Max 0.593653 +trainer/Q2 Predictions Min -87.9045 +trainer/Q Targets Mean -73.6045 +trainer/Q Targets Std 18.5247 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0443 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000666054 +trainer/policy/mean Std 0.734146 +trainer/policy/mean Max 0.999579 +trainer/policy/mean Min -0.999905 +trainer/policy/std Mean 0.410439 +trainer/policy/std Std 0.0190025 +trainer/policy/std Max 0.429032 +trainer/policy/std Min 0.382028 +trainer/Advantage Weights Mean 6.11648 +trainer/Advantage Weights Std 20.5464 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.053e-17 +trainer/Advantage Score Mean -0.290003 +trainer/Advantage Score Std 0.582537 +trainer/Advantage Score Max 1.09396 +trainer/Advantage Score Min -3.84246 +trainer/V1 Predictions Mean -73.3382 +trainer/V1 Predictions Std 18.6404 +trainer/V1 Predictions Max 2.10175 +trainer/V1 Predictions Min -87.9154 +trainer/VF Loss 0.0602541 +expl/num steps total 824000 +expl/num paths total 1125 +expl/path length Mean 500 +expl/path length Std 394 +expl/path length Max 894 +expl/path length Min 106 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0325764 +expl/Actions Std 0.821212 +expl/Actions Max 2.36441 +expl/Actions Min -2.34212 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 747792 +eval/num paths total 829 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0260661 +eval/Actions Std 0.71508 +eval/Actions Max 0.999841 +eval/Actions Min -0.999432 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.68083e-06 +time/evaluation sampling (s) 3.04747 +time/exploration sampling (s) 4.34942 +time/logging (s) 0.0102747 +time/saving (s) 0.0160726 +time/training (s) 16.2915 +time/epoch (s) 23.7147 +time/total (s) 21448.4 +Epoch -177 +------------------------------ ---------------- +2022-05-16 00:00:22.964401 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -176 finished +------------------------------ ---------------- +epoch -176 +replay_buffer/size 999047 +trainer/num train calls 825000 +trainer/QF1 Loss 1.0837 +trainer/QF2 Loss 0.904795 +trainer/Policy Loss 8.5415 +trainer/Q1 Predictions Mean -73.3647 +trainer/Q1 Predictions Std 18.0547 +trainer/Q1 Predictions Max -1.3975 +trainer/Q1 Predictions Min -88.4429 +trainer/Q2 Predictions Mean -73.3271 +trainer/Q2 Predictions Std 18.0529 +trainer/Q2 Predictions Max -0.881309 +trainer/Q2 Predictions Min -88.3179 +trainer/Q Targets Mean -73.1476 +trainer/Q Targets Std 18.3278 +trainer/Q Targets Max -0.815714 +trainer/Q Targets Min -87.6259 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00358896 +trainer/policy/mean Std 0.747162 +trainer/policy/mean Max 0.999596 +trainer/policy/mean Min -0.999845 +trainer/policy/std Mean 0.409428 +trainer/policy/std Std 0.020317 +trainer/policy/std Max 0.431568 +trainer/policy/std Min 0.375305 +trainer/Advantage Weights Mean 3.2375 +trainer/Advantage Weights Std 14.7032 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.74057e-29 +trainer/Advantage Score Mean -0.476356 +trainer/Advantage Score Std 0.76969 +trainer/Advantage Score Max 1.64893 +trainer/Advantage Score Min -6.52188 +trainer/V1 Predictions Mean -72.9368 +trainer/V1 Predictions Std 18.3984 +trainer/V1 Predictions Max 0.0760255 +trainer/V1 Predictions Min -87.3582 +trainer/VF Loss 0.0985738 +expl/num steps total 825000 +expl/num paths total 1126 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0595322 +expl/Actions Std 0.851427 +expl/Actions Max 2.37534 +expl/Actions Min -2.45652 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 748792 +eval/num paths total 830 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0897736 +eval/Actions Std 0.738118 +eval/Actions Max 0.999956 +eval/Actions Min -0.999683 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.13274e-06 +time/evaluation sampling (s) 3.6845 +time/exploration sampling (s) 3.8572 +time/logging (s) 0.00794407 +time/saving (s) 0.0112605 +time/training (s) 16.2892 +time/epoch (s) 23.8501 +time/total (s) 21472.2 +Epoch -176 +------------------------------ ---------------- +2022-05-16 00:00:47.399550 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -175 finished +------------------------------ ---------------- +epoch -175 +replay_buffer/size 999047 +trainer/num train calls 826000 +trainer/QF1 Loss 0.821065 +trainer/QF2 Loss 0.624466 +trainer/Policy Loss 28.2883 +trainer/Q1 Predictions Mean -73.5164 +trainer/Q1 Predictions Std 18.7243 +trainer/Q1 Predictions Max -0.203887 +trainer/Q1 Predictions Min -87.2781 +trainer/Q2 Predictions Mean -73.5799 +trainer/Q2 Predictions Std 18.7218 +trainer/Q2 Predictions Max -0.206718 +trainer/Q2 Predictions Min -87.1152 +trainer/Q Targets Mean -73.6796 +trainer/Q Targets Std 18.9392 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5899 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00166742 +trainer/policy/mean Std 0.735769 +trainer/policy/mean Max 0.999042 +trainer/policy/mean Min -0.999101 +trainer/policy/std Mean 0.410447 +trainer/policy/std Std 0.0208272 +trainer/policy/std Max 0.433179 +trainer/policy/std Min 0.378526 +trainer/Advantage Weights Mean 6.06664 +trainer/Advantage Weights Std 19.532 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.91802e-15 +trainer/Advantage Score Mean -0.287716 +trainer/Advantage Score Std 0.649518 +trainer/Advantage Score Max 1.59816 +trainer/Advantage Score Min -3.38875 +trainer/V1 Predictions Mean -73.4517 +trainer/V1 Predictions Std 18.978 +trainer/V1 Predictions Max 0.480911 +trainer/V1 Predictions Min -87.4649 +trainer/VF Loss 0.0788102 +expl/num steps total 826000 +expl/num paths total 1127 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0317822 +expl/Actions Std 0.821655 +expl/Actions Max 2.30437 +expl/Actions Min -2.33852 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 749792 +eval/num paths total 831 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.100437 +eval/Actions Std 0.777233 +eval/Actions Max 0.999818 +eval/Actions Min -0.999841 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.75066e-06 +time/evaluation sampling (s) 3.97136 +time/exploration sampling (s) 4.2699 +time/logging (s) 0.00716114 +time/saving (s) 0.010918 +time/training (s) 16.1612 +time/epoch (s) 24.4206 +time/total (s) 21496.7 +Epoch -175 +------------------------------ ---------------- +2022-05-16 00:01:11.351591 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -174 finished +------------------------------ ---------------- +epoch -174 +replay_buffer/size 999047 +trainer/num train calls 827000 +trainer/QF1 Loss 0.71547 +trainer/QF2 Loss 0.791747 +trainer/Policy Loss 19.6245 +trainer/Q1 Predictions Mean -74.4171 +trainer/Q1 Predictions Std 18.9594 +trainer/Q1 Predictions Max -2.0272 +trainer/Q1 Predictions Min -88.6057 +trainer/Q2 Predictions Mean -74.4308 +trainer/Q2 Predictions Std 18.9963 +trainer/Q2 Predictions Max -2.34516 +trainer/Q2 Predictions Min -88.0932 +trainer/Q Targets Mean -74.2392 +trainer/Q Targets Std 19.0391 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.3334 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00780856 +trainer/policy/mean Std 0.728837 +trainer/policy/mean Max 0.999329 +trainer/policy/mean Min -0.999258 +trainer/policy/std Mean 0.409654 +trainer/policy/std Std 0.0199921 +trainer/policy/std Max 0.432287 +trainer/policy/std Min 0.380381 +trainer/Advantage Weights Mean 2.7314 +trainer/Advantage Weights Std 15.2064 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.30207e-19 +trainer/Advantage Score Mean -0.422734 +trainer/Advantage Score Std 0.526204 +trainer/Advantage Score Max 0.994736 +trainer/Advantage Score Min -4.29153 +trainer/V1 Predictions Mean -74.0508 +trainer/V1 Predictions Std 19.1588 +trainer/V1 Predictions Max -1.83734 +trainer/V1 Predictions Min -88.072 +trainer/VF Loss 0.0566893 +expl/num steps total 827000 +expl/num paths total 1128 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0263903 +expl/Actions Std 0.848862 +expl/Actions Max 2.25963 +expl/Actions Min -2.45926 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 750792 +eval/num paths total 832 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.243854 +eval/Actions Std 0.707222 +eval/Actions Max 0.999625 +eval/Actions Min -0.999397 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.03383e-06 +time/evaluation sampling (s) 3.77011 +time/exploration sampling (s) 4.11672 +time/logging (s) 0.00722708 +time/saving (s) 0.011508 +time/training (s) 16.0328 +time/epoch (s) 23.9384 +time/total (s) 21520.6 +Epoch -174 +------------------------------ ---------------- +2022-05-16 00:01:35.542440 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -173 finished +------------------------------ ---------------- +epoch -173 +replay_buffer/size 999047 +trainer/num train calls 828000 +trainer/QF1 Loss 1.05879 +trainer/QF2 Loss 1.15526 +trainer/Policy Loss 6.31266 +trainer/Q1 Predictions Mean -72.514 +trainer/Q1 Predictions Std 20.6408 +trainer/Q1 Predictions Max -4.70905 +trainer/Q1 Predictions Min -88.0214 +trainer/Q2 Predictions Mean -72.5439 +trainer/Q2 Predictions Std 20.5406 +trainer/Q2 Predictions Max -6.24597 +trainer/Q2 Predictions Min -88.7567 +trainer/Q Targets Mean -72.0973 +trainer/Q Targets Std 20.5585 +trainer/Q Targets Max -6.00624 +trainer/Q Targets Min -88.0303 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00650742 +trainer/policy/mean Std 0.745851 +trainer/policy/mean Max 0.999965 +trainer/policy/mean Min -0.99963 +trainer/policy/std Mean 0.409305 +trainer/policy/std Std 0.0196284 +trainer/policy/std Max 0.432598 +trainer/policy/std Min 0.377519 +trainer/Advantage Weights Mean 2.52258 +trainer/Advantage Weights Std 13.8456 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.42837e-23 +trainer/Advantage Score Mean -0.620375 +trainer/Advantage Score Std 0.890949 +trainer/Advantage Score Max 1.57461 +trainer/Advantage Score Min -5.17274 +trainer/V1 Predictions Mean -71.7814 +trainer/V1 Predictions Std 21.0041 +trainer/V1 Predictions Max -5.23839 +trainer/V1 Predictions Min -88.0709 +trainer/VF Loss 0.133946 +expl/num steps total 828000 +expl/num paths total 1129 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0374353 +expl/Actions Std 0.812671 +expl/Actions Max 2.30419 +expl/Actions Min -2.55176 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 751792 +eval/num paths total 833 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.085408 +eval/Actions Std 0.672295 +eval/Actions Max 0.999491 +eval/Actions Min -0.999311 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.58116e-06 +time/evaluation sampling (s) 3.869 +time/exploration sampling (s) 4.6428 +time/logging (s) 0.0108167 +time/saving (s) 0.0172364 +time/training (s) 15.6425 +time/epoch (s) 24.1824 +time/total (s) 21544.8 +Epoch -173 +------------------------------ ---------------- +2022-05-16 00:01:59.942426 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -172 finished +------------------------------ ---------------- +epoch -172 +replay_buffer/size 999047 +trainer/num train calls 829000 +trainer/QF1 Loss 1.69158 +trainer/QF2 Loss 1.84655 +trainer/Policy Loss 45.0774 +trainer/Q1 Predictions Mean -73.0819 +trainer/Q1 Predictions Std 18.9634 +trainer/Q1 Predictions Max -0.279245 +trainer/Q1 Predictions Min -87.215 +trainer/Q2 Predictions Mean -73.0941 +trainer/Q2 Predictions Std 19.1061 +trainer/Q2 Predictions Max 0.0236427 +trainer/Q2 Predictions Min -87.6078 +trainer/Q Targets Mean -73.5587 +trainer/Q Targets Std 18.5927 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.826 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0132524 +trainer/policy/mean Std 0.729376 +trainer/policy/mean Max 0.999876 +trainer/policy/mean Min -0.999344 +trainer/policy/std Mean 0.409451 +trainer/policy/std Std 0.0196746 +trainer/policy/std Max 0.429992 +trainer/policy/std Min 0.37892 +trainer/Advantage Weights Mean 9.38992 +trainer/Advantage Weights Std 25.5383 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.87234e-14 +trainer/Advantage Score Mean -0.195729 +trainer/Advantage Score Std 0.571015 +trainer/Advantage Score Max 1.94173 +trainer/Advantage Score Min -3.1609 +trainer/V1 Predictions Mean -73.3311 +trainer/V1 Predictions Std 18.7105 +trainer/V1 Predictions Max -1.35501 +trainer/V1 Predictions Min -87.7082 +trainer/VF Loss 0.0904531 +expl/num steps total 829000 +expl/num paths total 1130 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00962671 +expl/Actions Std 0.819469 +expl/Actions Max 2.4714 +expl/Actions Min -2.3545 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 752234 +eval/num paths total 834 +eval/path length Mean 442 +eval/path length Std 0 +eval/path length Max 442 +eval/path length Min 442 +eval/Rewards Mean 0.00226244 +eval/Rewards Std 0.0475113 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0215854 +eval/Actions Std 0.737724 +eval/Actions Max 0.999892 +eval/Actions Min -0.999661 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.71993e-06 +time/evaluation sampling (s) 3.39139 +time/exploration sampling (s) 4.91607 +time/logging (s) 0.00749263 +time/saving (s) 0.01554 +time/training (s) 16.0443 +time/epoch (s) 24.3748 +time/total (s) 21569.2 +Epoch -172 +------------------------------ ---------------- +2022-05-16 00:02:23.187263 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -171 finished +------------------------------ ---------------- +epoch -171 +replay_buffer/size 999047 +trainer/num train calls 830000 +trainer/QF1 Loss 0.625808 +trainer/QF2 Loss 0.585311 +trainer/Policy Loss 13.4381 +trainer/Q1 Predictions Mean -73.9111 +trainer/Q1 Predictions Std 18.5135 +trainer/Q1 Predictions Max -0.959851 +trainer/Q1 Predictions Min -87.9157 +trainer/Q2 Predictions Mean -73.9215 +trainer/Q2 Predictions Std 18.5275 +trainer/Q2 Predictions Max -1.32832 +trainer/Q2 Predictions Min -87.6823 +trainer/Q Targets Mean -73.8443 +trainer/Q Targets Std 18.5915 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.583 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00690224 +trainer/policy/mean Std 0.732596 +trainer/policy/mean Max 0.999111 +trainer/policy/mean Min -0.999423 +trainer/policy/std Mean 0.410286 +trainer/policy/std Std 0.019896 +trainer/policy/std Max 0.43385 +trainer/policy/std Min 0.378375 +trainer/Advantage Weights Mean 5.29762 +trainer/Advantage Weights Std 19.5267 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.42721e-17 +trainer/Advantage Score Mean -0.340063 +trainer/Advantage Score Std 0.589131 +trainer/Advantage Score Max 1.40325 +trainer/Advantage Score Min -3.87882 +trainer/V1 Predictions Mean -73.6313 +trainer/V1 Predictions Std 18.6555 +trainer/V1 Predictions Max 0.262621 +trainer/V1 Predictions Min -87.431 +trainer/VF Loss 0.0734381 +expl/num steps total 830000 +expl/num paths total 1131 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0367872 +expl/Actions Std 0.844775 +expl/Actions Max 2.46326 +expl/Actions Min -2.73965 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 752744 +eval/num paths total 835 +eval/path length Mean 510 +eval/path length Std 0 +eval/path length Max 510 +eval/path length Min 510 +eval/Rewards Mean 0.00196078 +eval/Rewards Std 0.0442373 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00594118 +eval/Actions Std 0.734211 +eval/Actions Max 0.999697 +eval/Actions Min -0.999651 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.03099e-06 +time/evaluation sampling (s) 3.31117 +time/exploration sampling (s) 3.98847 +time/logging (s) 0.00842579 +time/saving (s) 0.0142375 +time/training (s) 15.9061 +time/epoch (s) 23.2285 +time/total (s) 21592.4 +Epoch -171 +------------------------------ ---------------- +2022-05-16 00:02:46.999844 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -170 finished +------------------------------ ---------------- +epoch -170 +replay_buffer/size 999047 +trainer/num train calls 831000 +trainer/QF1 Loss 0.622221 +trainer/QF2 Loss 0.589413 +trainer/Policy Loss 10.5443 +trainer/Q1 Predictions Mean -73.9672 +trainer/Q1 Predictions Std 18.1272 +trainer/Q1 Predictions Max -1.65324 +trainer/Q1 Predictions Min -87.2744 +trainer/Q2 Predictions Mean -73.976 +trainer/Q2 Predictions Std 18.1097 +trainer/Q2 Predictions Max -1.38889 +trainer/Q2 Predictions Min -87.0364 +trainer/Q Targets Mean -74.0134 +trainer/Q Targets Std 18.1876 +trainer/Q Targets Max -0.547463 +trainer/Q Targets Min -86.9883 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000983005 +trainer/policy/mean Std 0.727029 +trainer/policy/mean Max 0.999627 +trainer/policy/mean Min -0.999626 +trainer/policy/std Mean 0.407905 +trainer/policy/std Std 0.0198332 +trainer/policy/std Max 0.429345 +trainer/policy/std Min 0.377971 +trainer/Advantage Weights Mean 2.57365 +trainer/Advantage Weights Std 14.4026 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.21267e-17 +trainer/Advantage Score Mean -0.600372 +trainer/Advantage Score Std 0.572315 +trainer/Advantage Score Max 0.972763 +trainer/Advantage Score Min -3.70383 +trainer/V1 Predictions Mean -73.7743 +trainer/V1 Predictions Std 18.2992 +trainer/V1 Predictions Max -2.17745 +trainer/V1 Predictions Min -86.9525 +trainer/VF Loss 0.0782319 +expl/num steps total 831000 +expl/num paths total 1132 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0303562 +expl/Actions Std 0.829044 +expl/Actions Max 2.27735 +expl/Actions Min -2.30733 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 753483 +eval/num paths total 836 +eval/path length Mean 739 +eval/path length Std 0 +eval/path length Max 739 +eval/path length Min 739 +eval/Rewards Mean 0.00135318 +eval/Rewards Std 0.0367607 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0288846 +eval/Actions Std 0.734627 +eval/Actions Max 0.99966 +eval/Actions Min -0.99974 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.8559e-06 +time/evaluation sampling (s) 3.70273 +time/exploration sampling (s) 3.91214 +time/logging (s) 0.00630042 +time/saving (s) 0.00953545 +time/training (s) 16.1631 +time/epoch (s) 23.7938 +time/total (s) 21616.2 +Epoch -170 +------------------------------ ---------------- +2022-05-16 00:03:10.477915 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -169 finished +------------------------------ ---------------- +epoch -169 +replay_buffer/size 999047 +trainer/num train calls 832000 +trainer/QF1 Loss 0.409555 +trainer/QF2 Loss 0.426994 +trainer/Policy Loss 17.0469 +trainer/Q1 Predictions Mean -75.2941 +trainer/Q1 Predictions Std 16.0994 +trainer/Q1 Predictions Max -0.734 +trainer/Q1 Predictions Min -87.5885 +trainer/Q2 Predictions Mean -75.2023 +trainer/Q2 Predictions Std 16.1228 +trainer/Q2 Predictions Max 0.136528 +trainer/Q2 Predictions Min -87.5609 +trainer/Q Targets Mean -75.1135 +trainer/Q Targets Std 16.1338 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.402 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0125812 +trainer/policy/mean Std 0.724389 +trainer/policy/mean Max 0.99952 +trainer/policy/mean Min -0.998333 +trainer/policy/std Mean 0.408398 +trainer/policy/std Std 0.0198508 +trainer/policy/std Max 0.429693 +trainer/policy/std Min 0.380139 +trainer/Advantage Weights Mean 3.60445 +trainer/Advantage Weights Std 16.7852 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.77408e-15 +trainer/Advantage Score Mean -0.430756 +trainer/Advantage Score Std 0.5019 +trainer/Advantage Score Max 1.12901 +trainer/Advantage Score Min -3.32106 +trainer/V1 Predictions Mean -74.8859 +trainer/V1 Predictions Std 16.2351 +trainer/V1 Predictions Max 0.0240219 +trainer/V1 Predictions Min -87.0991 +trainer/VF Loss 0.0591656 +expl/num steps total 832000 +expl/num paths total 1134 +expl/path length Mean 500 +expl/path length Std 237 +expl/path length Max 737 +expl/path length Min 263 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0175766 +expl/Actions Std 0.826394 +expl/Actions Max 2.2254 +expl/Actions Min -2.3893 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 754129 +eval/num paths total 837 +eval/path length Mean 646 +eval/path length Std 0 +eval/path length Max 646 +eval/path length Min 646 +eval/Rewards Mean 0.00154799 +eval/Rewards Std 0.039314 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0249375 +eval/Actions Std 0.745154 +eval/Actions Max 0.999706 +eval/Actions Min -0.998803 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 8.7861e-06 +time/evaluation sampling (s) 3.41279 +time/exploration sampling (s) 4.25081 +time/logging (s) 0.00634234 +time/saving (s) 0.0100834 +time/training (s) 15.7861 +time/epoch (s) 23.4661 +time/total (s) 21639.7 +Epoch -169 +------------------------------ ---------------- +2022-05-16 00:03:34.607587 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -168 finished +------------------------------ ---------------- +epoch -168 +replay_buffer/size 999047 +trainer/num train calls 833000 +trainer/QF1 Loss 0.5985 +trainer/QF2 Loss 0.449742 +trainer/Policy Loss 4.94702 +trainer/Q1 Predictions Mean -74.5393 +trainer/Q1 Predictions Std 17.7986 +trainer/Q1 Predictions Max -0.204016 +trainer/Q1 Predictions Min -88.0515 +trainer/Q2 Predictions Mean -74.5518 +trainer/Q2 Predictions Std 17.7588 +trainer/Q2 Predictions Max -0.306805 +trainer/Q2 Predictions Min -88.1938 +trainer/Q Targets Mean -74.4845 +trainer/Q Targets Std 17.6849 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8135 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0249771 +trainer/policy/mean Std 0.724375 +trainer/policy/mean Max 0.999493 +trainer/policy/mean Min -0.999717 +trainer/policy/std Mean 0.408769 +trainer/policy/std Std 0.0203228 +trainer/policy/std Max 0.432159 +trainer/policy/std Min 0.377553 +trainer/Advantage Weights Mean 1.68178 +trainer/Advantage Weights Std 9.71827 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.20747e-16 +trainer/Advantage Score Mean -0.403841 +trainer/Advantage Score Std 0.552822 +trainer/Advantage Score Max 0.652901 +trainer/Advantage Score Min -3.66528 +trainer/V1 Predictions Mean -74.1848 +trainer/V1 Predictions Std 17.9145 +trainer/V1 Predictions Max 0.150536 +trainer/V1 Predictions Min -87.644 +trainer/VF Loss 0.0513652 +expl/num steps total 833000 +expl/num paths total 1135 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0301767 +expl/Actions Std 0.831366 +expl/Actions Max 2.76932 +expl/Actions Min -2.64904 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 754970 +eval/num paths total 838 +eval/path length Mean 841 +eval/path length Std 0 +eval/path length Max 841 +eval/path length Min 841 +eval/Rewards Mean 0.00118906 +eval/Rewards Std 0.0344623 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0341497 +eval/Actions Std 0.762841 +eval/Actions Max 0.999882 +eval/Actions Min -0.999905 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.77488e-06 +time/evaluation sampling (s) 3.28544 +time/exploration sampling (s) 4.35715 +time/logging (s) 0.00768516 +time/saving (s) 0.0123247 +time/training (s) 16.4578 +time/epoch (s) 24.1204 +time/total (s) 21663.8 +Epoch -168 +------------------------------ ---------------- +2022-05-16 00:03:58.560586 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -167 finished +------------------------------ ---------------- +epoch -167 +replay_buffer/size 999047 +trainer/num train calls 834000 +trainer/QF1 Loss 0.435487 +trainer/QF2 Loss 0.499256 +trainer/Policy Loss 23.3022 +trainer/Q1 Predictions Mean -73.1214 +trainer/Q1 Predictions Std 20.537 +trainer/Q1 Predictions Max -0.506618 +trainer/Q1 Predictions Min -86.7575 +trainer/Q2 Predictions Mean -73.1722 +trainer/Q2 Predictions Std 20.4912 +trainer/Q2 Predictions Max -0.71073 +trainer/Q2 Predictions Min -87.1498 +trainer/Q Targets Mean -73.319 +trainer/Q Targets Std 20.7632 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1967 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00570862 +trainer/policy/mean Std 0.733716 +trainer/policy/mean Max 0.99858 +trainer/policy/mean Min -0.999325 +trainer/policy/std Mean 0.40858 +trainer/policy/std Std 0.020037 +trainer/policy/std Max 0.432191 +trainer/policy/std Min 0.379795 +trainer/Advantage Weights Mean 4.46663 +trainer/Advantage Weights Std 15.834 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.71726e-23 +trainer/Advantage Score Mean -0.322526 +trainer/Advantage Score Std 0.686209 +trainer/Advantage Score Max 1.04869 +trainer/Advantage Score Min -5.16465 +trainer/V1 Predictions Mean -73.053 +trainer/V1 Predictions Std 20.8057 +trainer/V1 Predictions Max -0.18315 +trainer/V1 Predictions Min -86.966 +trainer/VF Loss 0.0719097 +expl/num steps total 834000 +expl/num paths total 1136 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0735047 +expl/Actions Std 0.822352 +expl/Actions Max 2.31369 +expl/Actions Min -2.59178 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 755970 +eval/num paths total 839 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.401293 +eval/Actions Std 0.778734 +eval/Actions Max 0.999699 +eval/Actions Min -0.999259 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.40609e-06 +time/evaluation sampling (s) 3.53109 +time/exploration sampling (s) 3.6579 +time/logging (s) 0.00846128 +time/saving (s) 0.0142169 +time/training (s) 16.7275 +time/epoch (s) 23.9392 +time/total (s) 21687.8 +Epoch -167 +------------------------------ ---------------- +2022-05-16 00:04:22.213205 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -166 finished +------------------------------ ---------------- +epoch -166 +replay_buffer/size 999047 +trainer/num train calls 835000 +trainer/QF1 Loss 0.883879 +trainer/QF2 Loss 0.853634 +trainer/Policy Loss 21.9699 +trainer/Q1 Predictions Mean -72.2353 +trainer/Q1 Predictions Std 19.7322 +trainer/Q1 Predictions Max -0.203589 +trainer/Q1 Predictions Min -87.7625 +trainer/Q2 Predictions Mean -72.2426 +trainer/Q2 Predictions Std 19.7749 +trainer/Q2 Predictions Max 0.293727 +trainer/Q2 Predictions Min -87.7796 +trainer/Q Targets Mean -72.7084 +trainer/Q Targets Std 19.6478 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9706 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00193635 +trainer/policy/mean Std 0.735842 +trainer/policy/mean Max 0.999261 +trainer/policy/mean Min -0.999736 +trainer/policy/std Mean 0.407773 +trainer/policy/std Std 0.019163 +trainer/policy/std Max 0.430569 +trainer/policy/std Min 0.379119 +trainer/Advantage Weights Mean 6.12534 +trainer/Advantage Weights Std 20.054 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.25059e-19 +trainer/Advantage Score Mean -0.326821 +trainer/Advantage Score Std 0.640837 +trainer/Advantage Score Max 1.22503 +trainer/Advantage Score Min -4.35255 +trainer/V1 Predictions Mean -72.3829 +trainer/V1 Predictions Std 19.8648 +trainer/V1 Predictions Max 1.23145 +trainer/V1 Predictions Min -87.8473 +trainer/VF Loss 0.0748916 +expl/num steps total 835000 +expl/num paths total 1137 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0894345 +expl/Actions Std 0.838622 +expl/Actions Max 2.29704 +expl/Actions Min -2.18913 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 756970 +eval/num paths total 840 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0932006 +eval/Actions Std 0.714354 +eval/Actions Max 0.998959 +eval/Actions Min -0.998989 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.32296e-06 +time/evaluation sampling (s) 3.41852 +time/exploration sampling (s) 4.11432 +time/logging (s) 0.00810436 +time/saving (s) 0.0120379 +time/training (s) 16.085 +time/epoch (s) 23.638 +time/total (s) 21711.4 +Epoch -166 +------------------------------ ---------------- +2022-05-16 00:04:46.659723 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -165 finished +------------------------------ ---------------- +epoch -165 +replay_buffer/size 999047 +trainer/num train calls 836000 +trainer/QF1 Loss 0.734105 +trainer/QF2 Loss 0.726573 +trainer/Policy Loss 35.9752 +trainer/Q1 Predictions Mean -74.9822 +trainer/Q1 Predictions Std 18.2967 +trainer/Q1 Predictions Max -1.16414 +trainer/Q1 Predictions Min -87.6035 +trainer/Q2 Predictions Mean -75.0438 +trainer/Q2 Predictions Std 18.238 +trainer/Q2 Predictions Max -1.92977 +trainer/Q2 Predictions Min -87.6059 +trainer/Q Targets Mean -75.0793 +trainer/Q Targets Std 17.9144 +trainer/Q Targets Max -1.69947 +trainer/Q Targets Min -87.3923 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0129988 +trainer/policy/mean Std 0.727849 +trainer/policy/mean Max 0.999085 +trainer/policy/mean Min -0.998198 +trainer/policy/std Mean 0.406432 +trainer/policy/std Std 0.0195013 +trainer/policy/std Max 0.428372 +trainer/policy/std Min 0.375587 +trainer/Advantage Weights Mean 6.91424 +trainer/Advantage Weights Std 22.012 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.21023e-21 +trainer/Advantage Score Mean -0.28522 +trainer/Advantage Score Std 0.645415 +trainer/Advantage Score Max 1.81867 +trainer/Advantage Score Min -4.67037 +trainer/V1 Predictions Mean -74.8136 +trainer/V1 Predictions Std 18.2514 +trainer/V1 Predictions Max -1.38388 +trainer/V1 Predictions Min -87.2011 +trainer/VF Loss 0.094243 +expl/num steps total 836000 +expl/num paths total 1139 +expl/path length Mean 500 +expl/path length Std 441 +expl/path length Max 941 +expl/path length Min 59 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0143641 +expl/Actions Std 0.824378 +expl/Actions Max 2.43763 +expl/Actions Min -2.24451 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 757970 +eval/num paths total 841 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0788722 +eval/Actions Std 0.78701 +eval/Actions Max 0.999008 +eval/Actions Min -0.999538 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.62214e-06 +time/evaluation sampling (s) 3.55062 +time/exploration sampling (s) 4.64072 +time/logging (s) 0.00836806 +time/saving (s) 0.013064 +time/training (s) 16.2206 +time/epoch (s) 24.4333 +time/total (s) 21735.8 +Epoch -165 +------------------------------ ---------------- +2022-05-16 00:05:10.309469 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -164 finished +------------------------------ ---------------- +epoch -164 +replay_buffer/size 999047 +trainer/num train calls 837000 +trainer/QF1 Loss 0.960298 +trainer/QF2 Loss 0.829376 +trainer/Policy Loss 4.08574 +trainer/Q1 Predictions Mean -74.7113 +trainer/Q1 Predictions Std 18.1661 +trainer/Q1 Predictions Max -0.252549 +trainer/Q1 Predictions Min -88.1387 +trainer/Q2 Predictions Mean -74.6726 +trainer/Q2 Predictions Std 18.1676 +trainer/Q2 Predictions Max 0.0575755 +trainer/Q2 Predictions Min -88.1557 +trainer/Q Targets Mean -74.5075 +trainer/Q Targets Std 17.8485 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5696 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0466093 +trainer/policy/mean Std 0.74082 +trainer/policy/mean Max 0.999675 +trainer/policy/mean Min -0.999828 +trainer/policy/std Mean 0.407699 +trainer/policy/std Std 0.0206766 +trainer/policy/std Max 0.429696 +trainer/policy/std Min 0.375233 +trainer/Advantage Weights Mean 1.69411 +trainer/Advantage Weights Std 11.0041 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.14099e-12 +trainer/Advantage Score Mean -0.48834 +trainer/Advantage Score Std 0.466593 +trainer/Advantage Score Max 1.6104 +trainer/Advantage Score Min -2.68698 +trainer/V1 Predictions Mean -74.1561 +trainer/V1 Predictions Std 18.0602 +trainer/V1 Predictions Max -0.0333512 +trainer/V1 Predictions Min -87.3172 +trainer/VF Loss 0.0598567 +expl/num steps total 837000 +expl/num paths total 1140 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0581843 +expl/Actions Std 0.801914 +expl/Actions Max 2.16372 +expl/Actions Min -2.4832 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 758970 +eval/num paths total 842 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0537714 +eval/Actions Std 0.737584 +eval/Actions Max 0.99995 +eval/Actions Min -0.999788 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.37418e-06 +time/evaluation sampling (s) 3.09641 +time/exploration sampling (s) 4.24978 +time/logging (s) 0.00748766 +time/saving (s) 0.0108807 +time/training (s) 16.2695 +time/epoch (s) 23.634 +time/total (s) 21759.5 +Epoch -164 +------------------------------ ---------------- +2022-05-16 00:05:33.831518 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -163 finished +------------------------------ ---------------- +epoch -163 +replay_buffer/size 999047 +trainer/num train calls 838000 +trainer/QF1 Loss 0.447773 +trainer/QF2 Loss 0.405357 +trainer/Policy Loss 23.3324 +trainer/Q1 Predictions Mean -76.5228 +trainer/Q1 Predictions Std 15.4008 +trainer/Q1 Predictions Max -0.158498 +trainer/Q1 Predictions Min -88.4754 +trainer/Q2 Predictions Mean -76.4074 +trainer/Q2 Predictions Std 15.4433 +trainer/Q2 Predictions Max -0.444853 +trainer/Q2 Predictions Min -88.2506 +trainer/Q Targets Mean -76.2535 +trainer/Q Targets Std 15.5543 +trainer/Q Targets Max -2.52802 +trainer/Q Targets Min -88.1647 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00687228 +trainer/policy/mean Std 0.729129 +trainer/policy/mean Max 0.999482 +trainer/policy/mean Min -0.999365 +trainer/policy/std Mean 0.407785 +trainer/policy/std Std 0.0185396 +trainer/policy/std Max 0.427814 +trainer/policy/std Min 0.376291 +trainer/Advantage Weights Mean 4.93213 +trainer/Advantage Weights Std 18.2126 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.94581e-15 +trainer/Advantage Score Mean -0.294652 +trainer/Advantage Score Std 0.61691 +trainer/Advantage Score Max 2.56631 +trainer/Advantage Score Min -3.31661 +trainer/V1 Predictions Mean -76.0445 +trainer/V1 Predictions Std 15.5894 +trainer/V1 Predictions Max -2.62598 +trainer/V1 Predictions Min -88.045 +trainer/VF Loss 0.102506 +expl/num steps total 838000 +expl/num paths total 1141 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0787846 +expl/Actions Std 0.88572 +expl/Actions Max 2.42205 +expl/Actions Min -2.62929 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 759970 +eval/num paths total 843 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.171072 +eval/Actions Std 0.556201 +eval/Actions Max 0.999748 +eval/Actions Min -0.999774 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.39843e-05 +time/evaluation sampling (s) 3.70498 +time/exploration sampling (s) 3.56863 +time/logging (s) 0.00736813 +time/saving (s) 0.0102113 +time/training (s) 16.2176 +time/epoch (s) 23.5088 +time/total (s) 21783 +Epoch -163 +------------------------------ ---------------- +2022-05-16 00:05:58.251857 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -162 finished +------------------------------ ---------------- +epoch -162 +replay_buffer/size 999047 +trainer/num train calls 839000 +trainer/QF1 Loss 0.694707 +trainer/QF2 Loss 0.535812 +trainer/Policy Loss 20.6716 +trainer/Q1 Predictions Mean -72.1939 +trainer/Q1 Predictions Std 19.5124 +trainer/Q1 Predictions Max -0.681693 +trainer/Q1 Predictions Min -88.146 +trainer/Q2 Predictions Mean -72.1763 +trainer/Q2 Predictions Std 19.5233 +trainer/Q2 Predictions Max -0.779696 +trainer/Q2 Predictions Min -88.0817 +trainer/Q Targets Mean -72.0571 +trainer/Q Targets Std 19.6746 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7486 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0275572 +trainer/policy/mean Std 0.730097 +trainer/policy/mean Max 0.99783 +trainer/policy/mean Min -0.999117 +trainer/policy/std Mean 0.406644 +trainer/policy/std Std 0.019415 +trainer/policy/std Max 0.429067 +trainer/policy/std Min 0.37537 +trainer/Advantage Weights Mean 4.90326 +trainer/Advantage Weights Std 18.917 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.07268e-23 +trainer/Advantage Score Mean -0.351157 +trainer/Advantage Score Std 0.637695 +trainer/Advantage Score Max 1.70496 +trainer/Advantage Score Min -5.0871 +trainer/V1 Predictions Mean -71.7869 +trainer/V1 Predictions Std 19.8324 +trainer/V1 Predictions Max 0.980946 +trainer/V1 Predictions Min -87.5206 +trainer/VF Loss 0.0818141 +expl/num steps total 839000 +expl/num paths total 1142 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0999765 +expl/Actions Std 0.859106 +expl/Actions Max 2.5812 +expl/Actions Min -2.26079 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 760970 +eval/num paths total 844 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.010482 +eval/Actions Std 0.723689 +eval/Actions Max 0.999931 +eval/Actions Min -0.999635 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.59373e-06 +time/evaluation sampling (s) 3.4447 +time/exploration sampling (s) 4.95904 +time/logging (s) 0.0116493 +time/saving (s) 0.0134272 +time/training (s) 15.9837 +time/epoch (s) 24.4125 +time/total (s) 21807.4 +Epoch -162 +------------------------------ ---------------- +2022-05-16 00:06:22.443508 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -161 finished +------------------------------ ---------------- +epoch -161 +replay_buffer/size 999047 +trainer/num train calls 840000 +trainer/QF1 Loss 0.899961 +trainer/QF2 Loss 0.751664 +trainer/Policy Loss 34.4354 +trainer/Q1 Predictions Mean -74.2246 +trainer/Q1 Predictions Std 18.5363 +trainer/Q1 Predictions Max -0.516086 +trainer/Q1 Predictions Min -88.3935 +trainer/Q2 Predictions Mean -74.2735 +trainer/Q2 Predictions Std 18.5575 +trainer/Q2 Predictions Max -0.38316 +trainer/Q2 Predictions Min -88.3058 +trainer/Q Targets Mean -74.1907 +trainer/Q Targets Std 18.9597 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.047 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00207721 +trainer/policy/mean Std 0.735011 +trainer/policy/mean Max 0.998508 +trainer/policy/mean Min -0.999573 +trainer/policy/std Mean 0.408781 +trainer/policy/std Std 0.0216502 +trainer/policy/std Max 0.430904 +trainer/policy/std Min 0.37446 +trainer/Advantage Weights Mean 8.47452 +trainer/Advantage Weights Std 22.6507 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.29612e-18 +trainer/Advantage Score Mean -0.231413 +trainer/Advantage Score Std 0.652127 +trainer/Advantage Score Max 0.950199 +trainer/Advantage Score Min -4.06153 +trainer/V1 Predictions Mean -74.068 +trainer/V1 Predictions Std 18.9673 +trainer/V1 Predictions Max 0.644172 +trainer/V1 Predictions Min -87.9268 +trainer/VF Loss 0.0729682 +expl/num steps total 840000 +expl/num paths total 1143 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.04712 +expl/Actions Std 0.8365 +expl/Actions Max 2.27902 +expl/Actions Min -2.71985 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 761485 +eval/num paths total 845 +eval/path length Mean 515 +eval/path length Std 0 +eval/path length Max 515 +eval/path length Min 515 +eval/Rewards Mean 0.00194175 +eval/Rewards Std 0.0440225 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.023681 +eval/Actions Std 0.720959 +eval/Actions Max 0.999038 +eval/Actions Min -0.999147 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.62586e-06 +time/evaluation sampling (s) 3.01696 +time/exploration sampling (s) 4.78434 +time/logging (s) 0.00683857 +time/saving (s) 0.0129213 +time/training (s) 16.3459 +time/epoch (s) 24.167 +time/total (s) 21831.6 +Epoch -161 +------------------------------ ---------------- +2022-05-16 00:06:45.484163 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -160 finished +------------------------------ ---------------- +epoch -160 +replay_buffer/size 999047 +trainer/num train calls 841000 +trainer/QF1 Loss 1.35967 +trainer/QF2 Loss 1.30518 +trainer/Policy Loss 31.7955 +trainer/Q1 Predictions Mean -73.3329 +trainer/Q1 Predictions Std 18.0086 +trainer/Q1 Predictions Max -1.60625 +trainer/Q1 Predictions Min -86.8747 +trainer/Q2 Predictions Mean -73.3447 +trainer/Q2 Predictions Std 18.0368 +trainer/Q2 Predictions Max -1.3845 +trainer/Q2 Predictions Min -86.9544 +trainer/Q Targets Mean -73.8015 +trainer/Q Targets Std 18.5922 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.315 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00399435 +trainer/policy/mean Std 0.727861 +trainer/policy/mean Max 0.998984 +trainer/policy/mean Min -0.999961 +trainer/policy/std Mean 0.410268 +trainer/policy/std Std 0.0213947 +trainer/policy/std Max 0.432113 +trainer/policy/std Min 0.377847 +trainer/Advantage Weights Mean 9.35206 +trainer/Advantage Weights Std 21.0229 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.62684e-12 +trainer/Advantage Score Mean -0.0983675 +trainer/Advantage Score Std 0.518469 +trainer/Advantage Score Max 2.29424 +trainer/Advantage Score Min -2.59035 +trainer/V1 Predictions Mean -73.7404 +trainer/V1 Predictions Std 18.2474 +trainer/V1 Predictions Max -0.470687 +trainer/V1 Predictions Min -87.2187 +trainer/VF Loss 0.0700413 +expl/num steps total 841000 +expl/num paths total 1145 +expl/path length Mean 500 +expl/path length Std 143 +expl/path length Max 643 +expl/path length Min 357 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0277945 +expl/Actions Std 0.834568 +expl/Actions Max 2.42126 +expl/Actions Min -2.30752 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 762085 +eval/num paths total 846 +eval/path length Mean 600 +eval/path length Std 0 +eval/path length Max 600 +eval/path length Min 600 +eval/Rewards Mean 0.00166667 +eval/Rewards Std 0.0407908 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0258115 +eval/Actions Std 0.724915 +eval/Actions Max 0.999341 +eval/Actions Min -0.999737 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 6.11134e-06 +time/evaluation sampling (s) 3.27858 +time/exploration sampling (s) 3.98103 +time/logging (s) 0.00804235 +time/saving (s) 0.0137788 +time/training (s) 15.7464 +time/epoch (s) 23.0278 +time/total (s) 21854.6 +Epoch -160 +------------------------------ ---------------- +2022-05-16 00:07:09.636266 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -159 finished +------------------------------ ---------------- +epoch -159 +replay_buffer/size 999047 +trainer/num train calls 842000 +trainer/QF1 Loss 0.445782 +trainer/QF2 Loss 0.484117 +trainer/Policy Loss 17.0489 +trainer/Q1 Predictions Mean -76.9355 +trainer/Q1 Predictions Std 13.6643 +trainer/Q1 Predictions Max -3.18199 +trainer/Q1 Predictions Min -88.3107 +trainer/Q2 Predictions Mean -77.0246 +trainer/Q2 Predictions Std 13.674 +trainer/Q2 Predictions Max -2.36932 +trainer/Q2 Predictions Min -88.0355 +trainer/Q Targets Mean -77.1486 +trainer/Q Targets Std 13.5603 +trainer/Q Targets Max -5.74079 +trainer/Q Targets Min -88.8009 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00758265 +trainer/policy/mean Std 0.731188 +trainer/policy/mean Max 0.99869 +trainer/policy/mean Min -0.99835 +trainer/policy/std Mean 0.408974 +trainer/policy/std Std 0.0212502 +trainer/policy/std Max 0.433355 +trainer/policy/std Min 0.378153 +trainer/Advantage Weights Mean 3.97892 +trainer/Advantage Weights Std 17.718 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.34186e-17 +trainer/Advantage Score Mean -0.396571 +trainer/Advantage Score Std 0.617741 +trainer/Advantage Score Max 1.08749 +trainer/Advantage Score Min -3.88499 +trainer/V1 Predictions Mean -76.8303 +trainer/V1 Predictions Std 13.8644 +trainer/V1 Predictions Max -4.98532 +trainer/V1 Predictions Min -88.4293 +trainer/VF Loss 0.0682118 +expl/num steps total 842000 +expl/num paths total 1146 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00930663 +expl/Actions Std 0.859393 +expl/Actions Max 2.42525 +expl/Actions Min -2.20976 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 763085 +eval/num paths total 847 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0724527 +eval/Actions Std 0.730939 +eval/Actions Max 0.999829 +eval/Actions Min -0.999561 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.6729e-06 +time/evaluation sampling (s) 3.70519 +time/exploration sampling (s) 3.91351 +time/logging (s) 0.00832956 +time/saving (s) 0.0114096 +time/training (s) 16.5016 +time/epoch (s) 24.1401 +time/total (s) 21878.8 +Epoch -159 +------------------------------ ---------------- +2022-05-16 00:07:33.993625 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -158 finished +------------------------------ ---------------- +epoch -158 +replay_buffer/size 999047 +trainer/num train calls 843000 +trainer/QF1 Loss 0.915982 +trainer/QF2 Loss 1.16485 +trainer/Policy Loss 16.6567 +trainer/Q1 Predictions Mean -72.0598 +trainer/Q1 Predictions Std 21.0703 +trainer/Q1 Predictions Max -0.218678 +trainer/Q1 Predictions Min -88.0113 +trainer/Q2 Predictions Mean -72.029 +trainer/Q2 Predictions Std 21.0531 +trainer/Q2 Predictions Max 0.0149571 +trainer/Q2 Predictions Min -87.7228 +trainer/Q Targets Mean -71.8789 +trainer/Q Targets Std 21.126 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.11 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00368702 +trainer/policy/mean Std 0.733372 +trainer/policy/mean Max 0.99978 +trainer/policy/mean Min -0.999568 +trainer/policy/std Mean 0.407789 +trainer/policy/std Std 0.0199602 +trainer/policy/std Max 0.430316 +trainer/policy/std Min 0.375657 +trainer/Advantage Weights Mean 4.698 +trainer/Advantage Weights Std 16.8969 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.36637e-20 +trainer/Advantage Score Mean -0.449863 +trainer/Advantage Score Std 0.775011 +trainer/Advantage Score Max 0.710285 +trainer/Advantage Score Min -4.51903 +trainer/V1 Predictions Mean -71.6358 +trainer/V1 Predictions Std 21.3026 +trainer/V1 Predictions Max 1.31069 +trainer/V1 Predictions Min -87.8834 +trainer/VF Loss 0.0931914 +expl/num steps total 843000 +expl/num paths total 1148 +expl/path length Mean 500 +expl/path length Std 221 +expl/path length Max 721 +expl/path length Min 279 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0394639 +expl/Actions Std 0.816847 +expl/Actions Max 2.31244 +expl/Actions Min -2.38432 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 764085 +eval/num paths total 848 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.675913 +eval/Actions Std 0.420048 +eval/Actions Max 0.99951 +eval/Actions Min -0.998856 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.08019e-05 +time/evaluation sampling (s) 3.52046 +time/exploration sampling (s) 4.30128 +time/logging (s) 0.00794634 +time/saving (s) 0.0141565 +time/training (s) 16.4989 +time/epoch (s) 24.3428 +time/total (s) 21903.1 +Epoch -158 +------------------------------ ---------------- +2022-05-16 00:07:57.507881 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -157 finished +------------------------------ ---------------- +epoch -157 +replay_buffer/size 999047 +trainer/num train calls 844000 +trainer/QF1 Loss 0.639485 +trainer/QF2 Loss 0.678434 +trainer/Policy Loss 14.9944 +trainer/Q1 Predictions Mean -74.9751 +trainer/Q1 Predictions Std 17.2112 +trainer/Q1 Predictions Max 0.14272 +trainer/Q1 Predictions Min -87.9923 +trainer/Q2 Predictions Mean -74.9702 +trainer/Q2 Predictions Std 17.1438 +trainer/Q2 Predictions Max -0.62852 +trainer/Q2 Predictions Min -87.5417 +trainer/Q Targets Mean -74.8459 +trainer/Q Targets Std 17.2579 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5896 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00196461 +trainer/policy/mean Std 0.740928 +trainer/policy/mean Max 0.999773 +trainer/policy/mean Min -0.999492 +trainer/policy/std Mean 0.408166 +trainer/policy/std Std 0.0188104 +trainer/policy/std Max 0.42896 +trainer/policy/std Min 0.378288 +trainer/Advantage Weights Mean 1.71755 +trainer/Advantage Weights Std 11.6796 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.35821e-33 +trainer/Advantage Score Mean -0.592483 +trainer/Advantage Score Std 0.77872 +trainer/Advantage Score Max 0.867619 +trainer/Advantage Score Min -7.43067 +trainer/V1 Predictions Mean -74.6403 +trainer/V1 Predictions Std 17.2776 +trainer/V1 Predictions Max -0.328932 +trainer/V1 Predictions Min -87.5275 +trainer/VF Loss 0.103069 +expl/num steps total 844000 +expl/num paths total 1150 +expl/path length Mean 500 +expl/path length Std 271 +expl/path length Max 771 +expl/path length Min 229 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0456441 +expl/Actions Std 0.839397 +expl/Actions Max 2.36985 +expl/Actions Min -2.29908 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 764863 +eval/num paths total 849 +eval/path length Mean 778 +eval/path length Std 0 +eval/path length Max 778 +eval/path length Min 778 +eval/Rewards Mean 0.00128535 +eval/Rewards Std 0.0358287 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0129973 +eval/Actions Std 0.745683 +eval/Actions Max 0.999954 +eval/Actions Min -0.999538 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.87127e-06 +time/evaluation sampling (s) 3.39398 +time/exploration sampling (s) 4.2418 +time/logging (s) 0.00642864 +time/saving (s) 0.0100903 +time/training (s) 15.8477 +time/epoch (s) 23.5 +time/total (s) 21926.6 +Epoch -157 +------------------------------ ---------------- +2022-05-16 00:08:21.719675 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -156 finished +------------------------------ ---------------- +epoch -156 +replay_buffer/size 999047 +trainer/num train calls 845000 +trainer/QF1 Loss 0.515463 +trainer/QF2 Loss 0.500108 +trainer/Policy Loss 12.7749 +trainer/Q1 Predictions Mean -74.7597 +trainer/Q1 Predictions Std 16.9383 +trainer/Q1 Predictions Max -0.46285 +trainer/Q1 Predictions Min -87.8913 +trainer/Q2 Predictions Mean -74.713 +trainer/Q2 Predictions Std 16.9903 +trainer/Q2 Predictions Max 0.0676023 +trainer/Q2 Predictions Min -88.0836 +trainer/Q Targets Mean -74.6719 +trainer/Q Targets Std 17.0758 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6514 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.020226 +trainer/policy/mean Std 0.739791 +trainer/policy/mean Max 0.999448 +trainer/policy/mean Min -0.998998 +trainer/policy/std Mean 0.406269 +trainer/policy/std Std 0.0186932 +trainer/policy/std Max 0.428316 +trainer/policy/std Min 0.374643 +trainer/Advantage Weights Mean 2.77921 +trainer/Advantage Weights Std 13.9431 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.69035e-12 +trainer/Advantage Score Mean -0.381345 +trainer/Advantage Score Std 0.513407 +trainer/Advantage Score Max 1.08749 +trainer/Advantage Score Min -2.66413 +trainer/V1 Predictions Mean -74.4717 +trainer/V1 Predictions Std 17.1498 +trainer/V1 Predictions Max -0.225828 +trainer/V1 Predictions Min -87.5186 +trainer/VF Loss 0.0541932 +expl/num steps total 845000 +expl/num paths total 1151 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.20748 +expl/Actions Std 0.84523 +expl/Actions Max 2.37503 +expl/Actions Min -2.33335 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 765601 +eval/num paths total 850 +eval/path length Mean 738 +eval/path length Std 0 +eval/path length Max 738 +eval/path length Min 738 +eval/Rewards Mean 0.00135501 +eval/Rewards Std 0.0367856 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0295899 +eval/Actions Std 0.744902 +eval/Actions Max 0.999896 +eval/Actions Min -0.999676 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.28734e-06 +time/evaluation sampling (s) 3.82813 +time/exploration sampling (s) 3.97178 +time/logging (s) 0.00719686 +time/saving (s) 0.0148671 +time/training (s) 16.377 +time/epoch (s) 24.199 +time/total (s) 21950.8 +Epoch -156 +------------------------------ ---------------- +2022-05-16 00:08:46.015177 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -155 finished +------------------------------ ---------------- +epoch -155 +replay_buffer/size 999047 +trainer/num train calls 846000 +trainer/QF1 Loss 0.607078 +trainer/QF2 Loss 0.75184 +trainer/Policy Loss 11.1776 +trainer/Q1 Predictions Mean -74.6514 +trainer/Q1 Predictions Std 17.2319 +trainer/Q1 Predictions Max -1.1975 +trainer/Q1 Predictions Min -88.1756 +trainer/Q2 Predictions Mean -74.5591 +trainer/Q2 Predictions Std 17.3229 +trainer/Q2 Predictions Max -0.227828 +trainer/Q2 Predictions Min -88.0904 +trainer/Q Targets Mean -74.7587 +trainer/Q Targets Std 17.0874 +trainer/Q Targets Max -1.53754 +trainer/Q Targets Min -87.7856 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0409624 +trainer/policy/mean Std 0.738199 +trainer/policy/mean Max 0.999496 +trainer/policy/mean Min -0.999008 +trainer/policy/std Mean 0.406523 +trainer/policy/std Std 0.0191809 +trainer/policy/std Max 0.429651 +trainer/policy/std Min 0.374435 +trainer/Advantage Weights Mean 2.89013 +trainer/Advantage Weights Std 14.1128 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.96347e-15 +trainer/Advantage Score Mean -0.378123 +trainer/Advantage Score Std 0.57862 +trainer/Advantage Score Max 2.08062 +trainer/Advantage Score Min -3.34524 +trainer/V1 Predictions Mean -74.4536 +trainer/V1 Predictions Std 17.2805 +trainer/V1 Predictions Max -1.61254 +trainer/V1 Predictions Min -87.7783 +trainer/VF Loss 0.0718842 +expl/num steps total 846000 +expl/num paths total 1152 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.00965316 +expl/Actions Std 0.814547 +expl/Actions Max 2.30035 +expl/Actions Min -2.1735 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 766601 +eval/num paths total 851 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.164212 +eval/Actions Std 0.834163 +eval/Actions Max 0.999132 +eval/Actions Min -0.999243 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.02827e-05 +time/evaluation sampling (s) 3.48964 +time/exploration sampling (s) 4.57358 +time/logging (s) 0.0114755 +time/saving (s) 0.0164535 +time/training (s) 16.1927 +time/epoch (s) 24.2839 +time/total (s) 21975.1 +Epoch -155 +------------------------------ ---------------- +2022-05-16 00:09:10.249819 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -154 finished +------------------------------ ---------------- +epoch -154 +replay_buffer/size 999047 +trainer/num train calls 847000 +trainer/QF1 Loss 0.602623 +trainer/QF2 Loss 0.492684 +trainer/Policy Loss 8.17496 +trainer/Q1 Predictions Mean -74.5323 +trainer/Q1 Predictions Std 17.1183 +trainer/Q1 Predictions Max -0.899837 +trainer/Q1 Predictions Min -87.4578 +trainer/Q2 Predictions Mean -74.5323 +trainer/Q2 Predictions Std 17.1172 +trainer/Q2 Predictions Max -0.977481 +trainer/Q2 Predictions Min -87.3489 +trainer/Q Targets Mean -74.4129 +trainer/Q Targets Std 16.925 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4591 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00283971 +trainer/policy/mean Std 0.72752 +trainer/policy/mean Max 0.999656 +trainer/policy/mean Min -0.999181 +trainer/policy/std Mean 0.407914 +trainer/policy/std Std 0.0197215 +trainer/policy/std Max 0.428987 +trainer/policy/std Min 0.376292 +trainer/Advantage Weights Mean 3.6992 +trainer/Advantage Weights Std 17.1476 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.05682e-14 +trainer/Advantage Score Mean -0.420389 +trainer/Advantage Score Std 0.532004 +trainer/Advantage Score Max 1.89177 +trainer/Advantage Score Min -3.06155 +trainer/V1 Predictions Mean -74.0919 +trainer/V1 Predictions Std 17.1336 +trainer/V1 Predictions Max 0.119828 +trainer/V1 Predictions Min -87.3179 +trainer/VF Loss 0.0697935 +expl/num steps total 847000 +expl/num paths total 1154 +expl/path length Mean 500 +expl/path length Std 216 +expl/path length Max 716 +expl/path length Min 284 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.02204 +expl/Actions Std 0.822086 +expl/Actions Max 2.24254 +expl/Actions Min -2.25334 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 767601 +eval/num paths total 852 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.117439 +eval/Actions Std 0.745629 +eval/Actions Max 0.999924 +eval/Actions Min -0.998724 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.86102e-06 +time/evaluation sampling (s) 3.73122 +time/exploration sampling (s) 4.33567 +time/logging (s) 0.00779815 +time/saving (s) 0.0128655 +time/training (s) 16.125 +time/epoch (s) 24.2125 +time/total (s) 21999.3 +Epoch -154 +------------------------------ ---------------- +2022-05-16 00:09:34.280053 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -153 finished +------------------------------ ---------------- +epoch -153 +replay_buffer/size 999047 +trainer/num train calls 848000 +trainer/QF1 Loss 0.605327 +trainer/QF2 Loss 0.521713 +trainer/Policy Loss 10.3451 +trainer/Q1 Predictions Mean -75.3033 +trainer/Q1 Predictions Std 17.8322 +trainer/Q1 Predictions Max -1.32469 +trainer/Q1 Predictions Min -87.7108 +trainer/Q2 Predictions Mean -75.2941 +trainer/Q2 Predictions Std 17.8139 +trainer/Q2 Predictions Max -0.602158 +trainer/Q2 Predictions Min -87.6435 +trainer/Q Targets Mean -75.4019 +trainer/Q Targets Std 18.038 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7327 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.038705 +trainer/policy/mean Std 0.742139 +trainer/policy/mean Max 0.999118 +trainer/policy/mean Min -0.998036 +trainer/policy/std Mean 0.409019 +trainer/policy/std Std 0.0194936 +trainer/policy/std Max 0.430347 +trainer/policy/std Min 0.378306 +trainer/Advantage Weights Mean 3.2998 +trainer/Advantage Weights Std 15.4993 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.7102e-15 +trainer/Advantage Score Mean -0.390576 +trainer/Advantage Score Std 0.520271 +trainer/Advantage Score Max 1.08718 +trainer/Advantage Score Min -3.32277 +trainer/V1 Predictions Mean -75.1124 +trainer/V1 Predictions Std 18.1331 +trainer/V1 Predictions Max -1.09134 +trainer/V1 Predictions Min -87.6207 +trainer/VF Loss 0.0537564 +expl/num steps total 848000 +expl/num paths total 1156 +expl/path length Mean 500 +expl/path length Std 8 +expl/path length Max 508 +expl/path length Min 492 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.00133929 +expl/Actions Std 0.831095 +expl/Actions Max 2.66074 +expl/Actions Min -2.19718 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 768209 +eval/num paths total 853 +eval/path length Mean 608 +eval/path length Std 0 +eval/path length Max 608 +eval/path length Min 608 +eval/Rewards Mean 0.00164474 +eval/Rewards Std 0.040522 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0027329 +eval/Actions Std 0.730265 +eval/Actions Max 0.999645 +eval/Actions Min -0.999977 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.99794e-06 +time/evaluation sampling (s) 3.23317 +time/exploration sampling (s) 4.66716 +time/logging (s) 0.0109854 +time/saving (s) 0.018927 +time/training (s) 16.0888 +time/epoch (s) 24.0191 +time/total (s) 22023.3 +Epoch -153 +------------------------------ ---------------- +2022-05-16 00:09:57.929640 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -152 finished +------------------------------ ---------------- +epoch -152 +replay_buffer/size 999047 +trainer/num train calls 849000 +trainer/QF1 Loss 1.04687 +trainer/QF2 Loss 1.03703 +trainer/Policy Loss 8.65259 +trainer/Q1 Predictions Mean -73.2935 +trainer/Q1 Predictions Std 18.7117 +trainer/Q1 Predictions Max -1.01851 +trainer/Q1 Predictions Min -88.1958 +trainer/Q2 Predictions Mean -73.273 +trainer/Q2 Predictions Std 18.7153 +trainer/Q2 Predictions Max -0.452371 +trainer/Q2 Predictions Min -88.1343 +trainer/Q Targets Mean -72.9724 +trainer/Q Targets Std 18.6654 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4291 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0253067 +trainer/policy/mean Std 0.735896 +trainer/policy/mean Max 0.999101 +trainer/policy/mean Min -0.999459 +trainer/policy/std Mean 0.407661 +trainer/policy/std Std 0.0201528 +trainer/policy/std Max 0.429636 +trainer/policy/std Min 0.373526 +trainer/Advantage Weights Mean 2.90108 +trainer/Advantage Weights Std 14.7401 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.67695e-16 +trainer/Advantage Score Mean -0.497105 +trainer/Advantage Score Std 0.558032 +trainer/Advantage Score Max 1.03325 +trainer/Advantage Score Min -3.63244 +trainer/V1 Predictions Mean -72.7418 +trainer/V1 Predictions Std 18.7358 +trainer/V1 Predictions Max -0.586242 +trainer/V1 Predictions Min -87.3652 +trainer/VF Loss 0.067043 +expl/num steps total 849000 +expl/num paths total 1158 +expl/path length Mean 500 +expl/path length Std 287 +expl/path length Max 787 +expl/path length Min 213 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.017354 +expl/Actions Std 0.809443 +expl/Actions Max 2.21101 +expl/Actions Min -2.67359 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 769209 +eval/num paths total 854 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0887523 +eval/Actions Std 0.696469 +eval/Actions Max 0.999593 +eval/Actions Min -0.999163 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.84798e-06 +time/evaluation sampling (s) 3.35483 +time/exploration sampling (s) 3.75361 +time/logging (s) 0.00809802 +time/saving (s) 0.0115259 +time/training (s) 16.5017 +time/epoch (s) 23.6297 +time/total (s) 22047 +Epoch -152 +------------------------------ ---------------- +2022-05-16 00:10:22.000836 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -151 finished +------------------------------ ---------------- +epoch -151 +replay_buffer/size 999047 +trainer/num train calls 850000 +trainer/QF1 Loss 0.588842 +trainer/QF2 Loss 0.573893 +trainer/Policy Loss 10.7554 +trainer/Q1 Predictions Mean -73.9041 +trainer/Q1 Predictions Std 19.5732 +trainer/Q1 Predictions Max -0.937193 +trainer/Q1 Predictions Min -89.0123 +trainer/Q2 Predictions Mean -73.8709 +trainer/Q2 Predictions Std 19.5145 +trainer/Q2 Predictions Max -0.787407 +trainer/Q2 Predictions Min -88.5593 +trainer/Q Targets Mean -73.8351 +trainer/Q Targets Std 19.6236 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7695 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00317478 +trainer/policy/mean Std 0.736621 +trainer/policy/mean Max 0.999656 +trainer/policy/mean Min -0.999203 +trainer/policy/std Mean 0.407064 +trainer/policy/std Std 0.0193889 +trainer/policy/std Max 0.428949 +trainer/policy/std Min 0.37352 +trainer/Advantage Weights Mean 2.80281 +trainer/Advantage Weights Std 13.538 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.31916e-26 +trainer/Advantage Score Mean -0.391388 +trainer/Advantage Score Std 0.627675 +trainer/Advantage Score Max 2.07668 +trainer/Advantage Score Min -5.81959 +trainer/V1 Predictions Mean -73.5983 +trainer/V1 Predictions Std 19.686 +trainer/V1 Predictions Max 0.380978 +trainer/V1 Predictions Min -87.6741 +trainer/VF Loss 0.0760699 +expl/num steps total 850000 +expl/num paths total 1160 +expl/path length Mean 500 +expl/path length Std 408 +expl/path length Max 908 +expl/path length Min 92 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0336215 +expl/Actions Std 0.821892 +expl/Actions Max 2.14275 +expl/Actions Min -2.43847 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 769780 +eval/num paths total 855 +eval/path length Mean 571 +eval/path length Std 0 +eval/path length Max 571 +eval/path length Min 571 +eval/Rewards Mean 0.00175131 +eval/Rewards Std 0.041812 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00102649 +eval/Actions Std 0.724652 +eval/Actions Max 0.999768 +eval/Actions Min -0.999692 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.96114e-06 +time/evaluation sampling (s) 3.40087 +time/exploration sampling (s) 4.35631 +time/logging (s) 0.0059418 +time/saving (s) 0.0115426 +time/training (s) 16.2806 +time/epoch (s) 24.0553 +time/total (s) 22071 +Epoch -151 +------------------------------ ---------------- +2022-05-16 00:10:46.581816 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -150 finished +------------------------------ ---------------- +epoch -150 +replay_buffer/size 999047 +trainer/num train calls 851000 +trainer/QF1 Loss 0.551978 +trainer/QF2 Loss 0.554354 +trainer/Policy Loss 3.33197 +trainer/Q1 Predictions Mean -75.9664 +trainer/Q1 Predictions Std 15.3793 +trainer/Q1 Predictions Max -0.233627 +trainer/Q1 Predictions Min -87.8092 +trainer/Q2 Predictions Mean -75.9303 +trainer/Q2 Predictions Std 15.4023 +trainer/Q2 Predictions Max 0.328164 +trainer/Q2 Predictions Min -87.5174 +trainer/Q Targets Mean -75.7021 +trainer/Q Targets Std 15.4092 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3808 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0166398 +trainer/policy/mean Std 0.733685 +trainer/policy/mean Max 0.999196 +trainer/policy/mean Min -0.999793 +trainer/policy/std Mean 0.409344 +trainer/policy/std Std 0.0192435 +trainer/policy/std Max 0.42943 +trainer/policy/std Min 0.377321 +trainer/Advantage Weights Mean 1.20205 +trainer/Advantage Weights Std 7.61758 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.12492e-19 +trainer/Advantage Score Mean -0.500354 +trainer/Advantage Score Std 0.617287 +trainer/Advantage Score Max 0.607437 +trainer/Advantage Score Min -4.2115 +trainer/V1 Predictions Mean -75.3954 +trainer/V1 Predictions Std 15.6876 +trainer/V1 Predictions Max 0.899951 +trainer/V1 Predictions Min -87.2969 +trainer/VF Loss 0.0663088 +expl/num steps total 851000 +expl/num paths total 1161 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.150518 +expl/Actions Std 0.827525 +expl/Actions Max 2.60418 +expl/Actions Min -2.57307 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 770780 +eval/num paths total 856 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.192131 +eval/Actions Std 0.744928 +eval/Actions Max 0.999969 +eval/Actions Min -0.99988 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.31016e-06 +time/evaluation sampling (s) 3.78 +time/exploration sampling (s) 4.63972 +time/logging (s) 0.00723657 +time/saving (s) 0.0184993 +time/training (s) 16.1238 +time/epoch (s) 24.5692 +time/total (s) 22095.6 +Epoch -150 +------------------------------ ---------------- +2022-05-16 00:11:10.722484 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -149 finished +------------------------------ ---------------- +epoch -149 +replay_buffer/size 999047 +trainer/num train calls 852000 +trainer/QF1 Loss 0.795277 +trainer/QF2 Loss 0.824247 +trainer/Policy Loss 37.9081 +trainer/Q1 Predictions Mean -73.3495 +trainer/Q1 Predictions Std 18.5846 +trainer/Q1 Predictions Max -5.84306 +trainer/Q1 Predictions Min -87.5546 +trainer/Q2 Predictions Mean -73.3476 +trainer/Q2 Predictions Std 18.5584 +trainer/Q2 Predictions Max -5.65456 +trainer/Q2 Predictions Min -87.5267 +trainer/Q Targets Mean -73.7173 +trainer/Q Targets Std 18.4459 +trainer/Q Targets Max -5.71746 +trainer/Q Targets Min -87.9555 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00744895 +trainer/policy/mean Std 0.731407 +trainer/policy/mean Max 0.999598 +trainer/policy/mean Min -0.998952 +trainer/policy/std Mean 0.409611 +trainer/policy/std Std 0.0195939 +trainer/policy/std Max 0.432222 +trainer/policy/std Min 0.378038 +trainer/Advantage Weights Mean 10.0766 +trainer/Advantage Weights Std 24.3602 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.05554e-15 +trainer/Advantage Score Mean -0.172363 +trainer/Advantage Score Std 0.690981 +trainer/Advantage Score Max 2.5622 +trainer/Advantage Score Min -3.34218 +trainer/V1 Predictions Mean -73.479 +trainer/V1 Predictions Std 18.6774 +trainer/V1 Predictions Max -3.58847 +trainer/V1 Predictions Min -87.7833 +trainer/VF Loss 0.113675 +expl/num steps total 852000 +expl/num paths total 1163 +expl/path length Mean 500 +expl/path length Std 360 +expl/path length Max 860 +expl/path length Min 140 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.0669459 +expl/Actions Std 0.87091 +expl/Actions Max 2.50129 +expl/Actions Min -2.1539 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 771780 +eval/num paths total 857 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00406807 +eval/Actions Std 0.769679 +eval/Actions Max 0.999102 +eval/Actions Min -0.998451 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.2247e-06 +time/evaluation sampling (s) 3.03909 +time/exploration sampling (s) 4.61146 +time/logging (s) 0.00828112 +time/saving (s) 0.0115174 +time/training (s) 16.4591 +time/epoch (s) 24.1294 +time/total (s) 22119.7 +Epoch -149 +------------------------------ ---------------- +2022-05-16 00:11:34.336976 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -148 finished +------------------------------ ---------------- +epoch -148 +replay_buffer/size 999047 +trainer/num train calls 853000 +trainer/QF1 Loss 0.79298 +trainer/QF2 Loss 0.569402 +trainer/Policy Loss 19.9628 +trainer/Q1 Predictions Mean -72.1891 +trainer/Q1 Predictions Std 19.3754 +trainer/Q1 Predictions Max -0.976787 +trainer/Q1 Predictions Min -87.4801 +trainer/Q2 Predictions Mean -72.0913 +trainer/Q2 Predictions Std 19.467 +trainer/Q2 Predictions Max -0.653775 +trainer/Q2 Predictions Min -87.5027 +trainer/Q Targets Mean -72.1647 +trainer/Q Targets Std 19.4027 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4138 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0121241 +trainer/policy/mean Std 0.733 +trainer/policy/mean Max 0.999498 +trainer/policy/mean Min -0.999329 +trainer/policy/std Mean 0.408839 +trainer/policy/std Std 0.0190787 +trainer/policy/std Max 0.431286 +trainer/policy/std Min 0.379573 +trainer/Advantage Weights Mean 5.51573 +trainer/Advantage Weights Std 18.7699 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.81248e-18 +trainer/Advantage Score Mean -0.259867 +trainer/Advantage Score Std 0.606846 +trainer/Advantage Score Max 2.22169 +trainer/Advantage Score Min -3.92704 +trainer/V1 Predictions Mean -71.8972 +trainer/V1 Predictions Std 19.4691 +trainer/V1 Predictions Max -1.98713 +trainer/V1 Predictions Min -87.2869 +trainer/VF Loss 0.0875916 +expl/num steps total 853000 +expl/num paths total 1164 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0805912 +expl/Actions Std 0.813446 +expl/Actions Max 2.13166 +expl/Actions Min -2.32053 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 772780 +eval/num paths total 858 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.204916 +eval/Actions Std 0.67285 +eval/Actions Max 0.999966 +eval/Actions Min -0.999315 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.74833e-06 +time/evaluation sampling (s) 3.30314 +time/exploration sampling (s) 4.2388 +time/logging (s) 0.00894145 +time/saving (s) 0.0107377 +time/training (s) 16.0384 +time/epoch (s) 23.6 +time/total (s) 22143.3 +Epoch -148 +------------------------------ ---------------- +2022-05-16 00:11:58.087821 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -147 finished +------------------------------ ---------------- +epoch -147 +replay_buffer/size 999047 +trainer/num train calls 854000 +trainer/QF1 Loss 1.11254 +trainer/QF2 Loss 1.11673 +trainer/Policy Loss 51.0771 +trainer/Q1 Predictions Mean -75.0864 +trainer/Q1 Predictions Std 16.7722 +trainer/Q1 Predictions Max -0.442181 +trainer/Q1 Predictions Min -86.9614 +trainer/Q2 Predictions Mean -75.0148 +trainer/Q2 Predictions Std 16.6745 +trainer/Q2 Predictions Max -0.775231 +trainer/Q2 Predictions Min -86.9641 +trainer/Q Targets Mean -75.7636 +trainer/Q Targets Std 16.8979 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5698 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0380063 +trainer/policy/mean Std 0.728811 +trainer/policy/mean Max 0.999201 +trainer/policy/mean Min -0.999568 +trainer/policy/std Mean 0.408784 +trainer/policy/std Std 0.0198698 +trainer/policy/std Max 0.430638 +trainer/policy/std Min 0.379706 +trainer/Advantage Weights Mean 10.33 +trainer/Advantage Weights Std 22.6334 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.69413e-16 +trainer/Advantage Score Mean -0.133296 +trainer/Advantage Score Std 0.642494 +trainer/Advantage Score Max 1.06354 +trainer/Advantage Score Min -3.5295 +trainer/V1 Predictions Mean -75.4634 +trainer/V1 Predictions Std 17.1984 +trainer/V1 Predictions Max -0.407814 +trainer/V1 Predictions Min -87.4427 +trainer/VF Loss 0.0815513 +expl/num steps total 854000 +expl/num paths total 1165 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0351715 +expl/Actions Std 0.834889 +expl/Actions Max 2.30552 +expl/Actions Min -2.33548 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 773438 +eval/num paths total 859 +eval/path length Mean 658 +eval/path length Std 0 +eval/path length Max 658 +eval/path length Min 658 +eval/Rewards Mean 0.00151976 +eval/Rewards Std 0.0389544 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00713958 +eval/Actions Std 0.744106 +eval/Actions Max 0.999896 +eval/Actions Min -0.99966 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.65009e-06 +time/evaluation sampling (s) 3.50951 +time/exploration sampling (s) 4.03916 +time/logging (s) 0.00816226 +time/saving (s) 0.0148777 +time/training (s) 16.1664 +time/epoch (s) 23.7381 +time/total (s) 22167.1 +Epoch -147 +------------------------------ ---------------- +2022-05-16 00:12:22.181020 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -146 finished +------------------------------ ---------------- +epoch -146 +replay_buffer/size 999047 +trainer/num train calls 855000 +trainer/QF1 Loss 0.868076 +trainer/QF2 Loss 0.834103 +trainer/Policy Loss 66.3851 +trainer/Q1 Predictions Mean -75.1437 +trainer/Q1 Predictions Std 16.633 +trainer/Q1 Predictions Max -2.05107 +trainer/Q1 Predictions Min -86.691 +trainer/Q2 Predictions Mean -74.9967 +trainer/Q2 Predictions Std 16.6515 +trainer/Q2 Predictions Max -2.72653 +trainer/Q2 Predictions Min -86.6599 +trainer/Q Targets Mean -75.4787 +trainer/Q Targets Std 16.6848 +trainer/Q Targets Max -2.27962 +trainer/Q Targets Min -87.4398 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00786243 +trainer/policy/mean Std 0.724796 +trainer/policy/mean Max 0.999088 +trainer/policy/mean Min -0.998798 +trainer/policy/std Mean 0.408752 +trainer/policy/std Std 0.0207354 +trainer/policy/std Max 0.430143 +trainer/policy/std Min 0.374144 +trainer/Advantage Weights Mean 12.7768 +trainer/Advantage Weights Std 25.967 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.25969e-13 +trainer/Advantage Score Mean -0.104798 +trainer/Advantage Score Std 0.574954 +trainer/Advantage Score Max 1.28117 +trainer/Advantage Score Min -2.97027 +trainer/V1 Predictions Mean -75.2364 +trainer/V1 Predictions Std 16.9102 +trainer/V1 Predictions Max -1.16936 +trainer/V1 Predictions Min -87.2216 +trainer/VF Loss 0.0828114 +expl/num steps total 855000 +expl/num paths total 1167 +expl/path length Mean 500 +expl/path length Std 310 +expl/path length Max 810 +expl/path length Min 190 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0144587 +expl/Actions Std 0.83931 +expl/Actions Max 2.39051 +expl/Actions Min -2.25509 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 774438 +eval/num paths total 860 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0434489 +eval/Actions Std 0.64542 +eval/Actions Max 0.99915 +eval/Actions Min -0.999307 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.8941e-06 +time/evaluation sampling (s) 3.70473 +time/exploration sampling (s) 3.99539 +time/logging (s) 0.00865137 +time/saving (s) 0.0133623 +time/training (s) 16.355 +time/epoch (s) 24.0772 +time/total (s) 22191.2 +Epoch -146 +------------------------------ ---------------- +2022-05-16 00:12:46.680354 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -145 finished +------------------------------ ---------------- +epoch -145 +replay_buffer/size 999047 +trainer/num train calls 856000 +trainer/QF1 Loss 1.48638 +trainer/QF2 Loss 1.50177 +trainer/Policy Loss 11.7387 +trainer/Q1 Predictions Mean -72.7215 +trainer/Q1 Predictions Std 19.722 +trainer/Q1 Predictions Max -0.983195 +trainer/Q1 Predictions Min -87.1655 +trainer/Q2 Predictions Mean -72.6644 +trainer/Q2 Predictions Std 19.7056 +trainer/Q2 Predictions Max -0.107629 +trainer/Q2 Predictions Min -87.1355 +trainer/Q Targets Mean -72.5476 +trainer/Q Targets Std 19.691 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0627 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0213179 +trainer/policy/mean Std 0.714151 +trainer/policy/mean Max 0.999939 +trainer/policy/mean Min -0.998904 +trainer/policy/std Mean 0.410475 +trainer/policy/std Std 0.0199858 +trainer/policy/std Max 0.433164 +trainer/policy/std Min 0.377017 +trainer/Advantage Weights Mean 2.81117 +trainer/Advantage Weights Std 15.2114 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.47731e-18 +trainer/Advantage Score Mean -0.494835 +trainer/Advantage Score Std 0.603361 +trainer/Advantage Score Max 0.877724 +trainer/Advantage Score Min -4.05394 +trainer/V1 Predictions Mean -72.1757 +trainer/V1 Predictions Std 19.8369 +trainer/V1 Predictions Max -0.670832 +trainer/V1 Predictions Min -86.9231 +trainer/VF Loss 0.0713459 +expl/num steps total 856000 +expl/num paths total 1168 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0529092 +expl/Actions Std 0.847256 +expl/Actions Max 2.49674 +expl/Actions Min -2.23557 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 775438 +eval/num paths total 861 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0159045 +eval/Actions Std 0.787548 +eval/Actions Max 0.999632 +eval/Actions Min -0.999846 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.06685e-06 +time/evaluation sampling (s) 3.67932 +time/exploration sampling (s) 4.48395 +time/logging (s) 0.0171005 +time/saving (s) 0.019786 +time/training (s) 16.2933 +time/epoch (s) 24.4935 +time/total (s) 22215.7 +Epoch -145 +------------------------------ ---------------- +2022-05-16 00:13:10.564488 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -144 finished +------------------------------ ---------------- +epoch -144 +replay_buffer/size 999047 +trainer/num train calls 857000 +trainer/QF1 Loss 0.722265 +trainer/QF2 Loss 0.612132 +trainer/Policy Loss 21.7047 +trainer/Q1 Predictions Mean -76.2815 +trainer/Q1 Predictions Std 16.4665 +trainer/Q1 Predictions Max -1.3526 +trainer/Q1 Predictions Min -88.4076 +trainer/Q2 Predictions Mean -76.284 +trainer/Q2 Predictions Std 16.475 +trainer/Q2 Predictions Max -2.21073 +trainer/Q2 Predictions Min -88.2991 +trainer/Q Targets Mean -76.1147 +trainer/Q Targets Std 16.3849 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8227 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.017442 +trainer/policy/mean Std 0.734829 +trainer/policy/mean Max 0.998989 +trainer/policy/mean Min -0.999811 +trainer/policy/std Mean 0.409571 +trainer/policy/std Std 0.0187995 +trainer/policy/std Max 0.430494 +trainer/policy/std Min 0.379229 +trainer/Advantage Weights Mean 4.4093 +trainer/Advantage Weights Std 18.4162 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.31972e-10 +trainer/Advantage Score Mean -0.347788 +trainer/Advantage Score Std 0.485775 +trainer/Advantage Score Max 1.77769 +trainer/Advantage Score Min -2.27484 +trainer/V1 Predictions Mean -75.8804 +trainer/V1 Predictions Std 16.3553 +trainer/V1 Predictions Max -0.871269 +trainer/V1 Predictions Min -87.6282 +trainer/VF Loss 0.0615123 +expl/num steps total 857000 +expl/num paths total 1169 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.430866 +expl/Actions Std 0.786875 +expl/Actions Max 2.11201 +expl/Actions Min -2.70332 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 776438 +eval/num paths total 862 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.272146 +eval/Actions Std 0.747837 +eval/Actions Max 0.999603 +eval/Actions Min -0.999892 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78186e-06 +time/evaluation sampling (s) 3.12023 +time/exploration sampling (s) 4.35141 +time/logging (s) 0.00746539 +time/saving (s) 0.0125771 +time/training (s) 16.3617 +time/epoch (s) 23.8534 +time/total (s) 22239.5 +Epoch -144 +------------------------------ ---------------- +2022-05-16 00:13:35.079946 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -143 finished +------------------------------ ---------------- +epoch -143 +replay_buffer/size 999047 +trainer/num train calls 858000 +trainer/QF1 Loss 2.08385 +trainer/QF2 Loss 1.95037 +trainer/Policy Loss 9.04117 +trainer/Q1 Predictions Mean -75.1356 +trainer/Q1 Predictions Std 16.8861 +trainer/Q1 Predictions Max -0.661653 +trainer/Q1 Predictions Min -87.4016 +trainer/Q2 Predictions Mean -75.0926 +trainer/Q2 Predictions Std 16.7252 +trainer/Q2 Predictions Max -1.12889 +trainer/Q2 Predictions Min -87.1234 +trainer/Q Targets Mean -75.0061 +trainer/Q Targets Std 16.5206 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7278 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0126054 +trainer/policy/mean Std 0.729672 +trainer/policy/mean Max 0.999425 +trainer/policy/mean Min -0.998228 +trainer/policy/std Mean 0.407418 +trainer/policy/std Std 0.0189704 +trainer/policy/std Max 0.428813 +trainer/policy/std Min 0.376658 +trainer/Advantage Weights Mean 2.33867 +trainer/Advantage Weights Std 12.7442 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.656e-20 +trainer/Advantage Score Mean -0.44373 +trainer/Advantage Score Std 0.566623 +trainer/Advantage Score Max 1.05297 +trainer/Advantage Score Min -4.55473 +trainer/V1 Predictions Mean -74.7685 +trainer/V1 Predictions Std 16.8524 +trainer/V1 Predictions Max -1.07548 +trainer/V1 Predictions Min -87.5326 +trainer/VF Loss 0.0630391 +expl/num steps total 858000 +expl/num paths total 1170 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0966595 +expl/Actions Std 0.853887 +expl/Actions Max 2.19467 +expl/Actions Min -2.3591 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 777032 +eval/num paths total 863 +eval/path length Mean 594 +eval/path length Std 0 +eval/path length Max 594 +eval/path length Min 594 +eval/Rewards Mean 0.0016835 +eval/Rewards Std 0.0409959 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.022624 +eval/Actions Std 0.750574 +eval/Actions Max 0.999644 +eval/Actions Min -0.999467 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.17115e-06 +time/evaluation sampling (s) 3.9056 +time/exploration sampling (s) 3.85311 +time/logging (s) 0.0065352 +time/saving (s) 0.0108253 +time/training (s) 16.7243 +time/epoch (s) 24.5004 +time/total (s) 22264 +Epoch -143 +------------------------------ ---------------- +2022-05-16 00:13:59.523110 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -142 finished +------------------------------ ---------------- +epoch -142 +replay_buffer/size 999047 +trainer/num train calls 859000 +trainer/QF1 Loss 0.619269 +trainer/QF2 Loss 0.608808 +trainer/Policy Loss 3.63522 +trainer/Q1 Predictions Mean -75.1575 +trainer/Q1 Predictions Std 17.53 +trainer/Q1 Predictions Max -0.22075 +trainer/Q1 Predictions Min -87.7766 +trainer/Q2 Predictions Mean -75.2172 +trainer/Q2 Predictions Std 17.5047 +trainer/Q2 Predictions Max -0.00172174 +trainer/Q2 Predictions Min -87.578 +trainer/Q Targets Mean -74.9878 +trainer/Q Targets Std 17.2898 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1445 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00387692 +trainer/policy/mean Std 0.733744 +trainer/policy/mean Max 0.999267 +trainer/policy/mean Min -0.999447 +trainer/policy/std Mean 0.408266 +trainer/policy/std Std 0.0192226 +trainer/policy/std Max 0.429894 +trainer/policy/std Min 0.37628 +trainer/Advantage Weights Mean 1.47414 +trainer/Advantage Weights Std 10.8426 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.66582e-13 +trainer/Advantage Score Mean -0.500351 +trainer/Advantage Score Std 0.484036 +trainer/Advantage Score Max 1.09339 +trainer/Advantage Score Min -2.80366 +trainer/V1 Predictions Mean -74.7094 +trainer/V1 Predictions Std 17.4286 +trainer/V1 Predictions Max -0.430152 +trainer/V1 Predictions Min -87.1369 +trainer/VF Loss 0.0565495 +expl/num steps total 859000 +expl/num paths total 1172 +expl/path length Mean 500 +expl/path length Std 247 +expl/path length Max 747 +expl/path length Min 253 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0450894 +expl/Actions Std 0.827571 +expl/Actions Max 2.66751 +expl/Actions Min -2.36198 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 777677 +eval/num paths total 864 +eval/path length Mean 645 +eval/path length Std 0 +eval/path length Max 645 +eval/path length Min 645 +eval/Rewards Mean 0.00155039 +eval/Rewards Std 0.0393444 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.017668 +eval/Actions Std 0.739091 +eval/Actions Max 0.999532 +eval/Actions Min -0.999746 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.93135e-06 +time/evaluation sampling (s) 3.86421 +time/exploration sampling (s) 4.08589 +time/logging (s) 0.00699545 +time/saving (s) 0.010536 +time/training (s) 16.4614 +time/epoch (s) 24.429 +time/total (s) 22288.5 +Epoch -142 +------------------------------ ---------------- +2022-05-16 00:14:23.727395 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -141 finished +------------------------------ ---------------- +epoch -141 +replay_buffer/size 999047 +trainer/num train calls 860000 +trainer/QF1 Loss 20.718 +trainer/QF2 Loss 21.0812 +trainer/Policy Loss 23.7814 +trainer/Q1 Predictions Mean -74.5388 +trainer/Q1 Predictions Std 18.904 +trainer/Q1 Predictions Max -0.608651 +trainer/Q1 Predictions Min -87.8667 +trainer/Q2 Predictions Mean -74.546 +trainer/Q2 Predictions Std 18.9381 +trainer/Q2 Predictions Max -0.629713 +trainer/Q2 Predictions Min -87.9624 +trainer/Q Targets Mean -74.3967 +trainer/Q Targets Std 18.3595 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1039 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00976288 +trainer/policy/mean Std 0.723399 +trainer/policy/mean Max 0.999951 +trainer/policy/mean Min -0.999463 +trainer/policy/std Mean 0.408596 +trainer/policy/std Std 0.020253 +trainer/policy/std Max 0.429305 +trainer/policy/std Min 0.375966 +trainer/Advantage Weights Mean 3.81894 +trainer/Advantage Weights Std 17.5914 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.76947e-21 +trainer/Advantage Score Mean -0.453596 +trainer/Advantage Score Std 0.596422 +trainer/Advantage Score Max 2.34983 +trainer/Advantage Score Min -4.63041 +trainer/V1 Predictions Mean -73.9092 +trainer/V1 Predictions Std 18.8478 +trainer/V1 Predictions Max -0.765337 +trainer/V1 Predictions Min -87.0568 +trainer/VF Loss 0.0881156 +expl/num steps total 860000 +expl/num paths total 1173 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.116088 +expl/Actions Std 0.832926 +expl/Actions Max 2.325 +expl/Actions Min -2.666 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 778598 +eval/num paths total 865 +eval/path length Mean 921 +eval/path length Std 0 +eval/path length Max 921 +eval/path length Min 921 +eval/Rewards Mean 0.00108578 +eval/Rewards Std 0.0329332 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.036551 +eval/Actions Std 0.760228 +eval/Actions Max 0.999959 +eval/Actions Min -0.999894 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.67569e-06 +time/evaluation sampling (s) 3.72753 +time/exploration sampling (s) 4.53324 +time/logging (s) 0.0143968 +time/saving (s) 0.0186469 +time/training (s) 15.9052 +time/epoch (s) 24.199 +time/total (s) 22312.7 +Epoch -141 +------------------------------ ---------------- +2022-05-16 00:14:47.654063 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -140 finished +------------------------------ ---------------- +epoch -140 +replay_buffer/size 999047 +trainer/num train calls 861000 +trainer/QF1 Loss 0.527094 +trainer/QF2 Loss 0.432611 +trainer/Policy Loss 11.8585 +trainer/Q1 Predictions Mean -75.2996 +trainer/Q1 Predictions Std 15.4051 +trainer/Q1 Predictions Max -5.61676 +trainer/Q1 Predictions Min -87.6457 +trainer/Q2 Predictions Mean -75.3267 +trainer/Q2 Predictions Std 15.5164 +trainer/Q2 Predictions Max -6.1079 +trainer/Q2 Predictions Min -88.0115 +trainer/Q Targets Mean -75.2164 +trainer/Q Targets Std 15.6611 +trainer/Q Targets Max -6.90447 +trainer/Q Targets Min -87.9902 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00960015 +trainer/policy/mean Std 0.730553 +trainer/policy/mean Max 0.998211 +trainer/policy/mean Min -0.999101 +trainer/policy/std Mean 0.408478 +trainer/policy/std Std 0.0201051 +trainer/policy/std Max 0.427709 +trainer/policy/std Min 0.377822 +trainer/Advantage Weights Mean 3.93957 +trainer/Advantage Weights Std 15.6472 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.27516e-22 +trainer/Advantage Score Mean -0.385272 +trainer/Advantage Score Std 0.730299 +trainer/Advantage Score Max 1.24884 +trainer/Advantage Score Min -4.98348 +trainer/V1 Predictions Mean -74.9751 +trainer/V1 Predictions Std 15.8393 +trainer/V1 Predictions Max -4.79314 +trainer/V1 Predictions Min -87.8673 +trainer/VF Loss 0.0823776 +expl/num steps total 861000 +expl/num paths total 1174 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0363425 +expl/Actions Std 0.798171 +expl/Actions Max 2.26284 +expl/Actions Min -2.45175 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 779598 +eval/num paths total 866 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.288747 +eval/Actions Std 0.67545 +eval/Actions Max 0.999967 +eval/Actions Min -0.999939 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.95183e-06 +time/evaluation sampling (s) 3.36822 +time/exploration sampling (s) 4.50249 +time/logging (s) 0.00814471 +time/saving (s) 0.0109714 +time/training (s) 16.0132 +time/epoch (s) 23.903 +time/total (s) 22336.6 +Epoch -140 +------------------------------ ---------------- +2022-05-16 00:15:11.741230 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -139 finished +------------------------------ ---------------- +epoch -139 +replay_buffer/size 999047 +trainer/num train calls 862000 +trainer/QF1 Loss 1.12193 +trainer/QF2 Loss 1.25418 +trainer/Policy Loss 29.6343 +trainer/Q1 Predictions Mean -73.2435 +trainer/Q1 Predictions Std 18.6123 +trainer/Q1 Predictions Max -0.242625 +trainer/Q1 Predictions Min -87.2817 +trainer/Q2 Predictions Mean -73.1671 +trainer/Q2 Predictions Std 18.7002 +trainer/Q2 Predictions Max 0.800356 +trainer/Q2 Predictions Min -87.1604 +trainer/Q Targets Mean -73.8031 +trainer/Q Targets Std 18.5635 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7712 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0204596 +trainer/policy/mean Std 0.729785 +trainer/policy/mean Max 0.999435 +trainer/policy/mean Min -0.999523 +trainer/policy/std Mean 0.408946 +trainer/policy/std Std 0.0195405 +trainer/policy/std Max 0.42918 +trainer/policy/std Min 0.381463 +trainer/Advantage Weights Mean 6.40257 +trainer/Advantage Weights Std 21.4485 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.09743e-21 +trainer/Advantage Score Mean -0.2898 +trainer/Advantage Score Std 0.581751 +trainer/Advantage Score Max 1.24016 +trainer/Advantage Score Min -4.82613 +trainer/V1 Predictions Mean -73.5508 +trainer/V1 Predictions Std 18.7918 +trainer/V1 Predictions Max 0.909239 +trainer/V1 Predictions Min -87.6479 +trainer/VF Loss 0.0731291 +expl/num steps total 862000 +expl/num paths total 1175 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0361666 +expl/Actions Std 0.837255 +expl/Actions Max 2.39505 +expl/Actions Min -2.24857 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 780598 +eval/num paths total 867 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.08031 +eval/Actions Std 0.63903 +eval/Actions Max 0.999968 +eval/Actions Min -0.999875 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.6878e-06 +time/evaluation sampling (s) 3.51185 +time/exploration sampling (s) 4.53898 +time/logging (s) 0.00688936 +time/saving (s) 0.00963724 +time/training (s) 16.005 +time/epoch (s) 24.0723 +time/total (s) 22360.7 +Epoch -139 +------------------------------ ---------------- +2022-05-16 00:15:35.578764 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -138 finished +------------------------------ ---------------- +epoch -138 +replay_buffer/size 999047 +trainer/num train calls 863000 +trainer/QF1 Loss 0.746682 +trainer/QF2 Loss 0.605699 +trainer/Policy Loss 13.2396 +trainer/Q1 Predictions Mean -74.8458 +trainer/Q1 Predictions Std 18.1151 +trainer/Q1 Predictions Max -1.1269 +trainer/Q1 Predictions Min -88.1223 +trainer/Q2 Predictions Mean -74.8624 +trainer/Q2 Predictions Std 18.1233 +trainer/Q2 Predictions Max -0.722172 +trainer/Q2 Predictions Min -88.0204 +trainer/Q Targets Mean -74.8661 +trainer/Q Targets Std 18.0107 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8456 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0132786 +trainer/policy/mean Std 0.742944 +trainer/policy/mean Max 0.999633 +trainer/policy/mean Min -0.998271 +trainer/policy/std Mean 0.408458 +trainer/policy/std Std 0.01859 +trainer/policy/std Max 0.429785 +trainer/policy/std Min 0.378721 +trainer/Advantage Weights Mean 3.19007 +trainer/Advantage Weights Std 13.8282 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.7682e-17 +trainer/Advantage Score Mean -0.338005 +trainer/Advantage Score Std 0.562986 +trainer/Advantage Score Max 0.626019 +trainer/Advantage Score Min -3.7582 +trainer/V1 Predictions Mean -74.694 +trainer/V1 Predictions Std 18.0333 +trainer/V1 Predictions Max -0.650875 +trainer/V1 Predictions Min -87.6402 +trainer/VF Loss 0.0511507 +expl/num steps total 863000 +expl/num paths total 1177 +expl/path length Mean 500 +expl/path length Std 186 +expl/path length Max 686 +expl/path length Min 314 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0427213 +expl/Actions Std 0.833514 +expl/Actions Max 2.1948 +expl/Actions Min -2.51094 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 781598 +eval/num paths total 868 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.101761 +eval/Actions Std 0.578869 +eval/Actions Max 0.999164 +eval/Actions Min -0.999823 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.88617e-06 +time/evaluation sampling (s) 3.42868 +time/exploration sampling (s) 4.17291 +time/logging (s) 0.00756933 +time/saving (s) 0.0103991 +time/training (s) 16.2062 +time/epoch (s) 23.8257 +time/total (s) 22384.5 +Epoch -138 +------------------------------ ---------------- +2022-05-16 00:15:59.381294 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -137 finished +------------------------------ ---------------- +epoch -137 +replay_buffer/size 999047 +trainer/num train calls 864000 +trainer/QF1 Loss 0.822177 +trainer/QF2 Loss 0.96165 +trainer/Policy Loss 9.36426 +trainer/Q1 Predictions Mean -74.2034 +trainer/Q1 Predictions Std 18.76 +trainer/Q1 Predictions Max -1.12843 +trainer/Q1 Predictions Min -87.6524 +trainer/Q2 Predictions Mean -74.2645 +trainer/Q2 Predictions Std 18.7109 +trainer/Q2 Predictions Max -0.51317 +trainer/Q2 Predictions Min -87.862 +trainer/Q Targets Mean -73.944 +trainer/Q Targets Std 18.8626 +trainer/Q Targets Max -1.33677 +trainer/Q Targets Min -87.6633 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000534239 +trainer/policy/mean Std 0.732231 +trainer/policy/mean Max 0.999815 +trainer/policy/mean Min -0.999599 +trainer/policy/std Mean 0.408577 +trainer/policy/std Std 0.0186738 +trainer/policy/std Max 0.427487 +trainer/policy/std Min 0.377528 +trainer/Advantage Weights Mean 2.59812 +trainer/Advantage Weights Std 14.3203 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.72018e-18 +trainer/Advantage Score Mean -0.46189 +trainer/Advantage Score Std 0.595237 +trainer/Advantage Score Max 1.65054 +trainer/Advantage Score Min -3.95414 +trainer/V1 Predictions Mean -73.6725 +trainer/V1 Predictions Std 19.0673 +trainer/V1 Predictions Max 0.149204 +trainer/V1 Predictions Min -87.5354 +trainer/VF Loss 0.0769917 +expl/num steps total 864000 +expl/num paths total 1179 +expl/path length Mean 500 +expl/path length Std 118 +expl/path length Max 618 +expl/path length Min 382 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0325581 +expl/Actions Std 0.821653 +expl/Actions Max 2.37964 +expl/Actions Min -2.24802 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 782598 +eval/num paths total 869 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.171921 +eval/Actions Std 0.740259 +eval/Actions Max 0.9998 +eval/Actions Min -0.999591 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.83215e-06 +time/evaluation sampling (s) 3.63544 +time/exploration sampling (s) 4.36867 +time/logging (s) 0.00802923 +time/saving (s) 0.0104654 +time/training (s) 15.7679 +time/epoch (s) 23.7905 +time/total (s) 22408.3 +Epoch -137 +------------------------------ ---------------- +2022-05-16 00:16:24.732355 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -136 finished +------------------------------ ---------------- +epoch -136 +replay_buffer/size 999047 +trainer/num train calls 865000 +trainer/QF1 Loss 2.1683 +trainer/QF2 Loss 2.15603 +trainer/Policy Loss 18.7595 +trainer/Q1 Predictions Mean -73.2005 +trainer/Q1 Predictions Std 19.6239 +trainer/Q1 Predictions Max -1.1817 +trainer/Q1 Predictions Min -87.3439 +trainer/Q2 Predictions Mean -73.1845 +trainer/Q2 Predictions Std 19.6705 +trainer/Q2 Predictions Max -0.5577 +trainer/Q2 Predictions Min -87.5768 +trainer/Q Targets Mean -72.8531 +trainer/Q Targets Std 19.3522 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9926 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0265995 +trainer/policy/mean Std 0.74053 +trainer/policy/mean Max 0.999998 +trainer/policy/mean Min -0.998697 +trainer/policy/std Mean 0.407506 +trainer/policy/std Std 0.0200854 +trainer/policy/std Max 0.430456 +trainer/policy/std Min 0.378749 +trainer/Advantage Weights Mean 4.92503 +trainer/Advantage Weights Std 19.9529 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.12766e-10 +trainer/Advantage Score Mean -0.45433 +trainer/Advantage Score Std 0.500503 +trainer/Advantage Score Max 1.56125 +trainer/Advantage Score Min -2.18856 +trainer/V1 Predictions Mean -72.6038 +trainer/V1 Predictions Std 19.5671 +trainer/V1 Predictions Max -1.46286 +trainer/V1 Predictions Min -86.7407 +trainer/VF Loss 0.0782434 +expl/num steps total 865000 +expl/num paths total 1180 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.142285 +expl/Actions Std 0.856477 +expl/Actions Max 2.35756 +expl/Actions Min -2.38536 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 783598 +eval/num paths total 870 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.334768 +eval/Actions Std 0.69759 +eval/Actions Max 0.999523 +eval/Actions Min -0.999803 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.73716e-06 +time/evaluation sampling (s) 3.9171 +time/exploration sampling (s) 5.12272 +time/logging (s) 0.00752701 +time/saving (s) 0.010857 +time/training (s) 16.2806 +time/epoch (s) 25.3388 +time/total (s) 22433.6 +Epoch -136 +------------------------------ ---------------- +2022-05-16 00:16:49.296617 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -135 finished +------------------------------ ---------------- +epoch -135 +replay_buffer/size 999047 +trainer/num train calls 866000 +trainer/QF1 Loss 0.846359 +trainer/QF2 Loss 0.649264 +trainer/Policy Loss 25.178 +trainer/Q1 Predictions Mean -72.6975 +trainer/Q1 Predictions Std 18.9734 +trainer/Q1 Predictions Max -1.1446 +trainer/Q1 Predictions Min -87.0765 +trainer/Q2 Predictions Mean -72.7052 +trainer/Q2 Predictions Std 18.9684 +trainer/Q2 Predictions Max -2.04916 +trainer/Q2 Predictions Min -87.1587 +trainer/Q Targets Mean -72.9323 +trainer/Q Targets Std 18.789 +trainer/Q Targets Max -3.5439 +trainer/Q Targets Min -87.4335 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00773242 +trainer/policy/mean Std 0.728276 +trainer/policy/mean Max 0.999904 +trainer/policy/mean Min -0.999858 +trainer/policy/std Mean 0.407059 +trainer/policy/std Std 0.0204787 +trainer/policy/std Max 0.430471 +trainer/policy/std Min 0.376903 +trainer/Advantage Weights Mean 5.08042 +trainer/Advantage Weights Std 19.6822 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.16136e-13 +trainer/Advantage Score Mean -0.345875 +trainer/Advantage Score Std 0.574214 +trainer/Advantage Score Max 2.15868 +trainer/Advantage Score Min -2.9784 +trainer/V1 Predictions Mean -72.6696 +trainer/V1 Predictions Std 18.9112 +trainer/V1 Predictions Max -1.75282 +trainer/V1 Predictions Min -87.5269 +trainer/VF Loss 0.0891158 +expl/num steps total 866000 +expl/num paths total 1182 +expl/path length Mean 500 +expl/path length Std 155 +expl/path length Max 655 +expl/path length Min 345 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0197617 +expl/Actions Std 0.840612 +expl/Actions Max 2.46144 +expl/Actions Min -2.2272 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 784598 +eval/num paths total 871 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.23464 +eval/Actions Std 0.693604 +eval/Actions Max 0.999947 +eval/Actions Min -0.999946 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.11661e-05 +time/evaluation sampling (s) 3.21426 +time/exploration sampling (s) 4.66811 +time/logging (s) 0.00808666 +time/saving (s) 0.0127854 +time/training (s) 16.6487 +time/epoch (s) 24.5519 +time/total (s) 22458.2 +Epoch -135 +------------------------------ ---------------- +2022-05-16 00:17:13.145826 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -134 finished +------------------------------ ---------------- +epoch -134 +replay_buffer/size 999047 +trainer/num train calls 867000 +trainer/QF1 Loss 0.920666 +trainer/QF2 Loss 0.773663 +trainer/Policy Loss 9.58922 +trainer/Q1 Predictions Mean -75.2804 +trainer/Q1 Predictions Std 16.2692 +trainer/Q1 Predictions Max -1.75597 +trainer/Q1 Predictions Min -87.119 +trainer/Q2 Predictions Mean -75.3598 +trainer/Q2 Predictions Std 16.2588 +trainer/Q2 Predictions Max -2.25313 +trainer/Q2 Predictions Min -87.262 +trainer/Q Targets Mean -75.2125 +trainer/Q Targets Std 16.1219 +trainer/Q Targets Max -4.52081 +trainer/Q Targets Min -87.3703 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.020677 +trainer/policy/mean Std 0.730434 +trainer/policy/mean Max 0.998793 +trainer/policy/mean Min -0.99962 +trainer/policy/std Mean 0.404725 +trainer/policy/std Std 0.0196103 +trainer/policy/std Max 0.426 +trainer/policy/std Min 0.374558 +trainer/Advantage Weights Mean 3.97086 +trainer/Advantage Weights Std 18.0348 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.03455e-20 +trainer/Advantage Score Mean -0.437614 +trainer/Advantage Score Std 0.637537 +trainer/Advantage Score Max 1.06379 +trainer/Advantage Score Min -4.60177 +trainer/V1 Predictions Mean -74.9602 +trainer/V1 Predictions Std 16.3053 +trainer/V1 Predictions Max -1.82178 +trainer/V1 Predictions Min -87.2297 +trainer/VF Loss 0.0748042 +expl/num steps total 867000 +expl/num paths total 1183 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.059332 +expl/Actions Std 0.817641 +expl/Actions Max 2.48534 +expl/Actions Min -2.25308 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 785162 +eval/num paths total 872 +eval/path length Mean 564 +eval/path length Std 0 +eval/path length Max 564 +eval/path length Min 564 +eval/Rewards Mean 0.00177305 +eval/Rewards Std 0.0420703 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0207831 +eval/Actions Std 0.736811 +eval/Actions Max 0.999867 +eval/Actions Min -0.999764 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 9.18983e-06 +time/evaluation sampling (s) 3.26095 +time/exploration sampling (s) 4.01252 +time/logging (s) 0.00785028 +time/saving (s) 0.0128065 +time/training (s) 16.54 +time/epoch (s) 23.8341 +time/total (s) 22482 +Epoch -134 +------------------------------ ---------------- +2022-05-16 00:17:37.118887 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -133 finished +------------------------------ ---------------- +epoch -133 +replay_buffer/size 999047 +trainer/num train calls 868000 +trainer/QF1 Loss 1.00333 +trainer/QF2 Loss 0.871737 +trainer/Policy Loss 16.3126 +trainer/Q1 Predictions Mean -75.3747 +trainer/Q1 Predictions Std 17.2097 +trainer/Q1 Predictions Max -1.26466 +trainer/Q1 Predictions Min -88.3779 +trainer/Q2 Predictions Mean -75.2906 +trainer/Q2 Predictions Std 17.2376 +trainer/Q2 Predictions Max -1.09767 +trainer/Q2 Predictions Min -88.2555 +trainer/Q Targets Mean -75.2892 +trainer/Q Targets Std 17.0048 +trainer/Q Targets Max 0 +trainer/Q Targets Min -89.1075 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0212721 +trainer/policy/mean Std 0.733564 +trainer/policy/mean Max 0.999676 +trainer/policy/mean Min -0.999194 +trainer/policy/std Mean 0.407277 +trainer/policy/std Std 0.0195919 +trainer/policy/std Max 0.427605 +trainer/policy/std Min 0.376358 +trainer/Advantage Weights Mean 4.73039 +trainer/Advantage Weights Std 18.4356 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.68671e-16 +trainer/Advantage Score Mean -0.366352 +trainer/Advantage Score Std 0.588869 +trainer/Advantage Score Max 1.45063 +trainer/Advantage Score Min -3.5853 +trainer/V1 Predictions Mean -75.0036 +trainer/V1 Predictions Std 17.2522 +trainer/V1 Predictions Max 0.872461 +trainer/V1 Predictions Min -88.5771 +trainer/VF Loss 0.068982 +expl/num steps total 868000 +expl/num paths total 1184 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.132966 +expl/Actions Std 0.845089 +expl/Actions Max 2.26538 +expl/Actions Min -2.34136 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 786162 +eval/num paths total 873 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.000494396 +eval/Actions Std 0.750137 +eval/Actions Max 0.999791 +eval/Actions Min -0.999913 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.64123e-06 +time/evaluation sampling (s) 3.61772 +time/exploration sampling (s) 4.41977 +time/logging (s) 0.0122104 +time/saving (s) 0.0156437 +time/training (s) 15.8989 +time/epoch (s) 23.9643 +time/total (s) 22506 +Epoch -133 +------------------------------ ---------------- +2022-05-16 00:18:01.010022 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -132 finished +------------------------------ ---------------- +epoch -132 +replay_buffer/size 999047 +trainer/num train calls 869000 +trainer/QF1 Loss 0.529275 +trainer/QF2 Loss 0.503442 +trainer/Policy Loss 8.50905 +trainer/Q1 Predictions Mean -73.1109 +trainer/Q1 Predictions Std 18.9394 +trainer/Q1 Predictions Max -1.51304 +trainer/Q1 Predictions Min -86.8873 +trainer/Q2 Predictions Mean -73.1022 +trainer/Q2 Predictions Std 19.031 +trainer/Q2 Predictions Max -0.227566 +trainer/Q2 Predictions Min -86.8522 +trainer/Q Targets Mean -73.2495 +trainer/Q Targets Std 19.1397 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.1399 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0194345 +trainer/policy/mean Std 0.721876 +trainer/policy/mean Max 0.999431 +trainer/policy/mean Min -0.998634 +trainer/policy/std Mean 0.408921 +trainer/policy/std Std 0.0192705 +trainer/policy/std Max 0.430068 +trainer/policy/std Min 0.378583 +trainer/Advantage Weights Mean 2.28943 +trainer/Advantage Weights Std 12.6202 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.62653e-23 +trainer/Advantage Score Mean -0.454811 +trainer/Advantage Score Std 0.66629 +trainer/Advantage Score Max 1.87652 +trainer/Advantage Score Min -5.14276 +trainer/V1 Predictions Mean -72.9468 +trainer/V1 Predictions Std 19.1995 +trainer/V1 Predictions Max -0.512763 +trainer/V1 Predictions Min -87.0616 +trainer/VF Loss 0.0836454 +expl/num steps total 869000 +expl/num paths total 1185 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0302306 +expl/Actions Std 0.934214 +expl/Actions Max 2.38798 +expl/Actions Min -2.26349 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 786766 +eval/num paths total 874 +eval/path length Mean 604 +eval/path length Std 0 +eval/path length Max 604 +eval/path length Min 604 +eval/Rewards Mean 0.00165563 +eval/Rewards Std 0.0406557 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00981054 +eval/Actions Std 0.735499 +eval/Actions Max 0.999956 +eval/Actions Min -0.999991 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.70201e-06 +time/evaluation sampling (s) 3.40839 +time/exploration sampling (s) 4.52454 +time/logging (s) 0.00643312 +time/saving (s) 0.0104678 +time/training (s) 15.9155 +time/epoch (s) 23.8654 +time/total (s) 22529.9 +Epoch -132 +------------------------------ ---------------- +2022-05-16 00:18:26.207354 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -131 finished +------------------------------ ---------------- +epoch -131 +replay_buffer/size 999047 +trainer/num train calls 870000 +trainer/QF1 Loss 0.602558 +trainer/QF2 Loss 0.824958 +trainer/Policy Loss 21.0675 +trainer/Q1 Predictions Mean -74.69 +trainer/Q1 Predictions Std 17.0602 +trainer/Q1 Predictions Max -0.194196 +trainer/Q1 Predictions Min -87.4938 +trainer/Q2 Predictions Mean -74.758 +trainer/Q2 Predictions Std 17.0509 +trainer/Q2 Predictions Max -0.774535 +trainer/Q2 Predictions Min -87.6565 +trainer/Q Targets Mean -74.3752 +trainer/Q Targets Std 16.9437 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.446 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00424065 +trainer/policy/mean Std 0.739689 +trainer/policy/mean Max 0.999877 +trainer/policy/mean Min -0.999199 +trainer/policy/std Mean 0.409867 +trainer/policy/std Std 0.0206485 +trainer/policy/std Max 0.429655 +trainer/policy/std Min 0.37997 +trainer/Advantage Weights Mean 2.44275 +trainer/Advantage Weights Std 12.9831 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.16278e-12 +trainer/Advantage Score Mean -0.512514 +trainer/Advantage Score Std 0.476467 +trainer/Advantage Score Max 1.31393 +trainer/Advantage Score Min -2.64796 +trainer/V1 Predictions Mean -74.1623 +trainer/V1 Predictions Std 16.9421 +trainer/V1 Predictions Max -0.18217 +trainer/V1 Predictions Min -87.1518 +trainer/VF Loss 0.0602806 +expl/num steps total 870000 +expl/num paths total 1186 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0309037 +expl/Actions Std 0.825032 +expl/Actions Max 2.37077 +expl/Actions Min -2.2766 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 787766 +eval/num paths total 875 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.552035 +eval/Actions Std 0.601255 +eval/Actions Max 0.999898 +eval/Actions Min -0.999963 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.06851e-05 +time/evaluation sampling (s) 4.12428 +time/exploration sampling (s) 4.68305 +time/logging (s) 0.00885512 +time/saving (s) 0.0146003 +time/training (s) 16.3573 +time/epoch (s) 25.1881 +time/total (s) 22555.1 +Epoch -131 +------------------------------ ---------------- +2022-05-16 00:18:51.116833 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -130 finished +------------------------------ ---------------- +epoch -130 +replay_buffer/size 999047 +trainer/num train calls 871000 +trainer/QF1 Loss 0.833298 +trainer/QF2 Loss 0.837499 +trainer/Policy Loss 18.7707 +trainer/Q1 Predictions Mean -75.1074 +trainer/Q1 Predictions Std 16.4608 +trainer/Q1 Predictions Max -1.18214 +trainer/Q1 Predictions Min -87.4174 +trainer/Q2 Predictions Mean -75.1581 +trainer/Q2 Predictions Std 16.5456 +trainer/Q2 Predictions Max -1.85633 +trainer/Q2 Predictions Min -87.4199 +trainer/Q Targets Mean -75.3177 +trainer/Q Targets Std 16.4032 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6115 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0312079 +trainer/policy/mean Std 0.741409 +trainer/policy/mean Max 0.999316 +trainer/policy/mean Min -0.99942 +trainer/policy/std Mean 0.407546 +trainer/policy/std Std 0.0213517 +trainer/policy/std Max 0.430792 +trainer/policy/std Min 0.375713 +trainer/Advantage Weights Mean 5.61735 +trainer/Advantage Weights Std 19.9064 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.93376e-21 +trainer/Advantage Score Mean -0.305751 +trainer/Advantage Score Std 0.593232 +trainer/Advantage Score Max 1.25676 +trainer/Advantage Score Min -4.69847 +trainer/V1 Predictions Mean -74.9641 +trainer/V1 Predictions Std 16.626 +trainer/V1 Predictions Max -1.70676 +trainer/V1 Predictions Min -87.4428 +trainer/VF Loss 0.0648541 +expl/num steps total 871000 +expl/num paths total 1188 +expl/path length Mean 500 +expl/path length Std 452 +expl/path length Max 952 +expl/path length Min 48 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00175857 +expl/Actions Std 0.844727 +expl/Actions Max 2.38955 +expl/Actions Min -2.51935 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 788284 +eval/num paths total 876 +eval/path length Mean 518 +eval/path length Std 0 +eval/path length Max 518 +eval/path length Min 518 +eval/Rewards Mean 0.0019305 +eval/Rewards Std 0.043895 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0143793 +eval/Actions Std 0.748226 +eval/Actions Max 0.999768 +eval/Actions Min -0.999515 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.71387e-06 +time/evaluation sampling (s) 3.24388 +time/exploration sampling (s) 5.12103 +time/logging (s) 0.00893016 +time/saving (s) 0.016237 +time/training (s) 16.5066 +time/epoch (s) 24.8967 +time/total (s) 22580 +Epoch -130 +------------------------------ ---------------- +2022-05-16 00:19:15.089982 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -129 finished +------------------------------ ---------------- +epoch -129 +replay_buffer/size 999047 +trainer/num train calls 872000 +trainer/QF1 Loss 0.674762 +trainer/QF2 Loss 0.768474 +trainer/Policy Loss 17.3417 +trainer/Q1 Predictions Mean -74.7524 +trainer/Q1 Predictions Std 16.5906 +trainer/Q1 Predictions Max -4.17465 +trainer/Q1 Predictions Min -87.3739 +trainer/Q2 Predictions Mean -74.79 +trainer/Q2 Predictions Std 16.5165 +trainer/Q2 Predictions Max -5.05187 +trainer/Q2 Predictions Min -87.3687 +trainer/Q Targets Mean -74.8965 +trainer/Q Targets Std 16.5979 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.656 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0236289 +trainer/policy/mean Std 0.734799 +trainer/policy/mean Max 0.999744 +trainer/policy/mean Min -0.999232 +trainer/policy/std Mean 0.408997 +trainer/policy/std Std 0.0213056 +trainer/policy/std Max 0.430204 +trainer/policy/std Min 0.379125 +trainer/Advantage Weights Mean 3.14283 +trainer/Advantage Weights Std 15.7194 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.21937e-17 +trainer/Advantage Score Mean -0.449678 +trainer/Advantage Score Std 0.543093 +trainer/Advantage Score Max 1.27182 +trainer/Advantage Score Min -3.74916 +trainer/V1 Predictions Mean -74.7267 +trainer/V1 Predictions Std 16.4529 +trainer/V1 Predictions Max -4.52743 +trainer/V1 Predictions Min -87.4954 +trainer/VF Loss 0.0638247 +expl/num steps total 872000 +expl/num paths total 1189 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0191697 +expl/Actions Std 0.820099 +expl/Actions Max 2.50969 +expl/Actions Min -2.18327 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 788789 +eval/num paths total 877 +eval/path length Mean 505 +eval/path length Std 0 +eval/path length Max 505 +eval/path length Min 505 +eval/Rewards Mean 0.0019802 +eval/Rewards Std 0.0444553 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0302007 +eval/Actions Std 0.73904 +eval/Actions Max 0.999515 +eval/Actions Min -0.999744 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.90573e-06 +time/evaluation sampling (s) 3.12206 +time/exploration sampling (s) 4.12156 +time/logging (s) 0.00908177 +time/saving (s) 0.0152401 +time/training (s) 16.6872 +time/epoch (s) 23.9551 +time/total (s) 22603.9 +Epoch -129 +------------------------------ ---------------- +2022-05-16 00:19:38.963323 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -128 finished +------------------------------ ---------------- +epoch -128 +replay_buffer/size 999047 +trainer/num train calls 873000 +trainer/QF1 Loss 0.50637 +trainer/QF2 Loss 0.474042 +trainer/Policy Loss 5.19269 +trainer/Q1 Predictions Mean -74.5135 +trainer/Q1 Predictions Std 20.0031 +trainer/Q1 Predictions Max -0.36397 +trainer/Q1 Predictions Min -87.6776 +trainer/Q2 Predictions Mean -74.3939 +trainer/Q2 Predictions Std 19.9911 +trainer/Q2 Predictions Max 0.259122 +trainer/Q2 Predictions Min -87.4437 +trainer/Q Targets Mean -74.5878 +trainer/Q Targets Std 20.0215 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.966 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00177091 +trainer/policy/mean Std 0.723183 +trainer/policy/mean Max 0.999382 +trainer/policy/mean Min -0.999297 +trainer/policy/std Mean 0.407665 +trainer/policy/std Std 0.0203298 +trainer/policy/std Max 0.429669 +trainer/policy/std Min 0.378683 +trainer/Advantage Weights Mean 2.27308 +trainer/Advantage Weights Std 13.054 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.84053e-17 +trainer/Advantage Score Mean -0.513508 +trainer/Advantage Score Std 0.606371 +trainer/Advantage Score Max 0.910683 +trainer/Advantage Score Min -3.81 +trainer/V1 Predictions Mean -74.3727 +trainer/V1 Predictions Std 20.1455 +trainer/V1 Predictions Max 0.327488 +trainer/V1 Predictions Min -87.8309 +trainer/VF Loss 0.0706343 +expl/num steps total 873000 +expl/num paths total 1190 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.191213 +expl/Actions Std 0.869512 +expl/Actions Max 2.39131 +expl/Actions Min -2.4349 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 789789 +eval/num paths total 878 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.194842 +eval/Actions Std 0.705941 +eval/Actions Max 0.999974 +eval/Actions Min -0.9998 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.746e-06 +time/evaluation sampling (s) 3.77858 +time/exploration sampling (s) 3.79855 +time/logging (s) 0.00791227 +time/saving (s) 0.0137991 +time/training (s) 16.2549 +time/epoch (s) 23.8537 +time/total (s) 22627.8 +Epoch -128 +------------------------------ ---------------- +2022-05-16 00:20:03.941764 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -127 finished +------------------------------ ---------------- +epoch -127 +replay_buffer/size 999047 +trainer/num train calls 874000 +trainer/QF1 Loss 0.592701 +trainer/QF2 Loss 0.686714 +trainer/Policy Loss 5.99966 +trainer/Q1 Predictions Mean -74.8479 +trainer/Q1 Predictions Std 17.1338 +trainer/Q1 Predictions Max -0.949023 +trainer/Q1 Predictions Min -87.9795 +trainer/Q2 Predictions Mean -74.891 +trainer/Q2 Predictions Std 17.1932 +trainer/Q2 Predictions Max -1.55956 +trainer/Q2 Predictions Min -88.0632 +trainer/Q Targets Mean -74.7014 +trainer/Q Targets Std 17.2934 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6209 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0138652 +trainer/policy/mean Std 0.7262 +trainer/policy/mean Max 0.999746 +trainer/policy/mean Min -0.999054 +trainer/policy/std Mean 0.409741 +trainer/policy/std Std 0.020073 +trainer/policy/std Max 0.429685 +trainer/policy/std Min 0.382044 +trainer/Advantage Weights Mean 1.09827 +trainer/Advantage Weights Std 9.03644 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.53159e-15 +trainer/Advantage Score Mean -0.60773 +trainer/Advantage Score Std 0.544483 +trainer/Advantage Score Max 0.736115 +trainer/Advantage Score Min -3.28283 +trainer/V1 Predictions Mean -74.4853 +trainer/V1 Predictions Std 17.3636 +trainer/V1 Predictions Max -0.806245 +trainer/V1 Predictions Min -87.4968 +trainer/VF Loss 0.0707586 +expl/num steps total 874000 +expl/num paths total 1192 +expl/path length Mean 500 +expl/path length Std 190 +expl/path length Max 690 +expl/path length Min 310 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0307495 +expl/Actions Std 0.830942 +expl/Actions Max 2.06327 +expl/Actions Min -2.34994 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 790632 +eval/num paths total 879 +eval/path length Mean 843 +eval/path length Std 0 +eval/path length Max 843 +eval/path length Min 843 +eval/Rewards Mean 0.00118624 +eval/Rewards Std 0.0344214 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0203901 +eval/Actions Std 0.737602 +eval/Actions Max 0.999976 +eval/Actions Min -0.999791 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.63262e-06 +time/evaluation sampling (s) 4.12848 +time/exploration sampling (s) 4.67188 +time/logging (s) 0.0112861 +time/saving (s) 0.0177265 +time/training (s) 16.1399 +time/epoch (s) 24.9692 +time/total (s) 22652.8 +Epoch -127 +------------------------------ ---------------- +2022-05-16 00:20:28.000394 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -126 finished +------------------------------ ---------------- +epoch -126 +replay_buffer/size 999047 +trainer/num train calls 875000 +trainer/QF1 Loss 1.00298 +trainer/QF2 Loss 0.9977 +trainer/Policy Loss 51.4783 +trainer/Q1 Predictions Mean -73.8988 +trainer/Q1 Predictions Std 19.1119 +trainer/Q1 Predictions Max -1.42964 +trainer/Q1 Predictions Min -87.4928 +trainer/Q2 Predictions Mean -73.9736 +trainer/Q2 Predictions Std 19.0578 +trainer/Q2 Predictions Max -2.59542 +trainer/Q2 Predictions Min -87.431 +trainer/Q Targets Mean -74.2169 +trainer/Q Targets Std 19.5101 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8047 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00453326 +trainer/policy/mean Std 0.737851 +trainer/policy/mean Max 0.999173 +trainer/policy/mean Min -0.999501 +trainer/policy/std Mean 0.409509 +trainer/policy/std Std 0.0195113 +trainer/policy/std Max 0.428084 +trainer/policy/std Min 0.377428 +trainer/Advantage Weights Mean 10.4729 +trainer/Advantage Weights Std 26.9195 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.40177e-14 +trainer/Advantage Score Mean -0.251399 +trainer/Advantage Score Std 0.607896 +trainer/Advantage Score Max 2.45978 +trainer/Advantage Score Min -3.07542 +trainer/V1 Predictions Mean -74.0514 +trainer/V1 Predictions Std 19.4322 +trainer/V1 Predictions Max -0.169726 +trainer/V1 Predictions Min -87.7542 +trainer/VF Loss 0.0969374 +expl/num steps total 875000 +expl/num paths total 1193 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.121213 +expl/Actions Std 0.830221 +expl/Actions Max 2.43296 +expl/Actions Min -2.38531 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 791236 +eval/num paths total 880 +eval/path length Mean 604 +eval/path length Std 0 +eval/path length Max 604 +eval/path length Min 604 +eval/Rewards Mean 0.00165563 +eval/Rewards Std 0.0406557 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0392925 +eval/Actions Std 0.748603 +eval/Actions Max 0.999948 +eval/Actions Min -0.999548 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.56814e-06 +time/evaluation sampling (s) 3.13182 +time/exploration sampling (s) 4.50494 +time/logging (s) 0.00725479 +time/saving (s) 0.0132185 +time/training (s) 16.3768 +time/epoch (s) 24.0341 +time/total (s) 22676.8 +Epoch -126 +------------------------------ ---------------- +2022-05-16 00:20:52.365421 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -125 finished +------------------------------ ---------------- +epoch -125 +replay_buffer/size 999047 +trainer/num train calls 876000 +trainer/QF1 Loss 0.465417 +trainer/QF2 Loss 0.498512 +trainer/Policy Loss 7.50534 +trainer/Q1 Predictions Mean -75.2339 +trainer/Q1 Predictions Std 16.3293 +trainer/Q1 Predictions Max 0.135534 +trainer/Q1 Predictions Min -88.5992 +trainer/Q2 Predictions Mean -75.192 +trainer/Q2 Predictions Std 16.3976 +trainer/Q2 Predictions Max 1.56409 +trainer/Q2 Predictions Min -88.5186 +trainer/Q Targets Mean -75.1126 +trainer/Q Targets Std 16.2461 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2446 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0258588 +trainer/policy/mean Std 0.739671 +trainer/policy/mean Max 0.99972 +trainer/policy/mean Min -0.998849 +trainer/policy/std Mean 0.409292 +trainer/policy/std Std 0.0185605 +trainer/policy/std Max 0.429908 +trainer/policy/std Min 0.380848 +trainer/Advantage Weights Mean 2.56265 +trainer/Advantage Weights Std 12.8526 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.69287e-21 +trainer/Advantage Score Mean -0.378789 +trainer/Advantage Score Std 0.62837 +trainer/Advantage Score Max 2.18637 +trainer/Advantage Score Min -4.68082 +trainer/V1 Predictions Mean -74.8601 +trainer/V1 Predictions Std 16.3259 +trainer/V1 Predictions Max 0.444448 +trainer/V1 Predictions Min -88.0794 +trainer/VF Loss 0.0860722 +expl/num steps total 876000 +expl/num paths total 1195 +expl/path length Mean 500 +expl/path length Std 190 +expl/path length Max 690 +expl/path length Min 310 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0364106 +expl/Actions Std 0.819609 +expl/Actions Max 2.29195 +expl/Actions Min -2.42589 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 792236 +eval/num paths total 881 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.250507 +eval/Actions Std 0.602276 +eval/Actions Max 0.999858 +eval/Actions Min -0.999475 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.0062e-05 +time/evaluation sampling (s) 3.47983 +time/exploration sampling (s) 4.32992 +time/logging (s) 0.00941985 +time/saving (s) 0.0182462 +time/training (s) 16.5152 +time/epoch (s) 24.3527 +time/total (s) 22701.2 +Epoch -125 +------------------------------ ---------------- +2022-05-16 00:21:16.021263 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -124 finished +------------------------------ ---------------- +epoch -124 +replay_buffer/size 999047 +trainer/num train calls 877000 +trainer/QF1 Loss 0.501723 +trainer/QF2 Loss 0.569963 +trainer/Policy Loss 11.8021 +trainer/Q1 Predictions Mean -76.1632 +trainer/Q1 Predictions Std 15.6458 +trainer/Q1 Predictions Max -0.377066 +trainer/Q1 Predictions Min -87.7561 +trainer/Q2 Predictions Mean -76.1776 +trainer/Q2 Predictions Std 15.571 +trainer/Q2 Predictions Max -0.0560946 +trainer/Q2 Predictions Min -88.331 +trainer/Q Targets Mean -76.0466 +trainer/Q Targets Std 15.7734 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6501 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.014898 +trainer/policy/mean Std 0.737808 +trainer/policy/mean Max 0.999622 +trainer/policy/mean Min -0.999736 +trainer/policy/std Mean 0.410005 +trainer/policy/std Std 0.0197762 +trainer/policy/std Max 0.430716 +trainer/policy/std Min 0.378588 +trainer/Advantage Weights Mean 2.29843 +trainer/Advantage Weights Std 10.7306 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.45018e-17 +trainer/Advantage Score Mean -0.365932 +trainer/Advantage Score Std 0.587785 +trainer/Advantage Score Max 0.919973 +trainer/Advantage Score Min -3.79055 +trainer/V1 Predictions Mean -75.7934 +trainer/V1 Predictions Std 16.0569 +trainer/V1 Predictions Max 0.157985 +trainer/V1 Predictions Min -87.4468 +trainer/VF Loss 0.0558952 +expl/num steps total 877000 +expl/num paths total 1197 +expl/path length Mean 500 +expl/path length Std 280 +expl/path length Max 780 +expl/path length Min 220 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00325552 +expl/Actions Std 0.814476 +expl/Actions Max 2.33823 +expl/Actions Min -2.44245 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 792922 +eval/num paths total 882 +eval/path length Mean 686 +eval/path length Std 0 +eval/path length Max 686 +eval/path length Min 686 +eval/Rewards Mean 0.00145773 +eval/Rewards Std 0.0381523 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.0188174 +eval/Actions Std 0.739903 +eval/Actions Max 0.999783 +eval/Actions Min -0.999243 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.27406e-06 +time/evaluation sampling (s) 3.72501 +time/exploration sampling (s) 3.92946 +time/logging (s) 0.00624235 +time/saving (s) 0.00977874 +time/training (s) 15.9655 +time/epoch (s) 23.636 +time/total (s) 22724.8 +Epoch -124 +------------------------------ ---------------- +2022-05-16 00:21:39.993180 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -123 finished +------------------------------ ---------------- +epoch -123 +replay_buffer/size 999047 +trainer/num train calls 878000 +trainer/QF1 Loss 0.885799 +trainer/QF2 Loss 0.723771 +trainer/Policy Loss 11.0445 +trainer/Q1 Predictions Mean -72.5 +trainer/Q1 Predictions Std 20.7128 +trainer/Q1 Predictions Max -0.364477 +trainer/Q1 Predictions Min -87.8473 +trainer/Q2 Predictions Mean -72.4402 +trainer/Q2 Predictions Std 20.6971 +trainer/Q2 Predictions Max -0.0886169 +trainer/Q2 Predictions Min -87.7179 +trainer/Q Targets Mean -72.3195 +trainer/Q Targets Std 20.7825 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9354 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0153717 +trainer/policy/mean Std 0.727947 +trainer/policy/mean Max 0.999871 +trainer/policy/mean Min -0.998101 +trainer/policy/std Mean 0.40947 +trainer/policy/std Std 0.0194898 +trainer/policy/std Max 0.428001 +trainer/policy/std Min 0.377708 +trainer/Advantage Weights Mean 2.25347 +trainer/Advantage Weights Std 11.3322 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.57924e-16 +trainer/Advantage Score Mean -0.435862 +trainer/Advantage Score Std 0.586209 +trainer/Advantage Score Max 0.724373 +trainer/Advantage Score Min -3.58939 +trainer/V1 Predictions Mean -72.0607 +trainer/V1 Predictions Std 20.8976 +trainer/V1 Predictions Max 0.119721 +trainer/V1 Predictions Min -87.8071 +trainer/VF Loss 0.0602752 +expl/num steps total 878000 +expl/num paths total 1198 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.039597 +expl/Actions Std 0.83481 +expl/Actions Max 2.39353 +expl/Actions Min -2.23644 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 793707 +eval/num paths total 883 +eval/path length Mean 785 +eval/path length Std 0 +eval/path length Max 785 +eval/path length Min 785 +eval/Rewards Mean 0.00127389 +eval/Rewards Std 0.0356688 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.0427304 +eval/Actions Std 0.747657 +eval/Actions Max 0.999792 +eval/Actions Min -0.999441 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.21889e-06 +time/evaluation sampling (s) 3.43095 +time/exploration sampling (s) 4.60169 +time/logging (s) 0.00894679 +time/saving (s) 0.0162979 +time/training (s) 15.904 +time/epoch (s) 23.9619 +time/total (s) 22748.8 +Epoch -123 +------------------------------ ---------------- +2022-05-16 00:22:04.027756 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -122 finished +------------------------------ ---------------- +epoch -122 +replay_buffer/size 999047 +trainer/num train calls 879000 +trainer/QF1 Loss 0.761873 +trainer/QF2 Loss 0.869185 +trainer/Policy Loss 20.9089 +trainer/Q1 Predictions Mean -73.0506 +trainer/Q1 Predictions Std 18.9712 +trainer/Q1 Predictions Max -0.988192 +trainer/Q1 Predictions Min -87.5455 +trainer/Q2 Predictions Mean -72.9654 +trainer/Q2 Predictions Std 18.9829 +trainer/Q2 Predictions Max 1.11301 +trainer/Q2 Predictions Min -87.3968 +trainer/Q Targets Mean -73.3676 +trainer/Q Targets Std 19.1618 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1737 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0023283 +trainer/policy/mean Std 0.72137 +trainer/policy/mean Max 0.999093 +trainer/policy/mean Min -0.998762 +trainer/policy/std Mean 0.408494 +trainer/policy/std Std 0.0184753 +trainer/policy/std Max 0.425768 +trainer/policy/std Min 0.379493 +trainer/Advantage Weights Mean 4.52579 +trainer/Advantage Weights Std 16.7337 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.90769e-17 +trainer/Advantage Score Mean -0.316951 +trainer/Advantage Score Std 0.634177 +trainer/Advantage Score Max 1.95144 +trainer/Advantage Score Min -3.70761 +trainer/V1 Predictions Mean -73.1801 +trainer/V1 Predictions Std 19.0968 +trainer/V1 Predictions Max -0.00107473 +trainer/V1 Predictions Min -87.9293 +trainer/VF Loss 0.0759146 +expl/num steps total 879000 +expl/num paths total 1200 +expl/path length Mean 500 +expl/path length Std 160 +expl/path length Max 660 +expl/path length Min 340 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0298789 +expl/Actions Std 0.834149 +expl/Actions Max 2.45913 +expl/Actions Min -2.51904 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 794217 +eval/num paths total 884 +eval/path length Mean 510 +eval/path length Std 0 +eval/path length Max 510 +eval/path length Min 510 +eval/Rewards Mean 0.00196078 +eval/Rewards Std 0.0442373 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0371654 +eval/Actions Std 0.737646 +eval/Actions Max 0.999797 +eval/Actions Min -0.999878 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.40328e-06 +time/evaluation sampling (s) 3.44253 +time/exploration sampling (s) 4.27764 +time/logging (s) 0.00990666 +time/saving (s) 0.0191637 +time/training (s) 16.2681 +time/epoch (s) 24.0173 +time/total (s) 22772.8 +Epoch -122 +------------------------------ ---------------- +2022-05-16 00:22:27.989686 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -121 finished +------------------------------ ---------------- +epoch -121 +replay_buffer/size 999047 +trainer/num train calls 880000 +trainer/QF1 Loss 0.554808 +trainer/QF2 Loss 0.612079 +trainer/Policy Loss 7.22723 +trainer/Q1 Predictions Mean -72.4797 +trainer/Q1 Predictions Std 19.7487 +trainer/Q1 Predictions Max -0.667603 +trainer/Q1 Predictions Min -87.8282 +trainer/Q2 Predictions Mean -72.4217 +trainer/Q2 Predictions Std 19.79 +trainer/Q2 Predictions Max -1.53124 +trainer/Q2 Predictions Min -87.9788 +trainer/Q Targets Mean -72.586 +trainer/Q Targets Std 19.8264 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9419 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00903723 +trainer/policy/mean Std 0.736876 +trainer/policy/mean Max 0.999293 +trainer/policy/mean Min -0.999843 +trainer/policy/std Mean 0.407912 +trainer/policy/std Std 0.0197177 +trainer/policy/std Max 0.428825 +trainer/policy/std Min 0.377796 +trainer/Advantage Weights Mean 2.86376 +trainer/Advantage Weights Std 13.2188 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.80489e-16 +trainer/Advantage Score Mean -0.398121 +trainer/Advantage Score Std 0.623898 +trainer/Advantage Score Max 1.28374 +trainer/Advantage Score Min -3.46661 +trainer/V1 Predictions Mean -72.332 +trainer/V1 Predictions Std 20.0097 +trainer/V1 Predictions Max -0.422997 +trainer/V1 Predictions Min -87.7911 +trainer/VF Loss 0.0718244 +expl/num steps total 880000 +expl/num paths total 1202 +expl/path length Mean 500 +expl/path length Std 276 +expl/path length Max 776 +expl/path length Min 224 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0276687 +expl/Actions Std 0.82414 +expl/Actions Max 2.33782 +expl/Actions Min -2.53952 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 794954 +eval/num paths total 885 +eval/path length Mean 737 +eval/path length Std 0 +eval/path length Max 737 +eval/path length Min 737 +eval/Rewards Mean 0.00135685 +eval/Rewards Std 0.0368105 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0313257 +eval/Actions Std 0.759246 +eval/Actions Max 0.999837 +eval/Actions Min -0.999708 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 1.0381e-05 +time/evaluation sampling (s) 3.19579 +time/exploration sampling (s) 5.04694 +time/logging (s) 0.0114364 +time/saving (s) 0.0122139 +time/training (s) 15.6756 +time/epoch (s) 23.942 +time/total (s) 22796.7 +Epoch -121 +------------------------------ ---------------- +2022-05-16 00:22:51.370619 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -120 finished +------------------------------ ---------------- +epoch -120 +replay_buffer/size 999047 +trainer/num train calls 881000 +trainer/QF1 Loss 0.612988 +trainer/QF2 Loss 0.597837 +trainer/Policy Loss 4.86946 +trainer/Q1 Predictions Mean -74.8257 +trainer/Q1 Predictions Std 17.6867 +trainer/Q1 Predictions Max -0.659798 +trainer/Q1 Predictions Min -87.8564 +trainer/Q2 Predictions Mean -74.8888 +trainer/Q2 Predictions Std 17.6645 +trainer/Q2 Predictions Max -0.371731 +trainer/Q2 Predictions Min -87.9685 +trainer/Q Targets Mean -74.7083 +trainer/Q Targets Std 17.9073 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7732 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0141252 +trainer/policy/mean Std 0.739412 +trainer/policy/mean Max 0.99943 +trainer/policy/mean Min -0.999592 +trainer/policy/std Mean 0.408171 +trainer/policy/std Std 0.0188351 +trainer/policy/std Max 0.428152 +trainer/policy/std Min 0.380531 +trainer/Advantage Weights Mean 1.85837 +trainer/Advantage Weights Std 12.4596 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.6155e-19 +trainer/Advantage Score Mean -0.622665 +trainer/Advantage Score Std 0.692121 +trainer/Advantage Score Max 1.09945 +trainer/Advantage Score Min -4.15956 +trainer/V1 Predictions Mean -74.4226 +trainer/V1 Predictions Std 18.1322 +trainer/V1 Predictions Max 0.133312 +trainer/V1 Predictions Min -87.6499 +trainer/VF Loss 0.0955741 +expl/num steps total 881000 +expl/num paths total 1204 +expl/path length Mean 500 +expl/path length Std 354 +expl/path length Max 854 +expl/path length Min 146 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0499571 +expl/Actions Std 0.824891 +expl/Actions Max 2.35618 +expl/Actions Min -2.41982 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 795853 +eval/num paths total 886 +eval/path length Mean 899 +eval/path length Std 0 +eval/path length Max 899 +eval/path length Min 899 +eval/Rewards Mean 0.00111235 +eval/Rewards Std 0.0333333 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0225896 +eval/Actions Std 0.735758 +eval/Actions Max 0.999959 +eval/Actions Min -0.999756 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.66079e-06 +time/evaluation sampling (s) 3.7915 +time/exploration sampling (s) 3.63119 +time/logging (s) 0.00774594 +time/saving (s) 0.0107324 +time/training (s) 15.9213 +time/epoch (s) 23.3625 +time/total (s) 22820.1 +Epoch -120 +------------------------------ ---------------- +2022-05-16 00:23:15.751392 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -119 finished +------------------------------ ---------------- +epoch -119 +replay_buffer/size 999047 +trainer/num train calls 882000 +trainer/QF1 Loss 0.86165 +trainer/QF2 Loss 1.002 +trainer/Policy Loss 11.619 +trainer/Q1 Predictions Mean -72.8463 +trainer/Q1 Predictions Std 18.6375 +trainer/Q1 Predictions Max -0.575254 +trainer/Q1 Predictions Min -87.5078 +trainer/Q2 Predictions Mean -72.9297 +trainer/Q2 Predictions Std 18.5938 +trainer/Q2 Predictions Max -0.709005 +trainer/Q2 Predictions Min -87.7002 +trainer/Q Targets Mean -72.5156 +trainer/Q Targets Std 18.7923 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4083 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0182886 +trainer/policy/mean Std 0.729305 +trainer/policy/mean Max 0.999283 +trainer/policy/mean Min -0.99944 +trainer/policy/std Mean 0.409215 +trainer/policy/std Std 0.0195948 +trainer/policy/std Max 0.429962 +trainer/policy/std Min 0.379613 +trainer/Advantage Weights Mean 1.7445 +trainer/Advantage Weights Std 11.5507 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.99294e-14 +trainer/Advantage Score Mean -0.651474 +trainer/Advantage Score Std 0.536432 +trainer/Advantage Score Max 1.1675 +trainer/Advantage Score Min -3.02913 +trainer/V1 Predictions Mean -72.3751 +trainer/V1 Predictions Std 18.5975 +trainer/V1 Predictions Max 0.463921 +trainer/V1 Predictions Min -87.251 +trainer/VF Loss 0.0790885 +expl/num steps total 882000 +expl/num paths total 1206 +expl/path length Mean 500 +expl/path length Std 288 +expl/path length Max 788 +expl/path length Min 212 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0321227 +expl/Actions Std 0.830284 +expl/Actions Max 2.65322 +expl/Actions Min -2.14408 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 796853 +eval/num paths total 887 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.106053 +eval/Actions Std 0.729181 +eval/Actions Max 0.997643 +eval/Actions Min -0.999193 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.96999e-06 +time/evaluation sampling (s) 3.67547 +time/exploration sampling (s) 4.63954 +time/logging (s) 0.00817288 +time/saving (s) 0.0125683 +time/training (s) 16.0313 +time/epoch (s) 24.3671 +time/total (s) 22844.5 +Epoch -119 +------------------------------ ---------------- +2022-05-16 00:23:39.671593 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -118 finished +------------------------------ ---------------- +epoch -118 +replay_buffer/size 999047 +trainer/num train calls 883000 +trainer/QF1 Loss 0.692521 +trainer/QF2 Loss 0.670785 +trainer/Policy Loss 6.01824 +trainer/Q1 Predictions Mean -73.9405 +trainer/Q1 Predictions Std 18.9208 +trainer/Q1 Predictions Max -0.231703 +trainer/Q1 Predictions Min -88.3227 +trainer/Q2 Predictions Mean -73.9908 +trainer/Q2 Predictions Std 18.8609 +trainer/Q2 Predictions Max -0.142997 +trainer/Q2 Predictions Min -88.2469 +trainer/Q Targets Mean -73.6129 +trainer/Q Targets Std 18.8775 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7507 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0344277 +trainer/policy/mean Std 0.742141 +trainer/policy/mean Max 0.999658 +trainer/policy/mean Min -0.999518 +trainer/policy/std Mean 0.409274 +trainer/policy/std Std 0.0191286 +trainer/policy/std Max 0.429064 +trainer/policy/std Min 0.379904 +trainer/Advantage Weights Mean 1.8616 +trainer/Advantage Weights Std 11.6804 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.7685e-15 +trainer/Advantage Score Mean -0.574069 +trainer/Advantage Score Std 0.623111 +trainer/Advantage Score Max 1.14775 +trainer/Advantage Score Min -3.35205 +trainer/V1 Predictions Mean -73.3382 +trainer/V1 Predictions Std 19.0614 +trainer/V1 Predictions Max 0.414673 +trainer/V1 Predictions Min -87.6277 +trainer/VF Loss 0.0812957 +expl/num steps total 883000 +expl/num paths total 1208 +expl/path length Mean 500 +expl/path length Std 136 +expl/path length Max 636 +expl/path length Min 364 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0286292 +expl/Actions Std 0.838218 +expl/Actions Max 2.24706 +expl/Actions Min -2.52256 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 797306 +eval/num paths total 888 +eval/path length Mean 453 +eval/path length Std 0 +eval/path length Max 453 +eval/path length Min 453 +eval/Rewards Mean 0.00220751 +eval/Rewards Std 0.0469322 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0287629 +eval/Actions Std 0.768138 +eval/Actions Max 0.999847 +eval/Actions Min -0.999452 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.65799e-06 +time/evaluation sampling (s) 2.87868 +time/exploration sampling (s) 4.70378 +time/logging (s) 0.00697804 +time/saving (s) 0.0109318 +time/training (s) 16.3055 +time/epoch (s) 23.9059 +time/total (s) 22868.4 +Epoch -118 +------------------------------ ---------------- +2022-05-16 00:24:03.972077 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -117 finished +------------------------------ ---------------- +epoch -117 +replay_buffer/size 999047 +trainer/num train calls 884000 +trainer/QF1 Loss 4.4075 +trainer/QF2 Loss 4.41322 +trainer/Policy Loss 36.0695 +trainer/Q1 Predictions Mean -75.7818 +trainer/Q1 Predictions Std 17.6384 +trainer/Q1 Predictions Max -2.17789 +trainer/Q1 Predictions Min -87.7223 +trainer/Q2 Predictions Mean -75.7761 +trainer/Q2 Predictions Std 17.6842 +trainer/Q2 Predictions Max -2.08199 +trainer/Q2 Predictions Min -87.6446 +trainer/Q Targets Mean -75.6913 +trainer/Q Targets Std 17.489 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7913 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0104991 +trainer/policy/mean Std 0.729557 +trainer/policy/mean Max 0.999713 +trainer/policy/mean Min -0.999136 +trainer/policy/std Mean 0.408919 +trainer/policy/std Std 0.0188731 +trainer/policy/std Max 0.430425 +trainer/policy/std Min 0.378864 +trainer/Advantage Weights Mean 6.53557 +trainer/Advantage Weights Std 22.3499 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.5889e-14 +trainer/Advantage Score Mean -0.321488 +trainer/Advantage Score Std 0.553489 +trainer/Advantage Score Max 1.82562 +trainer/Advantage Score Min -3.17731 +trainer/V1 Predictions Mean -75.5778 +trainer/V1 Predictions Std 17.6311 +trainer/V1 Predictions Max -1.92294 +trainer/V1 Predictions Min -87.7036 +trainer/VF Loss 0.0789604 +expl/num steps total 884000 +expl/num paths total 1210 +expl/path length Mean 500 +expl/path length Std 99 +expl/path length Max 599 +expl/path length Min 401 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0233385 +expl/Actions Std 0.836449 +expl/Actions Max 2.54624 +expl/Actions Min -2.33049 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 798306 +eval/num paths total 889 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0832571 +eval/Actions Std 0.71773 +eval/Actions Max 0.999876 +eval/Actions Min -0.999816 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.19211e-06 +time/evaluation sampling (s) 3.61533 +time/exploration sampling (s) 4.08952 +time/logging (s) 0.00709466 +time/saving (s) 0.00975818 +time/training (s) 16.5524 +time/epoch (s) 24.2741 +time/total (s) 22892.7 +Epoch -117 +------------------------------ ---------------- +2022-05-16 00:24:27.805701 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -116 finished +------------------------------ ---------------- +epoch -116 +replay_buffer/size 999047 +trainer/num train calls 885000 +trainer/QF1 Loss 0.70663 +trainer/QF2 Loss 0.640618 +trainer/Policy Loss 10.4256 +trainer/Q1 Predictions Mean -74.0136 +trainer/Q1 Predictions Std 17.17 +trainer/Q1 Predictions Max -2.01078 +trainer/Q1 Predictions Min -87.9839 +trainer/Q2 Predictions Mean -74.0456 +trainer/Q2 Predictions Std 17.2614 +trainer/Q2 Predictions Max -1.09965 +trainer/Q2 Predictions Min -87.8128 +trainer/Q Targets Mean -73.6569 +trainer/Q Targets Std 17.2399 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7071 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000146806 +trainer/policy/mean Std 0.734288 +trainer/policy/mean Max 0.999498 +trainer/policy/mean Min -0.997959 +trainer/policy/std Mean 0.408932 +trainer/policy/std Std 0.0190254 +trainer/policy/std Max 0.43207 +trainer/policy/std Min 0.379186 +trainer/Advantage Weights Mean 1.64096 +trainer/Advantage Weights Std 11.8498 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.10432e-16 +trainer/Advantage Score Mean -0.758238 +trainer/Advantage Score Std 0.622286 +trainer/Advantage Score Max 0.910187 +trainer/Advantage Score Min -3.67421 +trainer/V1 Predictions Mean -73.4172 +trainer/V1 Predictions Std 17.2876 +trainer/V1 Predictions Max -1.8292 +trainer/V1 Predictions Min -87.5815 +trainer/VF Loss 0.102331 +expl/num steps total 885000 +expl/num paths total 1211 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0369546 +expl/Actions Std 0.809009 +expl/Actions Max 2.21471 +expl/Actions Min -2.34784 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 799306 +eval/num paths total 890 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.148716 +eval/Actions Std 0.583591 +eval/Actions Max 0.999919 +eval/Actions Min -0.998731 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.13417e-06 +time/evaluation sampling (s) 3.86857 +time/exploration sampling (s) 4.23807 +time/logging (s) 0.00821261 +time/saving (s) 0.0129438 +time/training (s) 15.6944 +time/epoch (s) 23.8222 +time/total (s) 22916.5 +Epoch -116 +------------------------------ ---------------- +2022-05-16 00:24:52.129319 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -115 finished +------------------------------ ---------------- +epoch -115 +replay_buffer/size 999047 +trainer/num train calls 886000 +trainer/QF1 Loss 0.384091 +trainer/QF2 Loss 0.381719 +trainer/Policy Loss 3.89188 +trainer/Q1 Predictions Mean -76.2038 +trainer/Q1 Predictions Std 15.9025 +trainer/Q1 Predictions Max -0.374116 +trainer/Q1 Predictions Min -88.0273 +trainer/Q2 Predictions Mean -76.1866 +trainer/Q2 Predictions Std 15.8967 +trainer/Q2 Predictions Max -0.319807 +trainer/Q2 Predictions Min -87.9795 +trainer/Q Targets Mean -76.1354 +trainer/Q Targets Std 15.8557 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9834 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00329069 +trainer/policy/mean Std 0.724066 +trainer/policy/mean Max 0.999924 +trainer/policy/mean Min -0.998316 +trainer/policy/std Mean 0.407596 +trainer/policy/std Std 0.0185571 +trainer/policy/std Max 0.42989 +trainer/policy/std Min 0.379136 +trainer/Advantage Weights Mean 1.48346 +trainer/Advantage Weights Std 10.4542 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.94388e-18 +trainer/Advantage Score Mean -0.514604 +trainer/Advantage Score Std 0.566717 +trainer/Advantage Score Max 0.984185 +trainer/Advantage Score Min -4.00744 +trainer/V1 Predictions Mean -75.8287 +trainer/V1 Predictions Std 16.0816 +trainer/V1 Predictions Max 1.53591 +trainer/V1 Predictions Min -87.8512 +trainer/VF Loss 0.0657143 +expl/num steps total 886000 +expl/num paths total 1213 +expl/path length Mean 500 +expl/path length Std 317 +expl/path length Max 817 +expl/path length Min 183 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0430103 +expl/Actions Std 0.840777 +expl/Actions Max 2.1959 +expl/Actions Min -2.25824 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 800306 +eval/num paths total 891 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.213927 +eval/Actions Std 0.727323 +eval/Actions Max 0.999514 +eval/Actions Min -0.998835 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.39793e-06 +time/evaluation sampling (s) 3.42261 +time/exploration sampling (s) 4.83127 +time/logging (s) 0.00923053 +time/saving (s) 0.0127853 +time/training (s) 16.0345 +time/epoch (s) 24.3103 +time/total (s) 22940.8 +Epoch -115 +------------------------------ ---------------- +2022-05-16 00:25:16.642923 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -114 finished +------------------------------ ---------------- +epoch -114 +replay_buffer/size 999047 +trainer/num train calls 887000 +trainer/QF1 Loss 0.569003 +trainer/QF2 Loss 0.492407 +trainer/Policy Loss 5.27583 +trainer/Q1 Predictions Mean -73.943 +trainer/Q1 Predictions Std 18.9132 +trainer/Q1 Predictions Max -0.213366 +trainer/Q1 Predictions Min -88.0077 +trainer/Q2 Predictions Mean -73.9568 +trainer/Q2 Predictions Std 19.0298 +trainer/Q2 Predictions Max 0.0684525 +trainer/Q2 Predictions Min -87.9674 +trainer/Q Targets Mean -73.7887 +trainer/Q Targets Std 19.0961 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0078 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0048647 +trainer/policy/mean Std 0.727681 +trainer/policy/mean Max 0.999534 +trainer/policy/mean Min -0.99983 +trainer/policy/std Mean 0.408699 +trainer/policy/std Std 0.0191266 +trainer/policy/std Max 0.429673 +trainer/policy/std Min 0.381193 +trainer/Advantage Weights Mean 1.23493 +trainer/Advantage Weights Std 7.69953 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.67713e-21 +trainer/Advantage Score Mean -0.553498 +trainer/Advantage Score Std 0.704351 +trainer/Advantage Score Max 0.970032 +trainer/Advantage Score Min -4.70522 +trainer/V1 Predictions Mean -73.4637 +trainer/V1 Predictions Std 19.4045 +trainer/V1 Predictions Max 0.589557 +trainer/V1 Predictions Min -87.8867 +trainer/VF Loss 0.0854147 +expl/num steps total 887000 +expl/num paths total 1215 +expl/path length Mean 500 +expl/path length Std 307 +expl/path length Max 807 +expl/path length Min 193 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0104071 +expl/Actions Std 0.825536 +expl/Actions Max 2.19715 +expl/Actions Min -2.05308 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 800733 +eval/num paths total 892 +eval/path length Mean 427 +eval/path length Std 0 +eval/path length Max 427 +eval/path length Min 427 +eval/Rewards Mean 0.00234192 +eval/Rewards Std 0.0483367 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0186859 +eval/Actions Std 0.753226 +eval/Actions Max 0.999757 +eval/Actions Min -0.999675 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.82006e-06 +time/evaluation sampling (s) 3.57243 +time/exploration sampling (s) 4.32145 +time/logging (s) 0.00612148 +time/saving (s) 0.0106725 +time/training (s) 16.5858 +time/epoch (s) 24.4965 +time/total (s) 22965.3 +Epoch -114 +------------------------------ ---------------- +2022-05-16 00:25:41.333519 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -113 finished +------------------------------ ---------------- +epoch -113 +replay_buffer/size 999047 +trainer/num train calls 888000 +trainer/QF1 Loss 6.59176 +trainer/QF2 Loss 6.83902 +trainer/Policy Loss 11.1439 +trainer/Q1 Predictions Mean -76.1588 +trainer/Q1 Predictions Std 15.3488 +trainer/Q1 Predictions Max -2.16973 +trainer/Q1 Predictions Min -87.9241 +trainer/Q2 Predictions Mean -76.1266 +trainer/Q2 Predictions Std 15.4028 +trainer/Q2 Predictions Max -1.57412 +trainer/Q2 Predictions Min -87.9141 +trainer/Q Targets Mean -76.3337 +trainer/Q Targets Std 15.1589 +trainer/Q Targets Max -1.05907 +trainer/Q Targets Min -87.8174 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0369331 +trainer/policy/mean Std 0.731141 +trainer/policy/mean Max 0.999386 +trainer/policy/mean Min -0.998735 +trainer/policy/std Mean 0.41026 +trainer/policy/std Std 0.0196579 +trainer/policy/std Max 0.432883 +trainer/policy/std Min 0.381491 +trainer/Advantage Weights Mean 2.96396 +trainer/Advantage Weights Std 14.3862 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.49525e-15 +trainer/Advantage Score Mean -0.407362 +trainer/Advantage Score Std 0.599383 +trainer/Advantage Score Max 3.07225 +trainer/Advantage Score Min -3.30358 +trainer/V1 Predictions Mean -75.8898 +trainer/V1 Predictions Std 15.5889 +trainer/V1 Predictions Max -1.32483 +trainer/V1 Predictions Min -87.6957 +trainer/VF Loss 0.0885525 +expl/num steps total 888000 +expl/num paths total 1216 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0480704 +expl/Actions Std 0.865031 +expl/Actions Max 2.31437 +expl/Actions Min -2.4464 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 801274 +eval/num paths total 893 +eval/path length Mean 541 +eval/path length Std 0 +eval/path length Max 541 +eval/path length Min 541 +eval/Rewards Mean 0.00184843 +eval/Rewards Std 0.0429536 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0464419 +eval/Actions Std 0.752393 +eval/Actions Max 0.999878 +eval/Actions Min -0.999518 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.52575e-06 +time/evaluation sampling (s) 3.46698 +time/exploration sampling (s) 4.92843 +time/logging (s) 0.00625614 +time/saving (s) 0.010435 +time/training (s) 16.2671 +time/epoch (s) 24.6792 +time/total (s) 22990 +Epoch -113 +------------------------------ ---------------- +2022-05-16 00:26:06.171513 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -112 finished +------------------------------ ---------------- +epoch -112 +replay_buffer/size 999047 +trainer/num train calls 889000 +trainer/QF1 Loss 0.913111 +trainer/QF2 Loss 0.760844 +trainer/Policy Loss 4.45001 +trainer/Q1 Predictions Mean -73.9433 +trainer/Q1 Predictions Std 19.0152 +trainer/Q1 Predictions Max -0.526522 +trainer/Q1 Predictions Min -88.3996 +trainer/Q2 Predictions Mean -73.9372 +trainer/Q2 Predictions Std 18.9505 +trainer/Q2 Predictions Max 0.363712 +trainer/Q2 Predictions Min -88.2716 +trainer/Q Targets Mean -73.9389 +trainer/Q Targets Std 19.1118 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1551 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0247897 +trainer/policy/mean Std 0.736506 +trainer/policy/mean Max 0.999334 +trainer/policy/mean Min -0.999017 +trainer/policy/std Mean 0.409375 +trainer/policy/std Std 0.0202604 +trainer/policy/std Max 0.430434 +trainer/policy/std Min 0.379526 +trainer/Advantage Weights Mean 2.00681 +trainer/Advantage Weights Std 11.7295 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.87139e-19 +trainer/Advantage Score Mean -0.565329 +trainer/Advantage Score Std 0.635781 +trainer/Advantage Score Max 0.632778 +trainer/Advantage Score Min -4.21657 +trainer/V1 Predictions Mean -73.6504 +trainer/V1 Predictions Std 19.208 +trainer/V1 Predictions Max 0.0476872 +trainer/V1 Predictions Min -88.0038 +trainer/VF Loss 0.077965 +expl/num steps total 889000 +expl/num paths total 1218 +expl/path length Mean 500 +expl/path length Std 202 +expl/path length Max 702 +expl/path length Min 298 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0169999 +expl/Actions Std 0.836852 +expl/Actions Max 2.75277 +expl/Actions Min -2.51774 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 802274 +eval/num paths total 894 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.398903 +eval/Actions Std 0.656584 +eval/Actions Max 0.999904 +eval/Actions Min -0.999189 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.07616e-06 +time/evaluation sampling (s) 3.0209 +time/exploration sampling (s) 5.69258 +time/logging (s) 0.0105037 +time/saving (s) 0.0162338 +time/training (s) 16.0903 +time/epoch (s) 24.8306 +time/total (s) 23014.8 +Epoch -112 +------------------------------ ---------------- +2022-05-16 00:26:29.523684 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -111 finished +------------------------------ ---------------- +epoch -111 +replay_buffer/size 999047 +trainer/num train calls 890000 +trainer/QF1 Loss 0.594929 +trainer/QF2 Loss 0.558354 +trainer/Policy Loss 9.00923 +trainer/Q1 Predictions Mean -75.9735 +trainer/Q1 Predictions Std 15.1649 +trainer/Q1 Predictions Max -0.492241 +trainer/Q1 Predictions Min -87.3956 +trainer/Q2 Predictions Mean -75.9269 +trainer/Q2 Predictions Std 15.1432 +trainer/Q2 Predictions Max 0.536437 +trainer/Q2 Predictions Min -87.3366 +trainer/Q Targets Mean -75.9887 +trainer/Q Targets Std 15.1673 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7466 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0174469 +trainer/policy/mean Std 0.730197 +trainer/policy/mean Max 0.999878 +trainer/policy/mean Min -0.999664 +trainer/policy/std Mean 0.409546 +trainer/policy/std Std 0.0199822 +trainer/policy/std Max 0.429887 +trainer/policy/std Min 0.379213 +trainer/Advantage Weights Mean 2.41704 +trainer/Advantage Weights Std 13.9863 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.54831e-16 +trainer/Advantage Score Mean -0.475023 +trainer/Advantage Score Std 0.569637 +trainer/Advantage Score Max 1.90606 +trainer/Advantage Score Min -3.51279 +trainer/V1 Predictions Mean -75.7328 +trainer/V1 Predictions Std 15.3339 +trainer/V1 Predictions Max 0.105561 +trainer/V1 Predictions Min -87.6074 +trainer/VF Loss 0.0725264 +expl/num steps total 890000 +expl/num paths total 1219 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0395063 +expl/Actions Std 0.855949 +expl/Actions Max 2.52397 +expl/Actions Min -2.18211 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 803274 +eval/num paths total 895 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.302182 +eval/Actions Std 0.765759 +eval/Actions Max 0.999882 +eval/Actions Min -0.99976 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.92622e-06 +time/evaluation sampling (s) 3.24804 +time/exploration sampling (s) 4.16119 +time/logging (s) 0.0081319 +time/saving (s) 0.0126661 +time/training (s) 15.9088 +time/epoch (s) 23.3388 +time/total (s) 23038.2 +Epoch -111 +------------------------------ ---------------- +2022-05-16 00:26:53.918450 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -110 finished +------------------------------ ---------------- +epoch -110 +replay_buffer/size 999047 +trainer/num train calls 891000 +trainer/QF1 Loss 0.634678 +trainer/QF2 Loss 0.684623 +trainer/Policy Loss 22.6734 +trainer/Q1 Predictions Mean -75.7029 +trainer/Q1 Predictions Std 16.7309 +trainer/Q1 Predictions Max -0.405843 +trainer/Q1 Predictions Min -88.3108 +trainer/Q2 Predictions Mean -75.54 +trainer/Q2 Predictions Std 17.0172 +trainer/Q2 Predictions Max 0.3128 +trainer/Q2 Predictions Min -88.2367 +trainer/Q Targets Mean -75.8649 +trainer/Q Targets Std 16.7562 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.4734 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0331149 +trainer/policy/mean Std 0.735997 +trainer/policy/mean Max 0.999159 +trainer/policy/mean Min -0.998609 +trainer/policy/std Mean 0.408763 +trainer/policy/std Std 0.019892 +trainer/policy/std Max 0.430955 +trainer/policy/std Min 0.37743 +trainer/Advantage Weights Mean 6.45173 +trainer/Advantage Weights Std 21.9647 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.41491e-11 +trainer/Advantage Score Mean -0.235587 +trainer/Advantage Score Std 0.521011 +trainer/Advantage Score Max 2.25777 +trainer/Advantage Score Min -2.34698 +trainer/V1 Predictions Mean -75.6367 +trainer/V1 Predictions Std 16.7351 +trainer/V1 Predictions Max -0.474178 +trainer/V1 Predictions Min -88.2159 +trainer/VF Loss 0.0955879 +expl/num steps total 891000 +expl/num paths total 1221 +expl/path length Mean 500 +expl/path length Std 8 +expl/path length Max 508 +expl/path length Min 492 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0135991 +expl/Actions Std 0.838151 +expl/Actions Max 2.56231 +expl/Actions Min -2.74207 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 804274 +eval/num paths total 896 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00242368 +eval/Actions Std 0.715183 +eval/Actions Max 0.999979 +eval/Actions Min -0.999827 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.83588e-06 +time/evaluation sampling (s) 3.77128 +time/exploration sampling (s) 4.47655 +time/logging (s) 0.0128902 +time/saving (s) 0.0185961 +time/training (s) 16.1057 +time/epoch (s) 24.3851 +time/total (s) 23062.6 +Epoch -110 +------------------------------ ---------------- +2022-05-16 00:27:18.138624 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -109 finished +------------------------------ ---------------- +epoch -109 +replay_buffer/size 999047 +trainer/num train calls 892000 +trainer/QF1 Loss 1.33213 +trainer/QF2 Loss 1.26186 +trainer/Policy Loss 18.3268 +trainer/Q1 Predictions Mean -73.4905 +trainer/Q1 Predictions Std 19.0666 +trainer/Q1 Predictions Max -0.951202 +trainer/Q1 Predictions Min -87.4141 +trainer/Q2 Predictions Mean -73.4813 +trainer/Q2 Predictions Std 19.0521 +trainer/Q2 Predictions Max 0.299817 +trainer/Q2 Predictions Min -87.7669 +trainer/Q Targets Mean -73.8847 +trainer/Q Targets Std 18.8523 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7472 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0067468 +trainer/policy/mean Std 0.732816 +trainer/policy/mean Max 0.999873 +trainer/policy/mean Min -0.998863 +trainer/policy/std Mean 0.408062 +trainer/policy/std Std 0.0180001 +trainer/policy/std Max 0.428824 +trainer/policy/std Min 0.381603 +trainer/Advantage Weights Mean 5.10122 +trainer/Advantage Weights Std 17.4017 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.45168e-12 +trainer/Advantage Score Mean -0.248863 +trainer/Advantage Score Std 0.493974 +trainer/Advantage Score Max 1.23281 +trainer/Advantage Score Min -2.59351 +trainer/V1 Predictions Mean -73.6223 +trainer/V1 Predictions Std 19.0715 +trainer/V1 Predictions Max -0.469039 +trainer/V1 Predictions Min -87.7752 +trainer/VF Loss 0.0538482 +expl/num steps total 892000 +expl/num paths total 1222 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.046803 +expl/Actions Std 0.818973 +expl/Actions Max 2.45695 +expl/Actions Min -2.39194 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 805274 +eval/num paths total 897 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.179886 +eval/Actions Std 0.704853 +eval/Actions Max 0.99995 +eval/Actions Min -0.999552 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.01351e-05 +time/evaluation sampling (s) 3.27605 +time/exploration sampling (s) 4.461 +time/logging (s) 0.0076179 +time/saving (s) 0.0105601 +time/training (s) 16.4371 +time/epoch (s) 24.1923 +time/total (s) 23086.8 +Epoch -109 +------------------------------ ---------------- +2022-05-16 00:27:42.461411 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -108 finished +------------------------------ ---------------- +epoch -108 +replay_buffer/size 999047 +trainer/num train calls 893000 +trainer/QF1 Loss 1.60751 +trainer/QF2 Loss 1.62517 +trainer/Policy Loss 5.82225 +trainer/Q1 Predictions Mean -74.5079 +trainer/Q1 Predictions Std 17.9609 +trainer/Q1 Predictions Max -7.30505 +trainer/Q1 Predictions Min -88.2261 +trainer/Q2 Predictions Mean -74.5172 +trainer/Q2 Predictions Std 17.9675 +trainer/Q2 Predictions Max -7.37777 +trainer/Q2 Predictions Min -88.3836 +trainer/Q Targets Mean -74.0751 +trainer/Q Targets Std 18.2063 +trainer/Q Targets Max -5.54483 +trainer/Q Targets Min -87.8134 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00766307 +trainer/policy/mean Std 0.729465 +trainer/policy/mean Max 0.998765 +trainer/policy/mean Min -0.998087 +trainer/policy/std Mean 0.410283 +trainer/policy/std Std 0.0187484 +trainer/policy/std Max 0.42921 +trainer/policy/std Min 0.382934 +trainer/Advantage Weights Mean 2.27559 +trainer/Advantage Weights Std 13.5154 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.59366e-24 +trainer/Advantage Score Mean -0.499746 +trainer/Advantage Score Std 0.702016 +trainer/Advantage Score Max 2.81573 +trainer/Advantage Score Min -5.4796 +trainer/V1 Predictions Mean -73.7348 +trainer/V1 Predictions Std 18.5377 +trainer/V1 Predictions Max -4.63077 +trainer/V1 Predictions Min -87.7113 +trainer/VF Loss 0.104737 +expl/num steps total 893000 +expl/num paths total 1224 +expl/path length Mean 500 +expl/path length Std 325 +expl/path length Max 825 +expl/path length Min 175 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0187703 +expl/Actions Std 0.836376 +expl/Actions Max 2.41465 +expl/Actions Min -2.29158 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 805803 +eval/num paths total 898 +eval/path length Mean 529 +eval/path length Std 0 +eval/path length Max 529 +eval/path length Min 529 +eval/Rewards Mean 0.00189036 +eval/Rewards Std 0.0434371 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0170605 +eval/Actions Std 0.723118 +eval/Actions Max 0.999976 +eval/Actions Min -0.999661 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.19676e-06 +time/evaluation sampling (s) 3.33949 +time/exploration sampling (s) 4.64848 +time/logging (s) 0.00926478 +time/saving (s) 0.0146872 +time/training (s) 16.2995 +time/epoch (s) 24.3114 +time/total (s) 23111.1 +Epoch -108 +------------------------------ ---------------- +2022-05-16 00:28:06.670104 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -107 finished +------------------------------ ---------------- +epoch -107 +replay_buffer/size 999047 +trainer/num train calls 894000 +trainer/QF1 Loss 0.828876 +trainer/QF2 Loss 0.879811 +trainer/Policy Loss 10.3504 +trainer/Q1 Predictions Mean -74.238 +trainer/Q1 Predictions Std 18.2988 +trainer/Q1 Predictions Max -2.20051 +trainer/Q1 Predictions Min -87.6001 +trainer/Q2 Predictions Mean -74.1508 +trainer/Q2 Predictions Std 18.2909 +trainer/Q2 Predictions Max -1.47718 +trainer/Q2 Predictions Min -87.6718 +trainer/Q Targets Mean -74.1846 +trainer/Q Targets Std 18.4628 +trainer/Q Targets Max 0.231958 +trainer/Q Targets Min -87.4005 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0139247 +trainer/policy/mean Std 0.722005 +trainer/policy/mean Max 0.999563 +trainer/policy/mean Min -0.998111 +trainer/policy/std Mean 0.40836 +trainer/policy/std Std 0.0186666 +trainer/policy/std Max 0.426815 +trainer/policy/std Min 0.382427 +trainer/Advantage Weights Mean 3.34644 +trainer/Advantage Weights Std 14.1921 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.33785e-29 +trainer/Advantage Score Mean -0.390126 +trainer/Advantage Score Std 0.681435 +trainer/Advantage Score Max 0.835628 +trainer/Advantage Score Min -6.64839 +trainer/V1 Predictions Mean -73.9018 +trainer/V1 Predictions Std 18.6065 +trainer/V1 Predictions Max -0.922248 +trainer/V1 Predictions Min -87.3235 +trainer/VF Loss 0.0716985 +expl/num steps total 894000 +expl/num paths total 1226 +expl/path length Mean 500 +expl/path length Std 258 +expl/path length Max 758 +expl/path length Min 242 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0332058 +expl/Actions Std 0.839393 +expl/Actions Max 2.33935 +expl/Actions Min -2.6652 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 806341 +eval/num paths total 899 +eval/path length Mean 538 +eval/path length Std 0 +eval/path length Max 538 +eval/path length Min 538 +eval/Rewards Mean 0.00185874 +eval/Rewards Std 0.043073 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0141091 +eval/Actions Std 0.730463 +eval/Actions Max 0.999887 +eval/Actions Min -0.999647 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.32482e-06 +time/evaluation sampling (s) 3.60975 +time/exploration sampling (s) 4.82964 +time/logging (s) 0.00865215 +time/saving (s) 0.0144643 +time/training (s) 15.7271 +time/epoch (s) 24.1896 +time/total (s) 23135.3 +Epoch -107 +------------------------------ ---------------- +2022-05-16 00:28:30.388702 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -106 finished +------------------------------ ---------------- +epoch -106 +replay_buffer/size 999047 +trainer/num train calls 895000 +trainer/QF1 Loss 5.30978 +trainer/QF2 Loss 5.23623 +trainer/Policy Loss 32.6982 +trainer/Q1 Predictions Mean -73.8037 +trainer/Q1 Predictions Std 18.7475 +trainer/Q1 Predictions Max -0.425884 +trainer/Q1 Predictions Min -87.5438 +trainer/Q2 Predictions Mean -73.8139 +trainer/Q2 Predictions Std 18.7451 +trainer/Q2 Predictions Max -0.15343 +trainer/Q2 Predictions Min -87.3422 +trainer/Q Targets Mean -74.0872 +trainer/Q Targets Std 18.6327 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6617 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000469654 +trainer/policy/mean Std 0.724003 +trainer/policy/mean Max 0.999398 +trainer/policy/mean Min -0.999211 +trainer/policy/std Mean 0.407956 +trainer/policy/std Std 0.0194885 +trainer/policy/std Max 0.429423 +trainer/policy/std Min 0.380866 +trainer/Advantage Weights Mean 6.83964 +trainer/Advantage Weights Std 21.2477 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.36028e-18 +trainer/Advantage Score Mean -0.284527 +trainer/Advantage Score Std 0.669493 +trainer/Advantage Score Max 0.760033 +trainer/Advantage Score Min -4.11388 +trainer/V1 Predictions Mean -73.6606 +trainer/V1 Predictions Std 19.1223 +trainer/V1 Predictions Max 0.848648 +trainer/V1 Predictions Min -87.3904 +trainer/VF Loss 0.0720824 +expl/num steps total 895000 +expl/num paths total 1228 +expl/path length Mean 500 +expl/path length Std 202 +expl/path length Max 702 +expl/path length Min 298 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.012212 +expl/Actions Std 0.836498 +expl/Actions Max 2.29648 +expl/Actions Min -2.64655 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 807193 +eval/num paths total 900 +eval/path length Mean 852 +eval/path length Std 0 +eval/path length Max 852 +eval/path length Min 852 +eval/Rewards Mean 0.00117371 +eval/Rewards Std 0.0342393 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0155392 +eval/Actions Std 0.761176 +eval/Actions Max 0.999771 +eval/Actions Min -0.999497 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.76696e-06 +time/evaluation sampling (s) 3.34939 +time/exploration sampling (s) 4.29446 +time/logging (s) 0.00746093 +time/saving (s) 0.0130959 +time/training (s) 16.0343 +time/epoch (s) 23.6987 +time/total (s) 23159 +Epoch -106 +------------------------------ ---------------- +2022-05-16 00:28:54.068056 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -105 finished +------------------------------ ---------------- +epoch -105 +replay_buffer/size 999047 +trainer/num train calls 896000 +trainer/QF1 Loss 0.598702 +trainer/QF2 Loss 0.474275 +trainer/Policy Loss 10.2045 +trainer/Q1 Predictions Mean -74.8862 +trainer/Q1 Predictions Std 18.0596 +trainer/Q1 Predictions Max -0.322059 +trainer/Q1 Predictions Min -87.7249 +trainer/Q2 Predictions Mean -74.9136 +trainer/Q2 Predictions Std 18.0208 +trainer/Q2 Predictions Max 0.276815 +trainer/Q2 Predictions Min -87.8648 +trainer/Q Targets Mean -75.2313 +trainer/Q Targets Std 17.9417 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.036 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00876243 +trainer/policy/mean Std 0.743282 +trainer/policy/mean Max 0.999454 +trainer/policy/mean Min -0.999771 +trainer/policy/std Mean 0.407469 +trainer/policy/std Std 0.0193033 +trainer/policy/std Max 0.427322 +trainer/policy/std Min 0.380287 +trainer/Advantage Weights Mean 3.67649 +trainer/Advantage Weights Std 15.4807 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.57813e-14 +trainer/Advantage Score Mean -0.327384 +trainer/Advantage Score Std 0.463276 +trainer/Advantage Score Max 1.55251 +trainer/Advantage Score Min -3.0087 +trainer/V1 Predictions Mean -75.0525 +trainer/V1 Predictions Std 18.0117 +trainer/V1 Predictions Max -0.0154197 +trainer/V1 Predictions Min -87.797 +trainer/VF Loss 0.0494601 +expl/num steps total 896000 +expl/num paths total 1229 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.379524 +expl/Actions Std 0.844105 +expl/Actions Max 2.42803 +expl/Actions Min -2.25724 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 808053 +eval/num paths total 901 +eval/path length Mean 860 +eval/path length Std 0 +eval/path length Max 860 +eval/path length Min 860 +eval/Rewards Mean 0.00116279 +eval/Rewards Std 0.0340799 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0177503 +eval/Actions Std 0.731577 +eval/Actions Max 0.999952 +eval/Actions Min -0.999881 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.82098e-06 +time/evaluation sampling (s) 3.66375 +time/exploration sampling (s) 4.18145 +time/logging (s) 0.00673038 +time/saving (s) 0.00998617 +time/training (s) 15.8039 +time/epoch (s) 23.6658 +time/total (s) 23182.6 +Epoch -105 +------------------------------ ---------------- +2022-05-16 00:29:18.676821 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -104 finished +------------------------------ ---------------- +epoch -104 +replay_buffer/size 999047 +trainer/num train calls 897000 +trainer/QF1 Loss 2.66232 +trainer/QF2 Loss 2.85356 +trainer/Policy Loss 32.0564 +trainer/Q1 Predictions Mean -75.205 +trainer/Q1 Predictions Std 16.5214 +trainer/Q1 Predictions Max -1.96431 +trainer/Q1 Predictions Min -86.748 +trainer/Q2 Predictions Mean -75.1862 +trainer/Q2 Predictions Std 16.5612 +trainer/Q2 Predictions Max -2.02435 +trainer/Q2 Predictions Min -86.9522 +trainer/Q Targets Mean -75.2587 +trainer/Q Targets Std 16.5579 +trainer/Q Targets Max 0 +trainer/Q Targets Min -86.9982 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0134129 +trainer/policy/mean Std 0.730314 +trainer/policy/mean Max 0.999253 +trainer/policy/mean Min -0.999536 +trainer/policy/std Mean 0.407041 +trainer/policy/std Std 0.0195323 +trainer/policy/std Max 0.425908 +trainer/policy/std Min 0.379082 +trainer/Advantage Weights Mean 6.57561 +trainer/Advantage Weights Std 19.5477 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.87937e-16 +trainer/Advantage Score Mean -0.277561 +trainer/Advantage Score Std 0.650987 +trainer/Advantage Score Max 1.60213 +trainer/Advantage Score Min -3.49128 +trainer/V1 Predictions Mean -74.8801 +trainer/V1 Predictions Std 16.8679 +trainer/V1 Predictions Max -0.191115 +trainer/V1 Predictions Min -86.926 +trainer/VF Loss 0.0814262 +expl/num steps total 897000 +expl/num paths total 1231 +expl/path length Mean 500 +expl/path length Std 350 +expl/path length Max 850 +expl/path length Min 150 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0387054 +expl/Actions Std 0.825916 +expl/Actions Max 2.33354 +expl/Actions Min -2.09758 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 808733 +eval/num paths total 902 +eval/path length Mean 680 +eval/path length Std 0 +eval/path length Max 680 +eval/path length Min 680 +eval/Rewards Mean 0.00147059 +eval/Rewards Std 0.03832 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.0109337 +eval/Actions Std 0.73069 +eval/Actions Max 0.999788 +eval/Actions Min -0.999603 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.01003e-06 +time/evaluation sampling (s) 3.71047 +time/exploration sampling (s) 4.96258 +time/logging (s) 0.00746962 +time/saving (s) 0.0172937 +time/training (s) 15.8984 +time/epoch (s) 24.5962 +time/total (s) 23207.2 +Epoch -104 +------------------------------ ---------------- +2022-05-16 00:29:42.218776 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -103 finished +------------------------------ ---------------- +epoch -103 +replay_buffer/size 999047 +trainer/num train calls 898000 +trainer/QF1 Loss 0.489435 +trainer/QF2 Loss 0.513471 +trainer/Policy Loss 5.33891 +trainer/Q1 Predictions Mean -74.6449 +trainer/Q1 Predictions Std 17.3085 +trainer/Q1 Predictions Max -1.28494 +trainer/Q1 Predictions Min -87.5322 +trainer/Q2 Predictions Mean -74.6625 +trainer/Q2 Predictions Std 17.3049 +trainer/Q2 Predictions Max -1.25759 +trainer/Q2 Predictions Min -87.5861 +trainer/Q Targets Mean -74.5179 +trainer/Q Targets Std 17.3519 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6186 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00483506 +trainer/policy/mean Std 0.732411 +trainer/policy/mean Max 0.999943 +trainer/policy/mean Min -0.99867 +trainer/policy/std Mean 0.407907 +trainer/policy/std Std 0.0188398 +trainer/policy/std Max 0.428269 +trainer/policy/std Min 0.378583 +trainer/Advantage Weights Mean 1.25079 +trainer/Advantage Weights Std 8.19516 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.52704e-20 +trainer/Advantage Score Mean -0.540102 +trainer/Advantage Score Std 0.634699 +trainer/Advantage Score Max 0.744759 +trainer/Advantage Score Min -4.45416 +trainer/V1 Predictions Mean -74.2197 +trainer/V1 Predictions Std 17.6057 +trainer/V1 Predictions Max -0.66349 +trainer/V1 Predictions Min -87.4934 +trainer/VF Loss 0.0733976 +expl/num steps total 898000 +expl/num paths total 1233 +expl/path length Mean 500 +expl/path length Std 368 +expl/path length Max 868 +expl/path length Min 132 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.0693611 +expl/Actions Std 0.855651 +expl/Actions Max 2.32504 +expl/Actions Min -2.32422 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 809494 +eval/num paths total 903 +eval/path length Mean 761 +eval/path length Std 0 +eval/path length Max 761 +eval/path length Min 761 +eval/Rewards Mean 0.00131406 +eval/Rewards Std 0.0362261 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00226035 +eval/Actions Std 0.749094 +eval/Actions Max 0.999984 +eval/Actions Min -0.999904 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.7041e-06 +time/evaluation sampling (s) 3.04246 +time/exploration sampling (s) 4.57172 +time/logging (s) 0.00648599 +time/saving (s) 0.00965114 +time/training (s) 15.8964 +time/epoch (s) 23.5267 +time/total (s) 23230.8 +Epoch -103 +------------------------------ ---------------- +2022-05-16 00:30:05.572062 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -102 finished +------------------------------ ---------------- +epoch -102 +replay_buffer/size 999047 +trainer/num train calls 899000 +trainer/QF1 Loss 0.496549 +trainer/QF2 Loss 0.503959 +trainer/Policy Loss 10.9852 +trainer/Q1 Predictions Mean -75.0297 +trainer/Q1 Predictions Std 16.9902 +trainer/Q1 Predictions Max -0.668825 +trainer/Q1 Predictions Min -87.5668 +trainer/Q2 Predictions Mean -75.0067 +trainer/Q2 Predictions Std 16.9685 +trainer/Q2 Predictions Max -0.859875 +trainer/Q2 Predictions Min -87.5584 +trainer/Q Targets Mean -74.7648 +trainer/Q Targets Std 16.7944 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.0429 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0183485 +trainer/policy/mean Std 0.736844 +trainer/policy/mean Max 0.999074 +trainer/policy/mean Min -0.998761 +trainer/policy/std Mean 0.409212 +trainer/policy/std Std 0.0186358 +trainer/policy/std Max 0.432611 +trainer/policy/std Min 0.381026 +trainer/Advantage Weights Mean 3.3371 +trainer/Advantage Weights Std 16.5421 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.28756e-14 +trainer/Advantage Score Mean -0.509744 +trainer/Advantage Score Std 0.531033 +trainer/Advantage Score Max 1.02915 +trainer/Advantage Score Min -3.1046 +trainer/V1 Predictions Mean -74.4897 +trainer/V1 Predictions Std 17.0337 +trainer/V1 Predictions Max -0.285662 +trainer/V1 Predictions Min -86.9524 +trainer/VF Loss 0.0667895 +expl/num steps total 899000 +expl/num paths total 1234 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.015275 +expl/Actions Std 0.824873 +expl/Actions Max 2.15826 +expl/Actions Min -2.25943 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 810438 +eval/num paths total 904 +eval/path length Mean 944 +eval/path length Std 0 +eval/path length Max 944 +eval/path length Min 944 +eval/Rewards Mean 0.00105932 +eval/Rewards Std 0.03253 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0243545 +eval/Actions Std 0.758 +eval/Actions Max 0.999877 +eval/Actions Min -0.999702 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.61795e-06 +time/evaluation sampling (s) 3.41417 +time/exploration sampling (s) 4.03439 +time/logging (s) 0.00727462 +time/saving (s) 0.0121217 +time/training (s) 15.8734 +time/epoch (s) 23.3413 +time/total (s) 23254.1 +Epoch -102 +------------------------------ ---------------- +2022-05-16 00:30:29.774993 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -101 finished +------------------------------ ---------------- +epoch -101 +replay_buffer/size 999047 +trainer/num train calls 900000 +trainer/QF1 Loss 0.590194 +trainer/QF2 Loss 0.615941 +trainer/Policy Loss 20.3308 +trainer/Q1 Predictions Mean -73.1383 +trainer/Q1 Predictions Std 18.808 +trainer/Q1 Predictions Max -0.0428693 +trainer/Q1 Predictions Min -87.2601 +trainer/Q2 Predictions Mean -73.1985 +trainer/Q2 Predictions Std 18.6762 +trainer/Q2 Predictions Max 0.114568 +trainer/Q2 Predictions Min -87.1691 +trainer/Q Targets Mean -73.4221 +trainer/Q Targets Std 18.9819 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.511 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0125202 +trainer/policy/mean Std 0.741626 +trainer/policy/mean Max 0.999571 +trainer/policy/mean Min -0.997439 +trainer/policy/std Mean 0.40948 +trainer/policy/std Std 0.0189441 +trainer/policy/std Max 0.435965 +trainer/policy/std Min 0.380803 +trainer/Advantage Weights Mean 4.89173 +trainer/Advantage Weights Std 17.0422 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.61523e-13 +trainer/Advantage Score Mean -0.330239 +trainer/Advantage Score Std 0.550731 +trainer/Advantage Score Max 1.81785 +trainer/Advantage Score Min -2.77801 +trainer/V1 Predictions Mean -73.2102 +trainer/V1 Predictions Std 19.0075 +trainer/V1 Predictions Max 0.61 +trainer/V1 Predictions Min -87.5861 +trainer/VF Loss 0.0690822 +expl/num steps total 900000 +expl/num paths total 1235 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.208299 +expl/Actions Std 0.886485 +expl/Actions Max 2.41297 +expl/Actions Min -2.42086 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 811092 +eval/num paths total 905 +eval/path length Mean 654 +eval/path length Std 0 +eval/path length Max 654 +eval/path length Min 654 +eval/Rewards Mean 0.00152905 +eval/Rewards Std 0.0390732 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0557581 +eval/Actions Std 0.750792 +eval/Actions Max 0.99976 +eval/Actions Min -0.999604 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.56998e-06 +time/evaluation sampling (s) 3.34447 +time/exploration sampling (s) 5.20441 +time/logging (s) 0.00622291 +time/saving (s) 0.0110213 +time/training (s) 15.6213 +time/epoch (s) 24.1874 +time/total (s) 23278.3 +Epoch -101 +------------------------------ ---------------- +2022-05-16 00:30:53.786109 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -100 finished +------------------------------ ---------------- +epoch -100 +replay_buffer/size 999047 +trainer/num train calls 901000 +trainer/QF1 Loss 0.716718 +trainer/QF2 Loss 0.7804 +trainer/Policy Loss 49.2176 +trainer/Q1 Predictions Mean -75.7201 +trainer/Q1 Predictions Std 16.3779 +trainer/Q1 Predictions Max -0.967345 +trainer/Q1 Predictions Min -87.4265 +trainer/Q2 Predictions Mean -75.6482 +trainer/Q2 Predictions Std 16.3504 +trainer/Q2 Predictions Max -1.47026 +trainer/Q2 Predictions Min -87.2432 +trainer/Q Targets Mean -76.242 +trainer/Q Targets Std 16.3696 +trainer/Q Targets Max -2.23665 +trainer/Q Targets Min -88.0281 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0061791 +trainer/policy/mean Std 0.726046 +trainer/policy/mean Max 0.999358 +trainer/policy/mean Min -0.999702 +trainer/policy/std Mean 0.40774 +trainer/policy/std Std 0.0181057 +trainer/policy/std Max 0.432048 +trainer/policy/std Min 0.379531 +trainer/Advantage Weights Mean 12.8674 +trainer/Advantage Weights Std 27.9195 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.57689e-10 +trainer/Advantage Score Mean -0.0784854 +trainer/Advantage Score Std 0.485098 +trainer/Advantage Score Max 1.0952 +trainer/Advantage Score Min -2.08768 +trainer/V1 Predictions Mean -75.9931 +trainer/V1 Predictions Std 16.5626 +trainer/V1 Predictions Max -1.41847 +trainer/V1 Predictions Min -87.8576 +trainer/VF Loss 0.0676296 +expl/num steps total 901000 +expl/num paths total 1237 +expl/path length Mean 500 +expl/path length Std 387 +expl/path length Max 887 +expl/path length Min 113 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.036328 +expl/Actions Std 0.849911 +expl/Actions Max 2.28023 +expl/Actions Min -2.28392 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 811637 +eval/num paths total 906 +eval/path length Mean 545 +eval/path length Std 0 +eval/path length Max 545 +eval/path length Min 545 +eval/Rewards Mean 0.00183486 +eval/Rewards Std 0.042796 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0175261 +eval/Actions Std 0.737518 +eval/Actions Max 0.999835 +eval/Actions Min -0.999625 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.16044e-06 +time/evaluation sampling (s) 3.03514 +time/exploration sampling (s) 4.74601 +time/logging (s) 0.00589991 +time/saving (s) 0.0179802 +time/training (s) 16.1949 +time/epoch (s) 23.9999 +time/total (s) 23302.3 +Epoch -100 +------------------------------ ---------------- +2022-05-16 00:31:17.601568 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -99 finished +------------------------------ ---------------- +epoch -99 +replay_buffer/size 999047 +trainer/num train calls 902000 +trainer/QF1 Loss 0.698185 +trainer/QF2 Loss 0.663871 +trainer/Policy Loss 27.7644 +trainer/Q1 Predictions Mean -73.0563 +trainer/Q1 Predictions Std 21.4234 +trainer/Q1 Predictions Max -1.44309 +trainer/Q1 Predictions Min -87.6638 +trainer/Q2 Predictions Mean -73.0709 +trainer/Q2 Predictions Std 21.3534 +trainer/Q2 Predictions Max -1.10486 +trainer/Q2 Predictions Min -87.5227 +trainer/Q Targets Mean -73.1098 +trainer/Q Targets Std 21.5323 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6967 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0410697 +trainer/policy/mean Std 0.724417 +trainer/policy/mean Max 0.999228 +trainer/policy/mean Min -0.998344 +trainer/policy/std Mean 0.409098 +trainer/policy/std Std 0.0190885 +trainer/policy/std Max 0.432794 +trainer/policy/std Min 0.379401 +trainer/Advantage Weights Mean 7.43809 +trainer/Advantage Weights Std 22.3537 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.57862e-13 +trainer/Advantage Score Mean -0.201242 +trainer/Advantage Score Std 0.511685 +trainer/Advantage Score Max 2.43262 +trainer/Advantage Score Min -2.84122 +trainer/V1 Predictions Mean -72.9502 +trainer/V1 Predictions Std 21.4569 +trainer/V1 Predictions Max -0.523749 +trainer/V1 Predictions Min -87.6334 +trainer/VF Loss 0.0706766 +expl/num steps total 902000 +expl/num paths total 1239 +expl/path length Mean 500 +expl/path length Std 173 +expl/path length Max 673 +expl/path length Min 327 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0268484 +expl/Actions Std 0.835812 +expl/Actions Max 2.50478 +expl/Actions Min -2.20504 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 812637 +eval/num paths total 907 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0448601 +eval/Actions Std 0.676693 +eval/Actions Max 0.999618 +eval/Actions Min -0.999151 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.7772e-06 +time/evaluation sampling (s) 3.56041 +time/exploration sampling (s) 3.81694 +time/logging (s) 0.0112167 +time/saving (s) 0.0162561 +time/training (s) 16.4049 +time/epoch (s) 23.8097 +time/total (s) 23326.1 +Epoch -99 +------------------------------ ---------------- +2022-05-16 00:31:42.492148 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -98 finished +------------------------------ ---------------- +epoch -98 +replay_buffer/size 999047 +trainer/num train calls 903000 +trainer/QF1 Loss 0.783083 +trainer/QF2 Loss 0.749638 +trainer/Policy Loss 16.8095 +trainer/Q1 Predictions Mean -75.0378 +trainer/Q1 Predictions Std 16.7683 +trainer/Q1 Predictions Max -3.99004 +trainer/Q1 Predictions Min -88.2537 +trainer/Q2 Predictions Mean -74.9842 +trainer/Q2 Predictions Std 16.7987 +trainer/Q2 Predictions Max -3.47399 +trainer/Q2 Predictions Min -88.0699 +trainer/Q Targets Mean -75.0499 +trainer/Q Targets Std 16.8262 +trainer/Q Targets Max -3.09649 +trainer/Q Targets Min -88.4366 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0118802 +trainer/policy/mean Std 0.71914 +trainer/policy/mean Max 0.998935 +trainer/policy/mean Min -0.998876 +trainer/policy/std Mean 0.409045 +trainer/policy/std Std 0.0189178 +trainer/policy/std Max 0.43305 +trainer/policy/std Min 0.383883 +trainer/Advantage Weights Mean 4.85159 +trainer/Advantage Weights Std 18.9383 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.66569e-13 +trainer/Advantage Score Mean -0.354323 +trainer/Advantage Score Std 0.532439 +trainer/Advantage Score Max 0.817356 +trainer/Advantage Score Min -2.7665 +trainer/V1 Predictions Mean -74.8047 +trainer/V1 Predictions Std 17.049 +trainer/V1 Predictions Max -2.68201 +trainer/V1 Predictions Min -88.26 +trainer/VF Loss 0.055567 +expl/num steps total 903000 +expl/num paths total 1241 +expl/path length Mean 500 +expl/path length Std 350 +expl/path length Max 850 +expl/path length Min 150 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0293233 +expl/Actions Std 0.82872 +expl/Actions Max 2.26006 +expl/Actions Min -2.31974 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 813118 +eval/num paths total 908 +eval/path length Mean 481 +eval/path length Std 0 +eval/path length Max 481 +eval/path length Min 481 +eval/Rewards Mean 0.002079 +eval/Rewards Std 0.0455487 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0177878 +eval/Actions Std 0.738915 +eval/Actions Max 0.999353 +eval/Actions Min -0.999484 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.86987e-06 +time/evaluation sampling (s) 3.77151 +time/exploration sampling (s) 4.65232 +time/logging (s) 0.0099382 +time/saving (s) 0.0183219 +time/training (s) 16.4184 +time/epoch (s) 24.8705 +time/total (s) 23351 +Epoch -98 +------------------------------ ---------------- +2022-05-16 00:32:06.779307 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -97 finished +------------------------------ ---------------- +epoch -97 +replay_buffer/size 999047 +trainer/num train calls 904000 +trainer/QF1 Loss 0.692417 +trainer/QF2 Loss 0.700604 +trainer/Policy Loss 32.8786 +trainer/Q1 Predictions Mean -74.2554 +trainer/Q1 Predictions Std 18.7457 +trainer/Q1 Predictions Max -1.11612 +trainer/Q1 Predictions Min -87.8965 +trainer/Q2 Predictions Mean -74.3209 +trainer/Q2 Predictions Std 18.6447 +trainer/Q2 Predictions Max -0.592163 +trainer/Q2 Predictions Min -88.1142 +trainer/Q Targets Mean -74.0609 +trainer/Q Targets Std 18.7625 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9313 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0188538 +trainer/policy/mean Std 0.731193 +trainer/policy/mean Max 0.999321 +trainer/policy/mean Min -0.998455 +trainer/policy/std Mean 0.408765 +trainer/policy/std Std 0.0186263 +trainer/policy/std Max 0.431822 +trainer/policy/std Min 0.380521 +trainer/Advantage Weights Mean 4.55025 +trainer/Advantage Weights Std 17.6791 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.70675e-15 +trainer/Advantage Score Mean -0.35656 +trainer/Advantage Score Std 0.575991 +trainer/Advantage Score Max 1.17408 +trainer/Advantage Score Min -3.27971 +trainer/V1 Predictions Mean -73.8034 +trainer/V1 Predictions Std 18.827 +trainer/V1 Predictions Max -0.497692 +trainer/V1 Predictions Min -87.8245 +trainer/VF Loss 0.0665322 +expl/num steps total 904000 +expl/num paths total 1243 +expl/path length Mean 500 +expl/path length Std 186 +expl/path length Max 686 +expl/path length Min 314 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0309236 +expl/Actions Std 0.834973 +expl/Actions Max 2.38948 +expl/Actions Min -2.38782 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 814118 +eval/num paths total 909 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.121486 +eval/Actions Std 0.753723 +eval/Actions Max 0.999923 +eval/Actions Min -0.999965 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.15206e-06 +time/evaluation sampling (s) 3.23991 +time/exploration sampling (s) 4.75025 +time/logging (s) 0.00892597 +time/saving (s) 0.0116495 +time/training (s) 16.2624 +time/epoch (s) 24.2731 +time/total (s) 23375.3 +Epoch -97 +------------------------------ ---------------- +2022-05-16 00:32:30.409232 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -96 finished +------------------------------ ---------------- +epoch -96 +replay_buffer/size 999047 +trainer/num train calls 905000 +trainer/QF1 Loss 0.837977 +trainer/QF2 Loss 0.847607 +trainer/Policy Loss 25.7327 +trainer/Q1 Predictions Mean -74.9674 +trainer/Q1 Predictions Std 15.7247 +trainer/Q1 Predictions Max -0.982694 +trainer/Q1 Predictions Min -86.6856 +trainer/Q2 Predictions Mean -74.933 +trainer/Q2 Predictions Std 15.8789 +trainer/Q2 Predictions Max -0.099267 +trainer/Q2 Predictions Min -86.6411 +trainer/Q Targets Mean -75.4498 +trainer/Q Targets Std 16.0858 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3786 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.000725013 +trainer/policy/mean Std 0.730643 +trainer/policy/mean Max 0.999627 +trainer/policy/mean Min -0.999152 +trainer/policy/std Mean 0.409508 +trainer/policy/std Std 0.017139 +trainer/policy/std Max 0.427214 +trainer/policy/std Min 0.384818 +trainer/Advantage Weights Mean 6.03989 +trainer/Advantage Weights Std 18.8143 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.75862e-16 +trainer/Advantage Score Mean -0.25034 +trainer/Advantage Score Std 0.559962 +trainer/Advantage Score Max 1.44791 +trainer/Advantage Score Min -3.47926 +trainer/V1 Predictions Mean -75.2511 +trainer/V1 Predictions Std 16.0709 +trainer/V1 Predictions Max 0.71374 +trainer/V1 Predictions Min -87.3798 +trainer/VF Loss 0.0653262 +expl/num steps total 905000 +expl/num paths total 1244 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0697604 +expl/Actions Std 0.855153 +expl/Actions Max 2.58153 +expl/Actions Min -2.29906 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 815118 +eval/num paths total 910 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.017019 +eval/Actions Std 0.774929 +eval/Actions Max 0.999928 +eval/Actions Min -0.999988 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.75904e-06 +time/evaluation sampling (s) 3.07721 +time/exploration sampling (s) 4.52853 +time/logging (s) 0.0096817 +time/saving (s) 0.0143044 +time/training (s) 15.9868 +time/epoch (s) 23.6165 +time/total (s) 23398.9 +Epoch -96 +------------------------------ ---------------- +2022-05-16 00:32:54.031617 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -95 finished +------------------------------ ---------------- +epoch -95 +replay_buffer/size 999047 +trainer/num train calls 906000 +trainer/QF1 Loss 0.711311 +trainer/QF2 Loss 0.758964 +trainer/Policy Loss 6.57543 +trainer/Q1 Predictions Mean -74.2642 +trainer/Q1 Predictions Std 17.9965 +trainer/Q1 Predictions Max -3.60548 +trainer/Q1 Predictions Min -88.1415 +trainer/Q2 Predictions Mean -74.1982 +trainer/Q2 Predictions Std 17.9701 +trainer/Q2 Predictions Max -4.29808 +trainer/Q2 Predictions Min -87.737 +trainer/Q Targets Mean -73.9853 +trainer/Q Targets Std 17.8263 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5962 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0043805 +trainer/policy/mean Std 0.738647 +trainer/policy/mean Max 0.999822 +trainer/policy/mean Min -0.999669 +trainer/policy/std Mean 0.40802 +trainer/policy/std Std 0.0190425 +trainer/policy/std Max 0.428665 +trainer/policy/std Min 0.382551 +trainer/Advantage Weights Mean 1.72691 +trainer/Advantage Weights Std 11.4101 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.07307e-15 +trainer/Advantage Score Mean -0.594451 +trainer/Advantage Score Std 0.588548 +trainer/Advantage Score Max 0.602623 +trainer/Advantage Score Min -3.34161 +trainer/V1 Predictions Mean -73.6624 +trainer/V1 Predictions Std 18.047 +trainer/V1 Predictions Max -2.81855 +trainer/V1 Predictions Min -87.4379 +trainer/VF Loss 0.0741221 +expl/num steps total 906000 +expl/num paths total 1246 +expl/path length Mean 500 +expl/path length Std 122 +expl/path length Max 622 +expl/path length Min 378 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0265915 +expl/Actions Std 0.832307 +expl/Actions Max 2.88842 +expl/Actions Min -2.57856 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 816118 +eval/num paths total 911 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.251367 +eval/Actions Std 0.780723 +eval/Actions Max 0.99992 +eval/Actions Min -0.999865 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.80095e-06 +time/evaluation sampling (s) 3.49737 +time/exploration sampling (s) 4.00617 +time/logging (s) 0.012377 +time/saving (s) 0.0189565 +time/training (s) 16.072 +time/epoch (s) 23.6069 +time/total (s) 23422.5 +Epoch -95 +------------------------------ ---------------- +2022-05-16 00:33:17.413677 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -94 finished +------------------------------ ---------------- +epoch -94 +replay_buffer/size 999047 +trainer/num train calls 907000 +trainer/QF1 Loss 1.2675 +trainer/QF2 Loss 0.923205 +trainer/Policy Loss 11.7354 +trainer/Q1 Predictions Mean -73.6302 +trainer/Q1 Predictions Std 18.988 +trainer/Q1 Predictions Max -1.07777 +trainer/Q1 Predictions Min -87.3916 +trainer/Q2 Predictions Mean -73.6307 +trainer/Q2 Predictions Std 18.9889 +trainer/Q2 Predictions Max -0.903169 +trainer/Q2 Predictions Min -87.1964 +trainer/Q Targets Mean -73.659 +trainer/Q Targets Std 19.2048 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8376 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0234153 +trainer/policy/mean Std 0.723985 +trainer/policy/mean Max 0.998771 +trainer/policy/mean Min -0.998736 +trainer/policy/std Mean 0.408523 +trainer/policy/std Std 0.0215648 +trainer/policy/std Max 0.433748 +trainer/policy/std Min 0.376597 +trainer/Advantage Weights Mean 3.27837 +trainer/Advantage Weights Std 14.6614 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.52677e-16 +trainer/Advantage Score Mean -0.423954 +trainer/Advantage Score Std 0.595598 +trainer/Advantage Score Max 1.89777 +trainer/Advantage Score Min -3.5581 +trainer/V1 Predictions Mean -73.4436 +trainer/V1 Predictions Std 19.1057 +trainer/V1 Predictions Max -0.824817 +trainer/V1 Predictions Min -87.4375 +trainer/VF Loss 0.0726698 +expl/num steps total 907000 +expl/num paths total 1248 +expl/path length Mean 500 +expl/path length Std 166 +expl/path length Max 666 +expl/path length Min 334 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0181219 +expl/Actions Std 0.835926 +expl/Actions Max 2.18886 +expl/Actions Min -2.3336 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 816828 +eval/num paths total 912 +eval/path length Mean 710 +eval/path length Std 0 +eval/path length Max 710 +eval/path length Min 710 +eval/Rewards Mean 0.00140845 +eval/Rewards Std 0.0375029 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0222355 +eval/Actions Std 0.733195 +eval/Actions Max 0.999967 +eval/Actions Min -0.999718 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.09804e-06 +time/evaluation sampling (s) 3.55622 +time/exploration sampling (s) 4.18071 +time/logging (s) 0.00699172 +time/saving (s) 0.0108088 +time/training (s) 15.604 +time/epoch (s) 23.3588 +time/total (s) 23445.9 +Epoch -94 +------------------------------ ---------------- +2022-05-16 00:33:41.586385 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -93 finished +------------------------------ ---------------- +epoch -93 +replay_buffer/size 999047 +trainer/num train calls 908000 +trainer/QF1 Loss 0.686366 +trainer/QF2 Loss 0.806611 +trainer/Policy Loss 13.7123 +trainer/Q1 Predictions Mean -72.338 +trainer/Q1 Predictions Std 20.2037 +trainer/Q1 Predictions Max -0.684787 +trainer/Q1 Predictions Min -88.0797 +trainer/Q2 Predictions Mean -72.3554 +trainer/Q2 Predictions Std 20.2713 +trainer/Q2 Predictions Max -0.93968 +trainer/Q2 Predictions Min -88.1726 +trainer/Q Targets Mean -72.6811 +trainer/Q Targets Std 20.3605 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1323 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00460708 +trainer/policy/mean Std 0.738133 +trainer/policy/mean Max 0.99981 +trainer/policy/mean Min -0.999383 +trainer/policy/std Mean 0.408116 +trainer/policy/std Std 0.0200797 +trainer/policy/std Max 0.430217 +trainer/policy/std Min 0.37847 +trainer/Advantage Weights Mean 3.76499 +trainer/Advantage Weights Std 15.9679 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.86434e-19 +trainer/Advantage Score Mean -0.383756 +trainer/Advantage Score Std 0.562078 +trainer/Advantage Score Max 2.26607 +trainer/Advantage Score Min -4.21672 +trainer/V1 Predictions Mean -72.4595 +trainer/V1 Predictions Std 20.3713 +trainer/V1 Predictions Max -0.939579 +trainer/V1 Predictions Min -87.9491 +trainer/VF Loss 0.075266 +expl/num steps total 908000 +expl/num paths total 1250 +expl/path length Mean 500 +expl/path length Std 310 +expl/path length Max 810 +expl/path length Min 190 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0299673 +expl/Actions Std 0.837957 +expl/Actions Max 2.59116 +expl/Actions Min -2.07507 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 817828 +eval/num paths total 913 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.185229 +eval/Actions Std 0.700721 +eval/Actions Max 0.999848 +eval/Actions Min -0.998808 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.44707e-06 +time/evaluation sampling (s) 3.53928 +time/exploration sampling (s) 4.95054 +time/logging (s) 0.00691471 +time/saving (s) 0.00957449 +time/training (s) 15.6523 +time/epoch (s) 24.1586 +time/total (s) 23470 +Epoch -93 +------------------------------ ---------------- +2022-05-16 00:34:05.530736 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -92 finished +------------------------------ ---------------- +epoch -92 +replay_buffer/size 999047 +trainer/num train calls 909000 +trainer/QF1 Loss 1.12284 +trainer/QF2 Loss 1.20691 +trainer/Policy Loss 10.8604 +trainer/Q1 Predictions Mean -72.9907 +trainer/Q1 Predictions Std 18.2089 +trainer/Q1 Predictions Max -3.63234 +trainer/Q1 Predictions Min -87.3908 +trainer/Q2 Predictions Mean -73.0344 +trainer/Q2 Predictions Std 18.2776 +trainer/Q2 Predictions Max -5.00722 +trainer/Q2 Predictions Min -87.6681 +trainer/Q Targets Mean -73.1556 +trainer/Q Targets Std 18.2532 +trainer/Q Targets Max -2.9943 +trainer/Q Targets Min -87.382 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0203147 +trainer/policy/mean Std 0.738418 +trainer/policy/mean Max 0.999497 +trainer/policy/mean Min -0.999931 +trainer/policy/std Mean 0.407159 +trainer/policy/std Std 0.019981 +trainer/policy/std Max 0.428969 +trainer/policy/std Min 0.379194 +trainer/Advantage Weights Mean 2.8118 +trainer/Advantage Weights Std 13.0734 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.52223e-21 +trainer/Advantage Score Mean -0.503534 +trainer/Advantage Score Std 0.758758 +trainer/Advantage Score Max 0.7164 +trainer/Advantage Score Min -4.79341 +trainer/V1 Predictions Mean -72.7692 +trainer/V1 Predictions Std 18.5896 +trainer/V1 Predictions Max -2.27303 +trainer/V1 Predictions Min -87.3315 +trainer/VF Loss 0.0908181 +expl/num steps total 909000 +expl/num paths total 1252 +expl/path length Mean 500 +expl/path length Std 143 +expl/path length Max 643 +expl/path length Min 357 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0197764 +expl/Actions Std 0.843851 +expl/Actions Max 2.31233 +expl/Actions Min -2.22584 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 818403 +eval/num paths total 914 +eval/path length Mean 575 +eval/path length Std 0 +eval/path length Max 575 +eval/path length Min 575 +eval/Rewards Mean 0.00173913 +eval/Rewards Std 0.0416666 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean -0.00564789 +eval/Actions Std 0.753425 +eval/Actions Max 0.99988 +eval/Actions Min -0.999917 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.31435e-06 +time/evaluation sampling (s) 2.83556 +time/exploration sampling (s) 4.87318 +time/logging (s) 0.00665832 +time/saving (s) 0.0125802 +time/training (s) 16.2033 +time/epoch (s) 23.9312 +time/total (s) 23494 +Epoch -92 +------------------------------ ---------------- +2022-05-16 00:34:29.258530 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -91 finished +------------------------------ ---------------- +epoch -91 +replay_buffer/size 999047 +trainer/num train calls 910000 +trainer/QF1 Loss 0.605508 +trainer/QF2 Loss 0.497913 +trainer/Policy Loss 25.5771 +trainer/Q1 Predictions Mean -76.4729 +trainer/Q1 Predictions Std 15.2639 +trainer/Q1 Predictions Max -2.09417 +trainer/Q1 Predictions Min -88.5133 +trainer/Q2 Predictions Mean -76.5842 +trainer/Q2 Predictions Std 15.2813 +trainer/Q2 Predictions Max -2.25005 +trainer/Q2 Predictions Min -88.6093 +trainer/Q Targets Mean -76.7652 +trainer/Q Targets Std 15.187 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.4652 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.024919 +trainer/policy/mean Std 0.73584 +trainer/policy/mean Max 0.999139 +trainer/policy/mean Min -0.999713 +trainer/policy/std Mean 0.409254 +trainer/policy/std Std 0.0201316 +trainer/policy/std Max 0.432036 +trainer/policy/std Min 0.380217 +trainer/Advantage Weights Mean 4.44324 +trainer/Advantage Weights Std 17.1773 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.36497e-17 +trainer/Advantage Score Mean -0.302837 +trainer/Advantage Score Std 0.62755 +trainer/Advantage Score Max 1.12263 +trainer/Advantage Score Min -3.79305 +trainer/V1 Predictions Mean -76.5594 +trainer/V1 Predictions Std 15.3085 +trainer/V1 Predictions Max -2.1642 +trainer/V1 Predictions Min -88.3949 +trainer/VF Loss 0.0634219 +expl/num steps total 910000 +expl/num paths total 1253 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.16906 +expl/Actions Std 0.797027 +expl/Actions Max 2.37667 +expl/Actions Min -2.17009 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 819403 +eval/num paths total 915 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0280562 +eval/Actions Std 0.702511 +eval/Actions Max 0.999792 +eval/Actions Min -0.999807 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.55601e-06 +time/evaluation sampling (s) 3.56439 +time/exploration sampling (s) 3.92765 +time/logging (s) 0.00844252 +time/saving (s) 0.0131227 +time/training (s) 16.202 +time/epoch (s) 23.7156 +time/total (s) 23517.7 +Epoch -91 +------------------------------ ---------------- +2022-05-16 00:34:53.766417 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -90 finished +------------------------------ ---------------- +epoch -90 +replay_buffer/size 999047 +trainer/num train calls 911000 +trainer/QF1 Loss 0.76884 +trainer/QF2 Loss 0.814941 +trainer/Policy Loss 19.7844 +trainer/Q1 Predictions Mean -75.2083 +trainer/Q1 Predictions Std 17.7464 +trainer/Q1 Predictions Max -0.187726 +trainer/Q1 Predictions Min -87.8297 +trainer/Q2 Predictions Mean -75.2226 +trainer/Q2 Predictions Std 17.7433 +trainer/Q2 Predictions Max -0.115924 +trainer/Q2 Predictions Min -87.9447 +trainer/Q Targets Mean -75.367 +trainer/Q Targets Std 17.3974 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8585 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0121808 +trainer/policy/mean Std 0.740086 +trainer/policy/mean Max 0.999297 +trainer/policy/mean Min -0.999515 +trainer/policy/std Mean 0.408831 +trainer/policy/std Std 0.0193805 +trainer/policy/std Max 0.43292 +trainer/policy/std Min 0.376246 +trainer/Advantage Weights Mean 4.68485 +trainer/Advantage Weights Std 16.9334 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.29174e-17 +trainer/Advantage Score Mean -0.250551 +trainer/Advantage Score Std 0.604133 +trainer/Advantage Score Max 2.26425 +trainer/Advantage Score Min -3.8888 +trainer/V1 Predictions Mean -75.0234 +trainer/V1 Predictions Std 17.8245 +trainer/V1 Predictions Max 0.333963 +trainer/V1 Predictions Min -87.8818 +trainer/VF Loss 0.0818829 +expl/num steps total 911000 +expl/num paths total 1254 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0195872 +expl/Actions Std 0.82585 +expl/Actions Max 2.24589 +expl/Actions Min -2.33899 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 820403 +eval/num paths total 916 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.109625 +eval/Actions Std 0.826672 +eval/Actions Max 0.999867 +eval/Actions Min -0.99955 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.51597e-06 +time/evaluation sampling (s) 3.7792 +time/exploration sampling (s) 4.77242 +time/logging (s) 0.00877071 +time/saving (s) 0.014069 +time/training (s) 15.9179 +time/epoch (s) 24.4924 +time/total (s) 23542.2 +Epoch -90 +------------------------------ ---------------- +2022-05-16 00:35:18.215861 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -89 finished +------------------------------ ---------------- +epoch -89 +replay_buffer/size 999047 +trainer/num train calls 912000 +trainer/QF1 Loss 1.20773 +trainer/QF2 Loss 1.14986 +trainer/Policy Loss 34.9206 +trainer/Q1 Predictions Mean -73.7207 +trainer/Q1 Predictions Std 19.2194 +trainer/Q1 Predictions Max 0.414565 +trainer/Q1 Predictions Min -88.4731 +trainer/Q2 Predictions Mean -73.729 +trainer/Q2 Predictions Std 19.1394 +trainer/Q2 Predictions Max -0.54871 +trainer/Q2 Predictions Min -87.9001 +trainer/Q Targets Mean -74.1077 +trainer/Q Targets Std 18.7908 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0592 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0087323 +trainer/policy/mean Std 0.728123 +trainer/policy/mean Max 0.999525 +trainer/policy/mean Min -0.999702 +trainer/policy/std Mean 0.409221 +trainer/policy/std Std 0.0203543 +trainer/policy/std Max 0.43215 +trainer/policy/std Min 0.376265 +trainer/Advantage Weights Mean 8.55066 +trainer/Advantage Weights Std 24.7568 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.83287e-10 +trainer/Advantage Score Mean -0.165628 +trainer/Advantage Score Std 0.506766 +trainer/Advantage Score Max 1.42073 +trainer/Advantage Score Min -2.242 +trainer/V1 Predictions Mean -73.8366 +trainer/V1 Predictions Std 19.047 +trainer/V1 Predictions Max -1.57225 +trainer/V1 Predictions Min -87.9189 +trainer/VF Loss 0.0894419 +expl/num steps total 912000 +expl/num paths total 1255 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.119786 +expl/Actions Std 0.877298 +expl/Actions Max 2.37255 +expl/Actions Min -2.41078 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 821403 +eval/num paths total 917 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.31608 +eval/Actions Std 0.698006 +eval/Actions Max 0.999049 +eval/Actions Min -0.999208 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.61702e-06 +time/evaluation sampling (s) 3.37928 +time/exploration sampling (s) 5.02962 +time/logging (s) 0.00695511 +time/saving (s) 0.011465 +time/training (s) 16.0055 +time/epoch (s) 24.4328 +time/total (s) 23566.6 +Epoch -89 +------------------------------ ---------------- +2022-05-16 00:35:42.058647 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -88 finished +------------------------------ ---------------- +epoch -88 +replay_buffer/size 999047 +trainer/num train calls 913000 +trainer/QF1 Loss 0.587361 +trainer/QF2 Loss 0.688721 +trainer/Policy Loss 28.7841 +trainer/Q1 Predictions Mean -74.9159 +trainer/Q1 Predictions Std 16.8588 +trainer/Q1 Predictions Max -2.40636 +trainer/Q1 Predictions Min -88.9305 +trainer/Q2 Predictions Mean -74.9381 +trainer/Q2 Predictions Std 16.8027 +trainer/Q2 Predictions Max -2.65113 +trainer/Q2 Predictions Min -88.8409 +trainer/Q Targets Mean -74.8624 +trainer/Q Targets Std 16.9731 +trainer/Q Targets Max -2.11662 +trainer/Q Targets Min -88.7173 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0177858 +trainer/policy/mean Std 0.735285 +trainer/policy/mean Max 0.9997 +trainer/policy/mean Min -0.999283 +trainer/policy/std Mean 0.407361 +trainer/policy/std Std 0.01996 +trainer/policy/std Max 0.429626 +trainer/policy/std Min 0.374075 +trainer/Advantage Weights Mean 5.64267 +trainer/Advantage Weights Std 19.5986 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.77379e-19 +trainer/Advantage Score Mean -0.414422 +trainer/Advantage Score Std 0.754522 +trainer/Advantage Score Max 2.76999 +trainer/Advantage Score Min -4.2421 +trainer/V1 Predictions Mean -74.6046 +trainer/V1 Predictions Std 17.2055 +trainer/V1 Predictions Max -1.72253 +trainer/V1 Predictions Min -88.628 +trainer/VF Loss 0.112095 +expl/num steps total 913000 +expl/num paths total 1256 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.02167 +expl/Actions Std 0.835099 +expl/Actions Max 2.4281 +expl/Actions Min -2.23923 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 822403 +eval/num paths total 918 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0791995 +eval/Actions Std 0.729392 +eval/Actions Max 0.999793 +eval/Actions Min -0.999699 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.93413e-06 +time/evaluation sampling (s) 3.33817 +time/exploration sampling (s) 4.43866 +time/logging (s) 0.0106596 +time/saving (s) 0.0110852 +time/training (s) 16.0338 +time/epoch (s) 23.8324 +time/total (s) 23590.5 +Epoch -88 +------------------------------ ---------------- +2022-05-16 00:36:05.880769 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -87 finished +------------------------------ ---------------- +epoch -87 +replay_buffer/size 999047 +trainer/num train calls 914000 +trainer/QF1 Loss 0.969723 +trainer/QF2 Loss 0.818736 +trainer/Policy Loss 17.952 +trainer/Q1 Predictions Mean -75.5446 +trainer/Q1 Predictions Std 15.6059 +trainer/Q1 Predictions Max -0.253251 +trainer/Q1 Predictions Min -88.1172 +trainer/Q2 Predictions Mean -75.4977 +trainer/Q2 Predictions Std 15.6198 +trainer/Q2 Predictions Max 0.137942 +trainer/Q2 Predictions Min -88.1644 +trainer/Q Targets Mean -75.6011 +trainer/Q Targets Std 15.3041 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1324 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.013791 +trainer/policy/mean Std 0.735439 +trainer/policy/mean Max 0.999874 +trainer/policy/mean Min -0.99957 +trainer/policy/std Mean 0.407453 +trainer/policy/std Std 0.0199055 +trainer/policy/std Max 0.42947 +trainer/policy/std Min 0.375354 +trainer/Advantage Weights Mean 3.66528 +trainer/Advantage Weights Std 17.4913 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.11368e-27 +trainer/Advantage Score Mean -0.464436 +trainer/Advantage Score Std 0.63735 +trainer/Advantage Score Max 1.22036 +trainer/Advantage Score Min -6.03593 +trainer/V1 Predictions Mean -75.2594 +trainer/V1 Predictions Std 15.4716 +trainer/V1 Predictions Max 0.733961 +trainer/V1 Predictions Min -87.9762 +trainer/VF Loss 0.0822268 +expl/num steps total 914000 +expl/num paths total 1257 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.110925 +expl/Actions Std 0.824439 +expl/Actions Max 2.21584 +expl/Actions Min -2.45843 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 823403 +eval/num paths total 919 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.402375 +eval/Actions Std 0.638805 +eval/Actions Max 0.998459 +eval/Actions Min -0.998756 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.87592e-06 +time/evaluation sampling (s) 3.31242 +time/exploration sampling (s) 4.83734 +time/logging (s) 0.00682268 +time/saving (s) 0.0104387 +time/training (s) 15.6383 +time/epoch (s) 23.8053 +time/total (s) 23614.3 +Epoch -87 +------------------------------ ---------------- +2022-05-16 00:36:28.898541 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -86 finished +------------------------------ ---------------- +epoch -86 +replay_buffer/size 999047 +trainer/num train calls 915000 +trainer/QF1 Loss 0.746656 +trainer/QF2 Loss 0.697605 +trainer/Policy Loss 9.00487 +trainer/Q1 Predictions Mean -73.6334 +trainer/Q1 Predictions Std 19.4573 +trainer/Q1 Predictions Max -2.79159 +trainer/Q1 Predictions Min -87.8781 +trainer/Q2 Predictions Mean -73.5153 +trainer/Q2 Predictions Std 19.6785 +trainer/Q2 Predictions Max -0.388685 +trainer/Q2 Predictions Min -87.5568 +trainer/Q Targets Mean -73.7044 +trainer/Q Targets Std 19.672 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5743 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0163144 +trainer/policy/mean Std 0.742144 +trainer/policy/mean Max 0.999372 +trainer/policy/mean Min -0.999834 +trainer/policy/std Mean 0.407642 +trainer/policy/std Std 0.0196851 +trainer/policy/std Max 0.431157 +trainer/policy/std Min 0.376667 +trainer/Advantage Weights Mean 3.38795 +trainer/Advantage Weights Std 16.0139 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.46982e-16 +trainer/Advantage Score Mean -0.455444 +trainer/Advantage Score Std 0.571999 +trainer/Advantage Score Max 1.20483 +trainer/Advantage Score Min -3.51421 +trainer/V1 Predictions Mean -73.4309 +trainer/V1 Predictions Std 19.8168 +trainer/V1 Predictions Max -0.0345131 +trainer/V1 Predictions Min -87.5668 +trainer/VF Loss 0.0682414 +expl/num steps total 915000 +expl/num paths total 1258 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.043805 +expl/Actions Std 0.820204 +expl/Actions Max 2.93529 +expl/Actions Min -2.10559 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 824403 +eval/num paths total 920 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.538031 +eval/Actions Std 0.555876 +eval/Actions Max 0.999808 +eval/Actions Min -0.99974 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.18978e-06 +time/evaluation sampling (s) 3.56352 +time/exploration sampling (s) 3.77306 +time/logging (s) 0.00697935 +time/saving (s) 0.00965481 +time/training (s) 15.6523 +time/epoch (s) 23.0055 +time/total (s) 23637.3 +Epoch -86 +------------------------------ ---------------- +2022-05-16 00:36:52.514986 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -85 finished +------------------------------ ---------------- +epoch -85 +replay_buffer/size 999047 +trainer/num train calls 916000 +trainer/QF1 Loss 0.7239 +trainer/QF2 Loss 0.655927 +trainer/Policy Loss 16.3577 +trainer/Q1 Predictions Mean -73.968 +trainer/Q1 Predictions Std 19.6863 +trainer/Q1 Predictions Max -0.117964 +trainer/Q1 Predictions Min -88.0026 +trainer/Q2 Predictions Mean -73.9793 +trainer/Q2 Predictions Std 19.704 +trainer/Q2 Predictions Max 0.301083 +trainer/Q2 Predictions Min -87.8225 +trainer/Q Targets Mean -74.035 +trainer/Q Targets Std 19.5489 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8297 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0197691 +trainer/policy/mean Std 0.74089 +trainer/policy/mean Max 0.999663 +trainer/policy/mean Min -0.99972 +trainer/policy/std Mean 0.409441 +trainer/policy/std Std 0.0184195 +trainer/policy/std Max 0.429073 +trainer/policy/std Min 0.375745 +trainer/Advantage Weights Mean 4.79485 +trainer/Advantage Weights Std 17.9964 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.4036e-12 +trainer/Advantage Score Mean -0.314245 +trainer/Advantage Score Std 0.591641 +trainer/Advantage Score Max 1.60289 +trainer/Advantage Score Min -2.5944 +trainer/V1 Predictions Mean -73.7958 +trainer/V1 Predictions Std 19.6621 +trainer/V1 Predictions Max -0.0108308 +trainer/V1 Predictions Min -87.8028 +trainer/VF Loss 0.0755814 +expl/num steps total 916000 +expl/num paths total 1259 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0281757 +expl/Actions Std 0.804385 +expl/Actions Max 2.35024 +expl/Actions Min -2.14351 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 825403 +eval/num paths total 921 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.016408 +eval/Actions Std 0.735413 +eval/Actions Max 0.999876 +eval/Actions Min -0.999215 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 4.27477e-06 +time/evaluation sampling (s) 3.71214 +time/exploration sampling (s) 4.72859 +time/logging (s) 0.00994611 +time/saving (s) 0.0142703 +time/training (s) 15.1416 +time/epoch (s) 23.6066 +time/total (s) 23660.9 +Epoch -85 +------------------------------ ---------------- +2022-05-16 00:37:15.268898 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -84 finished +------------------------------ ---------------- +epoch -84 +replay_buffer/size 999047 +trainer/num train calls 917000 +trainer/QF1 Loss 0.87721 +trainer/QF2 Loss 0.753067 +trainer/Policy Loss 25.7483 +trainer/Q1 Predictions Mean -74.2412 +trainer/Q1 Predictions Std 18.3162 +trainer/Q1 Predictions Max -0.589224 +trainer/Q1 Predictions Min -88.2763 +trainer/Q2 Predictions Mean -74.2686 +trainer/Q2 Predictions Std 18.4007 +trainer/Q2 Predictions Max -0.478819 +trainer/Q2 Predictions Min -88.5014 +trainer/Q Targets Mean -74.068 +trainer/Q Targets Std 18.5318 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6466 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.017444 +trainer/policy/mean Std 0.736733 +trainer/policy/mean Max 0.999859 +trainer/policy/mean Min -0.999772 +trainer/policy/std Mean 0.409726 +trainer/policy/std Std 0.018192 +trainer/policy/std Max 0.429248 +trainer/policy/std Min 0.379697 +trainer/Advantage Weights Mean 4.73881 +trainer/Advantage Weights Std 18.496 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.51198e-21 +trainer/Advantage Score Mean -0.439411 +trainer/Advantage Score Std 0.791333 +trainer/Advantage Score Max 2.60744 +trainer/Advantage Score Min -4.79409 +trainer/V1 Predictions Mean -73.8618 +trainer/V1 Predictions Std 18.5767 +trainer/V1 Predictions Max -0.0631539 +trainer/V1 Predictions Min -87.6246 +trainer/VF Loss 0.131023 +expl/num steps total 917000 +expl/num paths total 1261 +expl/path length Mean 500 +expl/path length Std 415 +expl/path length Max 915 +expl/path length Min 85 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0177359 +expl/Actions Std 0.825524 +expl/Actions Max 2.21923 +expl/Actions Min -2.19787 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 825980 +eval/num paths total 922 +eval/path length Mean 577 +eval/path length Std 0 +eval/path length Max 577 +eval/path length Min 577 +eval/Rewards Mean 0.0017331 +eval/Rewards Std 0.0415945 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0197458 +eval/Actions Std 0.755616 +eval/Actions Max 0.999955 +eval/Actions Min -0.999871 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 4.16208e-06 +time/evaluation sampling (s) 3.19551 +time/exploration sampling (s) 3.9014 +time/logging (s) 0.00572032 +time/saving (s) 0.00922438 +time/training (s) 15.6196 +time/epoch (s) 22.7315 +time/total (s) 23683.6 +Epoch -84 +------------------------------ ---------------- +2022-05-16 00:37:39.718592 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -83 finished +------------------------------ ---------------- +epoch -83 +replay_buffer/size 999047 +trainer/num train calls 918000 +trainer/QF1 Loss 0.671007 +trainer/QF2 Loss 0.639984 +trainer/Policy Loss 49.025 +trainer/Q1 Predictions Mean -73.9815 +trainer/Q1 Predictions Std 18.7943 +trainer/Q1 Predictions Max -0.205008 +trainer/Q1 Predictions Min -88.267 +trainer/Q2 Predictions Mean -73.9668 +trainer/Q2 Predictions Std 18.7931 +trainer/Q2 Predictions Max -0.331756 +trainer/Q2 Predictions Min -88.0082 +trainer/Q Targets Mean -74.1419 +trainer/Q Targets Std 18.964 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9746 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0320227 +trainer/policy/mean Std 0.7316 +trainer/policy/mean Max 0.99944 +trainer/policy/mean Min -0.99834 +trainer/policy/std Mean 0.409989 +trainer/policy/std Std 0.0185536 +trainer/policy/std Max 0.429836 +trainer/policy/std Min 0.380223 +trainer/Advantage Weights Mean 9.58654 +trainer/Advantage Weights Std 23.2586 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.56539e-12 +trainer/Advantage Score Mean -0.167497 +trainer/Advantage Score Std 0.588851 +trainer/Advantage Score Max 1.02724 +trainer/Advantage Score Min -2.71829 +trainer/V1 Predictions Mean -73.9106 +trainer/V1 Predictions Std 18.9909 +trainer/V1 Predictions Max -1.21263 +trainer/V1 Predictions Min -88.2115 +trainer/VF Loss 0.0716433 +expl/num steps total 918000 +expl/num paths total 1263 +expl/path length Mean 500 +expl/path length Std 441 +expl/path length Max 941 +expl/path length Min 59 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0349192 +expl/Actions Std 0.799829 +expl/Actions Max 2.25967 +expl/Actions Min -2.20941 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 826980 +eval/num paths total 923 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.245984 +eval/Actions Std 0.596528 +eval/Actions Max 0.999841 +eval/Actions Min -0.999867 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.80282e-06 +time/evaluation sampling (s) 3.64061 +time/exploration sampling (s) 4.61745 +time/logging (s) 0.00843488 +time/saving (s) 0.0147228 +time/training (s) 16.16 +time/epoch (s) 24.4413 +time/total (s) 23708.1 +Epoch -83 +------------------------------ ---------------- +2022-05-16 00:38:03.294910 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -82 finished +------------------------------ ---------------- +epoch -82 +replay_buffer/size 999047 +trainer/num train calls 919000 +trainer/QF1 Loss 0.624916 +trainer/QF2 Loss 0.604813 +trainer/Policy Loss 36.7697 +trainer/Q1 Predictions Mean -74.7045 +trainer/Q1 Predictions Std 16.2692 +trainer/Q1 Predictions Max -8.32994 +trainer/Q1 Predictions Min -86.923 +trainer/Q2 Predictions Mean -74.6884 +trainer/Q2 Predictions Std 16.3169 +trainer/Q2 Predictions Max -7.14446 +trainer/Q2 Predictions Min -86.8834 +trainer/Q Targets Mean -75.0387 +trainer/Q Targets Std 16.3266 +trainer/Q Targets Max -7.161 +trainer/Q Targets Min -87.53 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00834239 +trainer/policy/mean Std 0.725984 +trainer/policy/mean Max 0.999723 +trainer/policy/mean Min -0.998986 +trainer/policy/std Mean 0.409227 +trainer/policy/std Std 0.0192308 +trainer/policy/std Max 0.429413 +trainer/policy/std Min 0.377947 +trainer/Advantage Weights Mean 6.96271 +trainer/Advantage Weights Std 20.3734 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.24275e-13 +trainer/Advantage Score Mean -0.253787 +trainer/Advantage Score Std 0.633507 +trainer/Advantage Score Max 1.6096 +trainer/Advantage Score Min -2.79536 +trainer/V1 Predictions Mean -74.8238 +trainer/V1 Predictions Std 16.4614 +trainer/V1 Predictions Max -9.92516 +trainer/V1 Predictions Min -87.3905 +trainer/VF Loss 0.0741801 +expl/num steps total 919000 +expl/num paths total 1264 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0243623 +expl/Actions Std 0.829035 +expl/Actions Max 2.28676 +expl/Actions Min -2.43728 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 827980 +eval/num paths total 924 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0863253 +eval/Actions Std 0.742442 +eval/Actions Max 0.999079 +eval/Actions Min -0.999191 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.59699e-06 +time/evaluation sampling (s) 3.85094 +time/exploration sampling (s) 4.35067 +time/logging (s) 0.00689913 +time/saving (s) 0.00956526 +time/training (s) 15.3423 +time/epoch (s) 23.5604 +time/total (s) 23731.7 +Epoch -82 +------------------------------ ---------------- +2022-05-16 00:38:26.932169 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -81 finished +------------------------------ ---------------- +epoch -81 +replay_buffer/size 999047 +trainer/num train calls 920000 +trainer/QF1 Loss 0.872541 +trainer/QF2 Loss 0.922476 +trainer/Policy Loss 17.938 +trainer/Q1 Predictions Mean -73.8536 +trainer/Q1 Predictions Std 18.2918 +trainer/Q1 Predictions Max 0.309909 +trainer/Q1 Predictions Min -87.8767 +trainer/Q2 Predictions Mean -73.8284 +trainer/Q2 Predictions Std 18.2782 +trainer/Q2 Predictions Max -0.969658 +trainer/Q2 Predictions Min -87.872 +trainer/Q Targets Mean -74.159 +trainer/Q Targets Std 18.1397 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7624 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0128287 +trainer/policy/mean Std 0.739461 +trainer/policy/mean Max 0.999772 +trainer/policy/mean Min -0.998 +trainer/policy/std Mean 0.410329 +trainer/policy/std Std 0.0194614 +trainer/policy/std Max 0.433287 +trainer/policy/std Min 0.380035 +trainer/Advantage Weights Mean 5.88279 +trainer/Advantage Weights Std 19.6977 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.08592e-19 +trainer/Advantage Score Mean -0.280382 +trainer/Advantage Score Std 0.626993 +trainer/Advantage Score Max 1.91682 +trainer/Advantage Score Min -4.36667 +trainer/V1 Predictions Mean -73.8889 +trainer/V1 Predictions Std 18.31 +trainer/V1 Predictions Max -0.952602 +trainer/V1 Predictions Min -87.8097 +trainer/VF Loss 0.0821941 +expl/num steps total 920000 +expl/num paths total 1266 +expl/path length Mean 500 +expl/path length Std 300 +expl/path length Max 800 +expl/path length Min 200 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.038143 +expl/Actions Std 0.808651 +expl/Actions Max 2.53028 +expl/Actions Min -2.41274 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 828571 +eval/num paths total 925 +eval/path length Mean 591 +eval/path length Std 0 +eval/path length Max 591 +eval/path length Min 591 +eval/Rewards Mean 0.00169205 +eval/Rewards Std 0.0410997 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0383535 +eval/Actions Std 0.73588 +eval/Actions Max 0.999523 +eval/Actions Min -0.999142 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.46079e-06 +time/evaluation sampling (s) 3.0817 +time/exploration sampling (s) 4.69024 +time/logging (s) 0.00695272 +time/saving (s) 0.0119186 +time/training (s) 15.8338 +time/epoch (s) 23.6246 +time/total (s) 23755.3 +Epoch -81 +------------------------------ ---------------- +2022-05-16 00:38:51.539574 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -80 finished +------------------------------ ---------------- +epoch -80 +replay_buffer/size 999047 +trainer/num train calls 921000 +trainer/QF1 Loss 0.912541 +trainer/QF2 Loss 1.01566 +trainer/Policy Loss 9.15622 +trainer/Q1 Predictions Mean -75.3594 +trainer/Q1 Predictions Std 16.4208 +trainer/Q1 Predictions Max -3.30113 +trainer/Q1 Predictions Min -87.8295 +trainer/Q2 Predictions Mean -75.3294 +trainer/Q2 Predictions Std 16.54 +trainer/Q2 Predictions Max 0.13413 +trainer/Q2 Predictions Min -87.9918 +trainer/Q Targets Mean -75.2019 +trainer/Q Targets Std 16.314 +trainer/Q Targets Max -1.97554 +trainer/Q Targets Min -87.8869 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0306086 +trainer/policy/mean Std 0.735859 +trainer/policy/mean Max 0.999286 +trainer/policy/mean Min -0.998886 +trainer/policy/std Mean 0.410433 +trainer/policy/std Std 0.0203958 +trainer/policy/std Max 0.431281 +trainer/policy/std Min 0.379942 +trainer/Advantage Weights Mean 1.66533 +trainer/Advantage Weights Std 12.4124 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.2718e-15 +trainer/Advantage Score Mean -0.669848 +trainer/Advantage Score Std 0.532578 +trainer/Advantage Score Max 1.0723 +trainer/Advantage Score Min -3.27027 +trainer/V1 Predictions Mean -74.9098 +trainer/V1 Predictions Std 16.4536 +trainer/V1 Predictions Max -5.82787 +trainer/V1 Predictions Min -87.5589 +trainer/VF Loss 0.0813 +expl/num steps total 921000 +expl/num paths total 1267 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0122267 +expl/Actions Std 0.835122 +expl/Actions Max 2.32225 +expl/Actions Min -2.27374 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 829571 +eval/num paths total 926 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0111032 +eval/Actions Std 0.718247 +eval/Actions Max 0.999899 +eval/Actions Min -0.9998 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.67709e-06 +time/evaluation sampling (s) 3.34662 +time/exploration sampling (s) 4.67976 +time/logging (s) 0.00919315 +time/saving (s) 0.0128499 +time/training (s) 16.5484 +time/epoch (s) 24.5968 +time/total (s) 23779.9 +Epoch -80 +------------------------------ ---------------- +2022-05-16 00:39:15.117935 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -79 finished +------------------------------ ---------------- +epoch -79 +replay_buffer/size 999047 +trainer/num train calls 922000 +trainer/QF1 Loss 4.55285 +trainer/QF2 Loss 4.54343 +trainer/Policy Loss 16.0483 +trainer/Q1 Predictions Mean -75.6468 +trainer/Q1 Predictions Std 14.4657 +trainer/Q1 Predictions Max -1.29318 +trainer/Q1 Predictions Min -87.7559 +trainer/Q2 Predictions Mean -75.5763 +trainer/Q2 Predictions Std 14.5349 +trainer/Q2 Predictions Max -1.48865 +trainer/Q2 Predictions Min -87.7923 +trainer/Q Targets Mean -75.49 +trainer/Q Targets Std 14.7432 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0199 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0124236 +trainer/policy/mean Std 0.740219 +trainer/policy/mean Max 0.99911 +trainer/policy/mean Min -0.999498 +trainer/policy/std Mean 0.408893 +trainer/policy/std Std 0.0205725 +trainer/policy/std Max 0.432017 +trainer/policy/std Min 0.381538 +trainer/Advantage Weights Mean 2.48704 +trainer/Advantage Weights Std 12.1521 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.69544e-12 +trainer/Advantage Score Mean -0.337725 +trainer/Advantage Score Std 0.457354 +trainer/Advantage Score Max 0.693253 +trainer/Advantage Score Min -2.71031 +trainer/V1 Predictions Mean -75.3689 +trainer/V1 Predictions Std 14.6826 +trainer/V1 Predictions Max -1.16027 +trainer/V1 Predictions Min -87.8663 +trainer/VF Loss 0.0392069 +expl/num steps total 922000 +expl/num paths total 1269 +expl/path length Mean 500 +expl/path length Std 361 +expl/path length Max 861 +expl/path length Min 139 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0007839 +expl/Actions Std 0.825846 +expl/Actions Max 2.74824 +expl/Actions Min -2.27802 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 830571 +eval/num paths total 927 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.100181 +eval/Actions Std 0.626225 +eval/Actions Max 0.999636 +eval/Actions Min -0.999168 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.76277e-06 +time/evaluation sampling (s) 3.32738 +time/exploration sampling (s) 4.02491 +time/logging (s) 0.0110511 +time/saving (s) 0.0153074 +time/training (s) 16.1868 +time/epoch (s) 23.5654 +time/total (s) 23803.5 +Epoch -79 +------------------------------ ---------------- +2022-05-16 00:39:38.630211 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -78 finished +------------------------------ ---------------- +epoch -78 +replay_buffer/size 999047 +trainer/num train calls 923000 +trainer/QF1 Loss 2.36628 +trainer/QF2 Loss 2.24078 +trainer/Policy Loss 22.4975 +trainer/Q1 Predictions Mean -73.5871 +trainer/Q1 Predictions Std 18.356 +trainer/Q1 Predictions Max -1.04433 +trainer/Q1 Predictions Min -87.804 +trainer/Q2 Predictions Mean -73.592 +trainer/Q2 Predictions Std 18.3366 +trainer/Q2 Predictions Max -1.00599 +trainer/Q2 Predictions Min -87.5965 +trainer/Q Targets Mean -73.4841 +trainer/Q Targets Std 18.1929 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5709 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00243903 +trainer/policy/mean Std 0.72916 +trainer/policy/mean Max 0.999483 +trainer/policy/mean Min -0.999673 +trainer/policy/std Mean 0.410148 +trainer/policy/std Std 0.0192558 +trainer/policy/std Max 0.433246 +trainer/policy/std Min 0.383114 +trainer/Advantage Weights Mean 3.95309 +trainer/Advantage Weights Std 17.6655 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.41175e-18 +trainer/Advantage Score Mean -0.420068 +trainer/Advantage Score Std 0.538279 +trainer/Advantage Score Max 1.12172 +trainer/Advantage Score Min -4.05662 +trainer/V1 Predictions Mean -73.3563 +trainer/V1 Predictions Std 18.2731 +trainer/V1 Predictions Max -1.27747 +trainer/V1 Predictions Min -87.9485 +trainer/VF Loss 0.060328 +expl/num steps total 923000 +expl/num paths total 1270 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0436681 +expl/Actions Std 0.794368 +expl/Actions Max 2.31698 +expl/Actions Min -2.45386 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 831352 +eval/num paths total 928 +eval/path length Mean 781 +eval/path length Std 0 +eval/path length Max 781 +eval/path length Min 781 +eval/Rewards Mean 0.00128041 +eval/Rewards Std 0.0357599 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0252509 +eval/Actions Std 0.722354 +eval/Actions Max 0.999564 +eval/Actions Min -0.999313 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.94577e-06 +time/evaluation sampling (s) 3.21505 +time/exploration sampling (s) 4.50496 +time/logging (s) 0.00975614 +time/saving (s) 0.011341 +time/training (s) 15.7531 +time/epoch (s) 23.4942 +time/total (s) 23827 +Epoch -78 +------------------------------ ---------------- +2022-05-16 00:40:02.791539 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -77 finished +------------------------------ ---------------- +epoch -77 +replay_buffer/size 999047 +trainer/num train calls 924000 +trainer/QF1 Loss 0.920414 +trainer/QF2 Loss 0.700885 +trainer/Policy Loss 17.8009 +trainer/Q1 Predictions Mean -74.3941 +trainer/Q1 Predictions Std 17.6884 +trainer/Q1 Predictions Max -0.39874 +trainer/Q1 Predictions Min -88.1721 +trainer/Q2 Predictions Mean -74.4339 +trainer/Q2 Predictions Std 17.7517 +trainer/Q2 Predictions Max -0.24551 +trainer/Q2 Predictions Min -88.3166 +trainer/Q Targets Mean -74.5803 +trainer/Q Targets Std 18.102 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2607 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00162209 +trainer/policy/mean Std 0.735095 +trainer/policy/mean Max 0.999968 +trainer/policy/mean Min -0.999945 +trainer/policy/std Mean 0.407433 +trainer/policy/std Std 0.0196849 +trainer/policy/std Max 0.427659 +trainer/policy/std Min 0.378881 +trainer/Advantage Weights Mean 6.26935 +trainer/Advantage Weights Std 21.5975 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.92049e-40 +trainer/Advantage Score Mean -0.302287 +trainer/Advantage Score Std 0.793875 +trainer/Advantage Score Max 1.97002 +trainer/Advantage Score Min -9.00339 +trainer/V1 Predictions Mean -74.3982 +trainer/V1 Predictions Std 18.0977 +trainer/V1 Predictions Max -0.920731 +trainer/V1 Predictions Min -88.1284 +trainer/VF Loss 0.103238 +expl/num steps total 924000 +expl/num paths total 1272 +expl/path length Mean 500 +expl/path length Std 42 +expl/path length Max 542 +expl/path length Min 458 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.108217 +expl/Actions Std 0.849089 +expl/Actions Max 2.37331 +expl/Actions Min -2.33997 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 832352 +eval/num paths total 929 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.190581 +eval/Actions Std 0.712841 +eval/Actions Max 0.999988 +eval/Actions Min -0.999857 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.60584e-06 +time/evaluation sampling (s) 3.50113 +time/exploration sampling (s) 5.1057 +time/logging (s) 0.0114948 +time/saving (s) 0.0174872 +time/training (s) 15.507 +time/epoch (s) 24.1428 +time/total (s) 23851.1 +Epoch -77 +------------------------------ ---------------- +2022-05-16 00:40:26.614215 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -76 finished +------------------------------ ---------------- +epoch -76 +replay_buffer/size 999047 +trainer/num train calls 925000 +trainer/QF1 Loss 1.15076 +trainer/QF2 Loss 1.13986 +trainer/Policy Loss 5.04123 +trainer/Q1 Predictions Mean -73.6226 +trainer/Q1 Predictions Std 19.1437 +trainer/Q1 Predictions Max -1.46938 +trainer/Q1 Predictions Min -87.7735 +trainer/Q2 Predictions Mean -73.6258 +trainer/Q2 Predictions Std 19.1467 +trainer/Q2 Predictions Max -1.92796 +trainer/Q2 Predictions Min -88.0657 +trainer/Q Targets Mean -73.6465 +trainer/Q Targets Std 18.7071 +trainer/Q Targets Max -4.58147 +trainer/Q Targets Min -87.5967 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0106278 +trainer/policy/mean Std 0.738512 +trainer/policy/mean Max 0.99972 +trainer/policy/mean Min -0.999795 +trainer/policy/std Mean 0.406758 +trainer/policy/std Std 0.0198265 +trainer/policy/std Max 0.428121 +trainer/policy/std Min 0.378054 +trainer/Advantage Weights Mean 1.26669 +trainer/Advantage Weights Std 9.38895 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.62366e-16 +trainer/Advantage Score Mean -0.613479 +trainer/Advantage Score Std 0.580497 +trainer/Advantage Score Max 0.93214 +trainer/Advantage Score Min -3.46869 +trainer/V1 Predictions Mean -73.2931 +trainer/V1 Predictions Std 19.0957 +trainer/V1 Predictions Max -2.47579 +trainer/V1 Predictions Min -87.3887 +trainer/VF Loss 0.077923 +expl/num steps total 925000 +expl/num paths total 1274 +expl/path length Mean 500 +expl/path length Std 288 +expl/path length Max 788 +expl/path length Min 212 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0351577 +expl/Actions Std 0.807552 +expl/Actions Max 2.37619 +expl/Actions Min -2.21388 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 833352 +eval/num paths total 930 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00752795 +eval/Actions Std 0.734234 +eval/Actions Max 0.998976 +eval/Actions Min -0.999797 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.35975e-06 +time/evaluation sampling (s) 2.81418 +time/exploration sampling (s) 4.95654 +time/logging (s) 0.00944187 +time/saving (s) 0.0149951 +time/training (s) 16.0022 +time/epoch (s) 23.7974 +time/total (s) 23874.9 +Epoch -76 +------------------------------ ---------------- +2022-05-16 00:40:49.996095 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -75 finished +------------------------------ ---------------- +epoch -75 +replay_buffer/size 999047 +trainer/num train calls 926000 +trainer/QF1 Loss 7.85989 +trainer/QF2 Loss 8.04367 +trainer/Policy Loss 20.9155 +trainer/Q1 Predictions Mean -72.3993 +trainer/Q1 Predictions Std 20.361 +trainer/Q1 Predictions Max -0.178516 +trainer/Q1 Predictions Min -88.3503 +trainer/Q2 Predictions Mean -72.4746 +trainer/Q2 Predictions Std 20.3787 +trainer/Q2 Predictions Max 0.0200111 +trainer/Q2 Predictions Min -88.2085 +trainer/Q Targets Mean -72.6623 +trainer/Q Targets Std 19.4984 +trainer/Q Targets Max -3.74839 +trainer/Q Targets Min -87.8793 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0121365 +trainer/policy/mean Std 0.733865 +trainer/policy/mean Max 0.999769 +trainer/policy/mean Min -0.999796 +trainer/policy/std Mean 0.407238 +trainer/policy/std Std 0.0207281 +trainer/policy/std Max 0.429334 +trainer/policy/std Min 0.378205 +trainer/Advantage Weights Mean 5.52502 +trainer/Advantage Weights Std 21.8223 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.40676e-16 +trainer/Advantage Score Mean -0.484431 +trainer/Advantage Score Std 0.666881 +trainer/Advantage Score Max 2.01992 +trainer/Advantage Score Min -3.65001 +trainer/V1 Predictions Mean -72.1917 +trainer/V1 Predictions Std 20.1093 +trainer/V1 Predictions Max -0.511011 +trainer/V1 Predictions Min -87.8173 +trainer/VF Loss 0.116335 +expl/num steps total 926000 +expl/num paths total 1275 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0201836 +expl/Actions Std 0.801999 +expl/Actions Max 2.35402 +expl/Actions Min -2.53643 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 834352 +eval/num paths total 931 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.308377 +eval/Actions Std 0.646907 +eval/Actions Max 0.999393 +eval/Actions Min -0.999374 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.58721e-06 +time/evaluation sampling (s) 3.49119 +time/exploration sampling (s) 3.81338 +time/logging (s) 0.00704209 +time/saving (s) 0.0150781 +time/training (s) 16.0385 +time/epoch (s) 23.3652 +time/total (s) 23898.3 +Epoch -75 +------------------------------ ---------------- +2022-05-16 00:41:14.066209 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -74 finished +------------------------------ ---------------- +epoch -74 +replay_buffer/size 999047 +trainer/num train calls 927000 +trainer/QF1 Loss 0.838212 +trainer/QF2 Loss 0.650577 +trainer/Policy Loss 11.1511 +trainer/Q1 Predictions Mean -75.5074 +trainer/Q1 Predictions Std 18.3316 +trainer/Q1 Predictions Max -0.344578 +trainer/Q1 Predictions Min -88.6586 +trainer/Q2 Predictions Mean -75.4531 +trainer/Q2 Predictions Std 18.3573 +trainer/Q2 Predictions Max 0.0374808 +trainer/Q2 Predictions Min -88.421 +trainer/Q Targets Mean -75.5563 +trainer/Q Targets Std 18.1513 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8748 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00462845 +trainer/policy/mean Std 0.728856 +trainer/policy/mean Max 0.998729 +trainer/policy/mean Min -0.999279 +trainer/policy/std Mean 0.407794 +trainer/policy/std Std 0.021593 +trainer/policy/std Max 0.431102 +trainer/policy/std Min 0.373114 +trainer/Advantage Weights Mean 3.41752 +trainer/Advantage Weights Std 15.8898 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.24782e-14 +trainer/Advantage Score Mean -0.364633 +trainer/Advantage Score Std 0.533029 +trainer/Advantage Score Max 2.09756 +trainer/Advantage Score Min -3.20148 +trainer/V1 Predictions Mean -75.3152 +trainer/V1 Predictions Std 18.2329 +trainer/V1 Predictions Max -0.190697 +trainer/V1 Predictions Min -88.0089 +trainer/VF Loss 0.0870265 +expl/num steps total 927000 +expl/num paths total 1276 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0616927 +expl/Actions Std 0.820254 +expl/Actions Max 2.26223 +expl/Actions Min -2.25025 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 834966 +eval/num paths total 932 +eval/path length Mean 614 +eval/path length Std 0 +eval/path length Max 614 +eval/path length Min 614 +eval/Rewards Mean 0.00162866 +eval/Rewards Std 0.0403238 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.020959 +eval/Actions Std 0.744831 +eval/Actions Max 0.999319 +eval/Actions Min -0.999709 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.83401e-06 +time/evaluation sampling (s) 3.58832 +time/exploration sampling (s) 4.43288 +time/logging (s) 0.00784246 +time/saving (s) 0.0152032 +time/training (s) 16.0135 +time/epoch (s) 24.0577 +time/total (s) 23922.3 +Epoch -74 +------------------------------ ---------------- +2022-05-16 00:41:37.676007 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -73 finished +------------------------------ ---------------- +epoch -73 +replay_buffer/size 999047 +trainer/num train calls 928000 +trainer/QF1 Loss 0.597375 +trainer/QF2 Loss 0.629615 +trainer/Policy Loss 17.7305 +trainer/Q1 Predictions Mean -74.2843 +trainer/Q1 Predictions Std 18.5186 +trainer/Q1 Predictions Max 0.144867 +trainer/Q1 Predictions Min -87.3087 +trainer/Q2 Predictions Mean -74.3328 +trainer/Q2 Predictions Std 18.5074 +trainer/Q2 Predictions Max -0.946834 +trainer/Q2 Predictions Min -87.2647 +trainer/Q Targets Mean -74.2857 +trainer/Q Targets Std 18.312 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4543 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0144324 +trainer/policy/mean Std 0.72594 +trainer/policy/mean Max 0.998544 +trainer/policy/mean Min -0.999777 +trainer/policy/std Mean 0.406516 +trainer/policy/std Std 0.0203542 +trainer/policy/std Max 0.426577 +trainer/policy/std Min 0.371955 +trainer/Advantage Weights Mean 3.62244 +trainer/Advantage Weights Std 15.876 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.10783e-21 +trainer/Advantage Score Mean -0.46864 +trainer/Advantage Score Std 0.618563 +trainer/Advantage Score Max 1.17914 +trainer/Advantage Score Min -4.63931 +trainer/V1 Predictions Mean -73.9863 +trainer/V1 Predictions Std 18.5722 +trainer/V1 Predictions Max -1.53693 +trainer/V1 Predictions Min -87.3685 +trainer/VF Loss 0.0735682 +expl/num steps total 928000 +expl/num paths total 1277 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0341406 +expl/Actions Std 0.835232 +expl/Actions Max 2.35175 +expl/Actions Min -2.26834 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 835966 +eval/num paths total 933 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0585239 +eval/Actions Std 0.734076 +eval/Actions Max 0.999147 +eval/Actions Min -0.997784 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.89087e-06 +time/evaluation sampling (s) 3.27953 +time/exploration sampling (s) 4.85092 +time/logging (s) 0.0115519 +time/saving (s) 0.0148944 +time/training (s) 15.4389 +time/epoch (s) 23.5958 +time/total (s) 23945.9 +Epoch -73 +------------------------------ ---------------- +2022-05-16 00:42:01.977878 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -72 finished +------------------------------ ---------------- +epoch -72 +replay_buffer/size 999047 +trainer/num train calls 929000 +trainer/QF1 Loss 0.914174 +trainer/QF2 Loss 0.921821 +trainer/Policy Loss 17.4738 +trainer/Q1 Predictions Mean -73.0385 +trainer/Q1 Predictions Std 19.6225 +trainer/Q1 Predictions Max -0.203201 +trainer/Q1 Predictions Min -87.9438 +trainer/Q2 Predictions Mean -72.9166 +trainer/Q2 Predictions Std 19.6209 +trainer/Q2 Predictions Max 0.657456 +trainer/Q2 Predictions Min -87.72 +trainer/Q Targets Mean -73.2201 +trainer/Q Targets Std 20.027 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2355 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0483761 +trainer/policy/mean Std 0.743722 +trainer/policy/mean Max 0.998911 +trainer/policy/mean Min -0.999496 +trainer/policy/std Mean 0.406107 +trainer/policy/std Std 0.0197207 +trainer/policy/std Max 0.427489 +trainer/policy/std Min 0.375366 +trainer/Advantage Weights Mean 3.66488 +trainer/Advantage Weights Std 16.04 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.48645e-18 +trainer/Advantage Score Mean -0.435256 +trainer/Advantage Score Std 0.697981 +trainer/Advantage Score Max 0.967262 +trainer/Advantage Score Min -3.95768 +trainer/V1 Predictions Mean -72.9891 +trainer/V1 Predictions Std 20.0074 +trainer/V1 Predictions Max 0.563726 +trainer/V1 Predictions Min -88.0351 +trainer/VF Loss 0.0797625 +expl/num steps total 929000 +expl/num paths total 1279 +expl/path length Mean 500 +expl/path length Std 221 +expl/path length Max 721 +expl/path length Min 279 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0286522 +expl/Actions Std 0.822901 +expl/Actions Max 2.28893 +expl/Actions Min -2.06068 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 836966 +eval/num paths total 934 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.297815 +eval/Actions Std 0.636211 +eval/Actions Max 0.999986 +eval/Actions Min -0.999871 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.97185e-06 +time/evaluation sampling (s) 3.55321 +time/exploration sampling (s) 4.48562 +time/logging (s) 0.00759398 +time/saving (s) 0.0110548 +time/training (s) 16.2268 +time/epoch (s) 24.2843 +time/total (s) 23970.2 +Epoch -72 +------------------------------ ---------------- +2022-05-16 00:42:26.320103 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -71 finished +------------------------------ ---------------- +epoch -71 +replay_buffer/size 999047 +trainer/num train calls 930000 +trainer/QF1 Loss 0.481121 +trainer/QF2 Loss 0.598835 +trainer/Policy Loss 31.7466 +trainer/Q1 Predictions Mean -75.3835 +trainer/Q1 Predictions Std 16.7162 +trainer/Q1 Predictions Max -2.51639 +trainer/Q1 Predictions Min -87.7212 +trainer/Q2 Predictions Mean -75.3191 +trainer/Q2 Predictions Std 16.6254 +trainer/Q2 Predictions Max -3.26813 +trainer/Q2 Predictions Min -87.3906 +trainer/Q Targets Mean -75.6689 +trainer/Q Targets Std 16.6242 +trainer/Q Targets Max -2.74922 +trainer/Q Targets Min -87.6597 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0187693 +trainer/policy/mean Std 0.739282 +trainer/policy/mean Max 0.998911 +trainer/policy/mean Min -0.999266 +trainer/policy/std Mean 0.408356 +trainer/policy/std Std 0.0180102 +trainer/policy/std Max 0.426669 +trainer/policy/std Min 0.383552 +trainer/Advantage Weights Mean 8.09679 +trainer/Advantage Weights Std 23.7047 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.21906e-13 +trainer/Advantage Score Mean -0.164302 +trainer/Advantage Score Std 0.520466 +trainer/Advantage Score Max 1.85637 +trainer/Advantage Score Min -2.97355 +trainer/V1 Predictions Mean -75.4703 +trainer/V1 Predictions Std 16.6239 +trainer/V1 Predictions Max -2.46863 +trainer/V1 Predictions Min -87.4991 +trainer/VF Loss 0.0714164 +expl/num steps total 930000 +expl/num paths total 1281 +expl/path length Mean 500 +expl/path length Std 296 +expl/path length Max 796 +expl/path length Min 204 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0150953 +expl/Actions Std 0.834981 +expl/Actions Max 2.47054 +expl/Actions Min -2.27176 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 837966 +eval/num paths total 935 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.165242 +eval/Actions Std 0.763016 +eval/Actions Max 0.999996 +eval/Actions Min -0.999403 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.06297e-05 +time/evaluation sampling (s) 2.87177 +time/exploration sampling (s) 5.41415 +time/logging (s) 0.00826579 +time/saving (s) 0.0116469 +time/training (s) 16.0235 +time/epoch (s) 24.3294 +time/total (s) 23994.6 +Epoch -71 +------------------------------ ---------------- +2022-05-16 00:42:50.134458 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -70 finished +------------------------------ ---------------- +epoch -70 +replay_buffer/size 999047 +trainer/num train calls 931000 +trainer/QF1 Loss 0.571877 +trainer/QF2 Loss 0.604971 +trainer/Policy Loss 21.7667 +trainer/Q1 Predictions Mean -76.5084 +trainer/Q1 Predictions Std 15.4035 +trainer/Q1 Predictions Max -2.81892 +trainer/Q1 Predictions Min -88.283 +trainer/Q2 Predictions Mean -76.4671 +trainer/Q2 Predictions Std 15.4793 +trainer/Q2 Predictions Max -3.14566 +trainer/Q2 Predictions Min -88.2843 +trainer/Q Targets Mean -76.6709 +trainer/Q Targets Std 15.4114 +trainer/Q Targets Max -3.57595 +trainer/Q Targets Min -88.194 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0166633 +trainer/policy/mean Std 0.740603 +trainer/policy/mean Max 0.999717 +trainer/policy/mean Min -0.999069 +trainer/policy/std Mean 0.406806 +trainer/policy/std Std 0.0174595 +trainer/policy/std Max 0.425003 +trainer/policy/std Min 0.382097 +trainer/Advantage Weights Mean 5.60447 +trainer/Advantage Weights Std 19.2304 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.47568e-13 +trainer/Advantage Score Mean -0.23257 +trainer/Advantage Score Std 0.527274 +trainer/Advantage Score Max 1.89822 +trainer/Advantage Score Min -2.95445 +trainer/V1 Predictions Mean -76.428 +trainer/V1 Predictions Std 15.4366 +trainer/V1 Predictions Max -2.69952 +trainer/V1 Predictions Min -88.2419 +trainer/VF Loss 0.0631389 +expl/num steps total 931000 +expl/num paths total 1282 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0199482 +expl/Actions Std 0.855044 +expl/Actions Max 2.25431 +expl/Actions Min -2.37698 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 838433 +eval/num paths total 936 +eval/path length Mean 467 +eval/path length Std 0 +eval/path length Max 467 +eval/path length Min 467 +eval/Rewards Mean 0.00214133 +eval/Rewards Std 0.0462249 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0623655 +eval/Actions Std 0.740616 +eval/Actions Max 0.999713 +eval/Actions Min -0.999668 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.82004e-06 +time/evaluation sampling (s) 3.4626 +time/exploration sampling (s) 4.37156 +time/logging (s) 0.00792715 +time/saving (s) 0.0152708 +time/training (s) 15.9424 +time/epoch (s) 23.7998 +time/total (s) 24018.4 +Epoch -70 +------------------------------ ---------------- +2022-05-16 00:43:14.046989 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -69 finished +------------------------------ ---------------- +epoch -69 +replay_buffer/size 999047 +trainer/num train calls 932000 +trainer/QF1 Loss 0.664275 +trainer/QF2 Loss 0.66269 +trainer/Policy Loss 15.3831 +trainer/Q1 Predictions Mean -74.3358 +trainer/Q1 Predictions Std 18.2769 +trainer/Q1 Predictions Max -1.25371 +trainer/Q1 Predictions Min -87.6624 +trainer/Q2 Predictions Mean -74.3008 +trainer/Q2 Predictions Std 18.2835 +trainer/Q2 Predictions Max -1.48753 +trainer/Q2 Predictions Min -87.4887 +trainer/Q Targets Mean -74.6734 +trainer/Q Targets Std 18.1863 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.4987 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00902389 +trainer/policy/mean Std 0.736899 +trainer/policy/mean Max 0.9992 +trainer/policy/mean Min -0.999874 +trainer/policy/std Mean 0.407109 +trainer/policy/std Std 0.0191595 +trainer/policy/std Max 0.428532 +trainer/policy/std Min 0.379924 +trainer/Advantage Weights Mean 5.20975 +trainer/Advantage Weights Std 19.8616 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.66689e-23 +trainer/Advantage Score Mean -0.371913 +trainer/Advantage Score Std 0.634386 +trainer/Advantage Score Max 1.14424 +trainer/Advantage Score Min -5.09225 +trainer/V1 Predictions Mean -74.3984 +trainer/V1 Predictions Std 18.235 +trainer/V1 Predictions Max -1.85469 +trainer/V1 Predictions Min -87.501 +trainer/VF Loss 0.0762711 +expl/num steps total 932000 +expl/num paths total 1283 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0201052 +expl/Actions Std 0.821067 +expl/Actions Max 2.53817 +expl/Actions Min -2.46503 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 839433 +eval/num paths total 937 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0117151 +eval/Actions Std 0.747751 +eval/Actions Max 0.999802 +eval/Actions Min -0.999837 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.35281e-06 +time/evaluation sampling (s) 3.52701 +time/exploration sampling (s) 4.41593 +time/logging (s) 0.00873392 +time/saving (s) 0.0180845 +time/training (s) 15.9322 +time/epoch (s) 23.902 +time/total (s) 24042.3 +Epoch -69 +------------------------------ ---------------- +2022-05-16 00:43:38.595563 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -68 finished +------------------------------ ---------------- +epoch -68 +replay_buffer/size 999047 +trainer/num train calls 933000 +trainer/QF1 Loss 0.972642 +trainer/QF2 Loss 1.01603 +trainer/Policy Loss 17.8453 +trainer/Q1 Predictions Mean -75.3854 +trainer/Q1 Predictions Std 16.9921 +trainer/Q1 Predictions Max -1.53072 +trainer/Q1 Predictions Min -89.5854 +trainer/Q2 Predictions Mean -75.3448 +trainer/Q2 Predictions Std 17.0448 +trainer/Q2 Predictions Max -0.150222 +trainer/Q2 Predictions Min -89.5048 +trainer/Q Targets Mean -74.8629 +trainer/Q Targets Std 17.0675 +trainer/Q Targets Max -2.73258 +trainer/Q Targets Min -88.2131 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0202288 +trainer/policy/mean Std 0.73021 +trainer/policy/mean Max 0.999178 +trainer/policy/mean Min -0.998845 +trainer/policy/std Mean 0.408195 +trainer/policy/std Std 0.0201804 +trainer/policy/std Max 0.431648 +trainer/policy/std Min 0.377636 +trainer/Advantage Weights Mean 4.53397 +trainer/Advantage Weights Std 17.6033 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.97869e-22 +trainer/Advantage Score Mean -0.376525 +trainer/Advantage Score Std 0.64485 +trainer/Advantage Score Max 1.40058 +trainer/Advantage Score Min -4.90517 +trainer/V1 Predictions Mean -74.6532 +trainer/V1 Predictions Std 17.2071 +trainer/V1 Predictions Max -2.40263 +trainer/V1 Predictions Min -88.0313 +trainer/VF Loss 0.0783023 +expl/num steps total 933000 +expl/num paths total 1285 +expl/path length Mean 500 +expl/path length Std 299 +expl/path length Max 799 +expl/path length Min 201 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.029848 +expl/Actions Std 0.821863 +expl/Actions Max 2.2781 +expl/Actions Min -2.40034 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 839979 +eval/num paths total 938 +eval/path length Mean 546 +eval/path length Std 0 +eval/path length Max 546 +eval/path length Min 546 +eval/Rewards Mean 0.0018315 +eval/Rewards Std 0.0427568 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0313032 +eval/Actions Std 0.722544 +eval/Actions Max 0.999292 +eval/Actions Min -0.999851 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.50596e-06 +time/evaluation sampling (s) 3.63861 +time/exploration sampling (s) 4.9752 +time/logging (s) 0.00586506 +time/saving (s) 0.00950724 +time/training (s) 15.9034 +time/epoch (s) 24.5326 +time/total (s) 24066.8 +Epoch -68 +------------------------------ ---------------- +2022-05-16 00:44:02.417770 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -67 finished +------------------------------ ---------------- +epoch -67 +replay_buffer/size 999047 +trainer/num train calls 934000 +trainer/QF1 Loss 0.601635 +trainer/QF2 Loss 0.662075 +trainer/Policy Loss 20.1837 +trainer/Q1 Predictions Mean -74.1194 +trainer/Q1 Predictions Std 17.3194 +trainer/Q1 Predictions Max -1.66937 +trainer/Q1 Predictions Min -88.278 +trainer/Q2 Predictions Mean -74.2378 +trainer/Q2 Predictions Std 17.3238 +trainer/Q2 Predictions Max -0.72581 +trainer/Q2 Predictions Min -87.9699 +trainer/Q Targets Mean -74.2454 +trainer/Q Targets Std 17.2875 +trainer/Q Targets Max -2.14002 +trainer/Q Targets Min -87.8012 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -5.46884e-05 +trainer/policy/mean Std 0.725201 +trainer/policy/mean Max 0.998829 +trainer/policy/mean Min -0.998747 +trainer/policy/std Mean 0.406268 +trainer/policy/std Std 0.0198989 +trainer/policy/std Max 0.427749 +trainer/policy/std Min 0.376675 +trainer/Advantage Weights Mean 3.50263 +trainer/Advantage Weights Std 14.65 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.04106e-13 +trainer/Advantage Score Mean -0.419632 +trainer/Advantage Score Std 0.601614 +trainer/Advantage Score Max 2.22838 +trainer/Advantage Score Min -2.8316 +trainer/V1 Predictions Mean -73.9255 +trainer/V1 Predictions Std 17.4784 +trainer/V1 Predictions Max -0.185567 +trainer/V1 Predictions Min -88.0574 +trainer/VF Loss 0.0780146 +expl/num steps total 934000 +expl/num paths total 1286 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.00674812 +expl/Actions Std 0.786201 +expl/Actions Max 2.30744 +expl/Actions Min -2.57389 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 840403 +eval/num paths total 939 +eval/path length Mean 424 +eval/path length Std 0 +eval/path length Max 424 +eval/path length Min 424 +eval/Rewards Mean 0.00235849 +eval/Rewards Std 0.048507 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0354638 +eval/Actions Std 0.746456 +eval/Actions Max 0.999934 +eval/Actions Min -0.999188 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.09804e-06 +time/evaluation sampling (s) 3.00356 +time/exploration sampling (s) 5.0479 +time/logging (s) 0.00984227 +time/saving (s) 0.0159368 +time/training (s) 15.7378 +time/epoch (s) 23.8151 +time/total (s) 24090.6 +Epoch -67 +------------------------------ ---------------- +2022-05-16 00:44:25.714304 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -66 finished +------------------------------ ---------------- +epoch -66 +replay_buffer/size 999047 +trainer/num train calls 935000 +trainer/QF1 Loss 0.426299 +trainer/QF2 Loss 0.480922 +trainer/Policy Loss 10.805 +trainer/Q1 Predictions Mean -75.3033 +trainer/Q1 Predictions Std 17.536 +trainer/Q1 Predictions Max -0.674356 +trainer/Q1 Predictions Min -87.6785 +trainer/Q2 Predictions Mean -75.234 +trainer/Q2 Predictions Std 17.5467 +trainer/Q2 Predictions Max -1.04259 +trainer/Q2 Predictions Min -87.5676 +trainer/Q Targets Mean -75.5132 +trainer/Q Targets Std 17.4626 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1358 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0163279 +trainer/policy/mean Std 0.738275 +trainer/policy/mean Max 0.999724 +trainer/policy/mean Min -0.998738 +trainer/policy/std Mean 0.406897 +trainer/policy/std Std 0.0204123 +trainer/policy/std Max 0.427983 +trainer/policy/std Min 0.374984 +trainer/Advantage Weights Mean 3.59189 +trainer/Advantage Weights Std 16.2202 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.43469e-17 +trainer/Advantage Score Mean -0.33498 +trainer/Advantage Score Std 0.582339 +trainer/Advantage Score Max 0.787201 +trainer/Advantage Score Min -3.8783 +trainer/V1 Predictions Mean -75.2813 +trainer/V1 Predictions Std 17.6276 +trainer/V1 Predictions Max -1.21333 +trainer/V1 Predictions Min -87.9548 +trainer/VF Loss 0.0561459 +expl/num steps total 935000 +expl/num paths total 1287 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0287811 +expl/Actions Std 0.859922 +expl/Actions Max 2.67228 +expl/Actions Min -2.24871 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 841206 +eval/num paths total 940 +eval/path length Mean 803 +eval/path length Std 0 +eval/path length Max 803 +eval/path length Min 803 +eval/Rewards Mean 0.00124533 +eval/Rewards Std 0.0352673 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0236407 +eval/Actions Std 0.753199 +eval/Actions Max 0.999748 +eval/Actions Min -0.999631 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.72226e-06 +time/evaluation sampling (s) 3.31112 +time/exploration sampling (s) 4.08419 +time/logging (s) 0.00659327 +time/saving (s) 0.00956345 +time/training (s) 15.8634 +time/epoch (s) 23.2748 +time/total (s) 24113.9 +Epoch -66 +------------------------------ ---------------- +2022-05-16 00:44:50.031079 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -65 finished +------------------------------ ---------------- +epoch -65 +replay_buffer/size 999047 +trainer/num train calls 936000 +trainer/QF1 Loss 0.727772 +trainer/QF2 Loss 0.724389 +trainer/Policy Loss 30.7212 +trainer/Q1 Predictions Mean -73.9775 +trainer/Q1 Predictions Std 18.7785 +trainer/Q1 Predictions Max -0.90338 +trainer/Q1 Predictions Min -87.7853 +trainer/Q2 Predictions Mean -73.8794 +trainer/Q2 Predictions Std 18.6968 +trainer/Q2 Predictions Max -0.31223 +trainer/Q2 Predictions Min -87.6612 +trainer/Q Targets Mean -73.9491 +trainer/Q Targets Std 18.6497 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3277 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0242759 +trainer/policy/mean Std 0.740711 +trainer/policy/mean Max 0.999945 +trainer/policy/mean Min -0.999591 +trainer/policy/std Mean 0.407829 +trainer/policy/std Std 0.019048 +trainer/policy/std Max 0.42921 +trainer/policy/std Min 0.37918 +trainer/Advantage Weights Mean 8.27292 +trainer/Advantage Weights Std 22.6866 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.79939e-09 +trainer/Advantage Score Mean -0.147612 +trainer/Advantage Score Std 0.475459 +trainer/Advantage Score Max 1.95451 +trainer/Advantage Score Min -1.91548 +trainer/V1 Predictions Mean -73.6946 +trainer/V1 Predictions Std 18.7134 +trainer/V1 Predictions Max -1.11341 +trainer/V1 Predictions Min -87.271 +trainer/VF Loss 0.0675899 +expl/num steps total 936000 +expl/num paths total 1288 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0385389 +expl/Actions Std 0.821196 +expl/Actions Max 2.42836 +expl/Actions Min -2.35156 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 842206 +eval/num paths total 941 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.184354 +eval/Actions Std 0.653328 +eval/Actions Max 0.999582 +eval/Actions Min -0.999381 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.92901e-06 +time/evaluation sampling (s) 3.78627 +time/exploration sampling (s) 4.34728 +time/logging (s) 0.00749652 +time/saving (s) 0.0103094 +time/training (s) 16.1534 +time/epoch (s) 24.3047 +time/total (s) 24138.2 +Epoch -65 +------------------------------ ---------------- +2022-05-16 00:45:13.678636 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -64 finished +------------------------------ ---------------- +epoch -64 +replay_buffer/size 999047 +trainer/num train calls 937000 +trainer/QF1 Loss 0.608138 +trainer/QF2 Loss 0.529377 +trainer/Policy Loss 18.71 +trainer/Q1 Predictions Mean -75.146 +trainer/Q1 Predictions Std 15.7722 +trainer/Q1 Predictions Max -3.95665 +trainer/Q1 Predictions Min -87.6544 +trainer/Q2 Predictions Mean -75.1138 +trainer/Q2 Predictions Std 15.7488 +trainer/Q2 Predictions Max -3.87689 +trainer/Q2 Predictions Min -87.7733 +trainer/Q Targets Mean -75.1361 +trainer/Q Targets Std 15.9349 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9976 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0227674 +trainer/policy/mean Std 0.730078 +trainer/policy/mean Max 0.999247 +trainer/policy/mean Min -0.999599 +trainer/policy/std Mean 0.408856 +trainer/policy/std Std 0.0185061 +trainer/policy/std Max 0.431474 +trainer/policy/std Min 0.382147 +trainer/Advantage Weights Mean 3.94818 +trainer/Advantage Weights Std 15.0968 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.33208e-18 +trainer/Advantage Score Mean -0.393279 +trainer/Advantage Score Std 0.662216 +trainer/Advantage Score Max 0.742195 +trainer/Advantage Score Min -4.11598 +trainer/V1 Predictions Mean -74.8155 +trainer/V1 Predictions Std 16.1686 +trainer/V1 Predictions Max -2.87308 +trainer/V1 Predictions Min -87.6808 +trainer/VF Loss 0.0714688 +expl/num steps total 937000 +expl/num paths total 1290 +expl/path length Mean 500 +expl/path length Std 379 +expl/path length Max 879 +expl/path length Min 121 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0300603 +expl/Actions Std 0.81894 +expl/Actions Max 2.22624 +expl/Actions Min -2.27751 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 842866 +eval/num paths total 942 +eval/path length Mean 660 +eval/path length Std 0 +eval/path length Max 660 +eval/path length Min 660 +eval/Rewards Mean 0.00151515 +eval/Rewards Std 0.0388954 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0275517 +eval/Actions Std 0.730181 +eval/Actions Max 0.999598 +eval/Actions Min -0.999566 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.69664e-06 +time/evaluation sampling (s) 3.20386 +time/exploration sampling (s) 4.49818 +time/logging (s) 0.00888665 +time/saving (s) 0.0099726 +time/training (s) 15.9149 +time/epoch (s) 23.6358 +time/total (s) 24161.9 +Epoch -64 +------------------------------ ---------------- +2022-05-16 00:45:37.744835 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -63 finished +------------------------------ ---------------- +epoch -63 +replay_buffer/size 999047 +trainer/num train calls 938000 +trainer/QF1 Loss 1.44464 +trainer/QF2 Loss 1.47966 +trainer/Policy Loss 11.88 +trainer/Q1 Predictions Mean -72.4161 +trainer/Q1 Predictions Std 20.9127 +trainer/Q1 Predictions Max 0.405456 +trainer/Q1 Predictions Min -88.3328 +trainer/Q2 Predictions Mean -72.4029 +trainer/Q2 Predictions Std 20.9279 +trainer/Q2 Predictions Max -0.422159 +trainer/Q2 Predictions Min -88.4593 +trainer/Q Targets Mean -72.0565 +trainer/Q Targets Std 20.7816 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.989 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00950782 +trainer/policy/mean Std 0.735591 +trainer/policy/mean Max 0.998213 +trainer/policy/mean Min -0.999356 +trainer/policy/std Mean 0.408737 +trainer/policy/std Std 0.0191059 +trainer/policy/std Max 0.428034 +trainer/policy/std Min 0.379738 +trainer/Advantage Weights Mean 4.0014 +trainer/Advantage Weights Std 18.6786 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.00095e-27 +trainer/Advantage Score Mean -0.600459 +trainer/Advantage Score Std 0.715267 +trainer/Advantage Score Max 1.62256 +trainer/Advantage Score Min -6.21688 +trainer/V1 Predictions Mean -71.7565 +trainer/V1 Predictions Std 21.0255 +trainer/V1 Predictions Max 0.891802 +trainer/V1 Predictions Min -87.8612 +trainer/VF Loss 0.123525 +expl/num steps total 938000 +expl/num paths total 1291 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0250193 +expl/Actions Std 0.819082 +expl/Actions Max 2.62741 +expl/Actions Min -2.30473 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 843866 +eval/num paths total 943 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0847721 +eval/Actions Std 0.743758 +eval/Actions Max 0.999892 +eval/Actions Min -0.999678 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.45729e-06 +time/evaluation sampling (s) 3.62458 +time/exploration sampling (s) 4.47418 +time/logging (s) 0.00768433 +time/saving (s) 0.0121696 +time/training (s) 15.9332 +time/epoch (s) 24.0518 +time/total (s) 24185.9 +Epoch -63 +------------------------------ ---------------- +2022-05-16 00:46:01.383749 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -62 finished +------------------------------ ---------------- +epoch -62 +replay_buffer/size 999047 +trainer/num train calls 939000 +trainer/QF1 Loss 4.35082 +trainer/QF2 Loss 4.0301 +trainer/Policy Loss 35.4673 +trainer/Q1 Predictions Mean -73.8164 +trainer/Q1 Predictions Std 17.7376 +trainer/Q1 Predictions Max -0.576652 +trainer/Q1 Predictions Min -88.2968 +trainer/Q2 Predictions Mean -73.8519 +trainer/Q2 Predictions Std 17.8014 +trainer/Q2 Predictions Max -0.929928 +trainer/Q2 Predictions Min -88.3249 +trainer/Q Targets Mean -73.7823 +trainer/Q Targets Std 17.3241 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8212 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0112148 +trainer/policy/mean Std 0.731847 +trainer/policy/mean Max 0.99959 +trainer/policy/mean Min -0.999622 +trainer/policy/std Mean 0.407277 +trainer/policy/std Std 0.0199089 +trainer/policy/std Max 0.428645 +trainer/policy/std Min 0.38058 +trainer/Advantage Weights Mean 7.44002 +trainer/Advantage Weights Std 24.0528 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.14195e-33 +trainer/Advantage Score Mean -0.317697 +trainer/Advantage Score Std 0.721668 +trainer/Advantage Score Max 1.67334 +trainer/Advantage Score Min -7.58526 +trainer/V1 Predictions Mean -73.3728 +trainer/V1 Predictions Std 17.7424 +trainer/V1 Predictions Max -0.689571 +trainer/V1 Predictions Min -87.8697 +trainer/VF Loss 0.112003 +expl/num steps total 939000 +expl/num paths total 1293 +expl/path length Mean 500 +expl/path length Std 148 +expl/path length Max 648 +expl/path length Min 352 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0323166 +expl/Actions Std 0.829679 +expl/Actions Max 2.32434 +expl/Actions Min -2.16857 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 844409 +eval/num paths total 944 +eval/path length Mean 543 +eval/path length Std 0 +eval/path length Max 543 +eval/path length Min 543 +eval/Rewards Mean 0.00184162 +eval/Rewards Std 0.0428746 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.048 +eval/Actions Std 0.728534 +eval/Actions Max 0.999803 +eval/Actions Min -0.999358 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 8.86898e-06 +time/evaluation sampling (s) 3.16781 +time/exploration sampling (s) 4.70445 +time/logging (s) 0.00603942 +time/saving (s) 0.0122725 +time/training (s) 15.7319 +time/epoch (s) 23.6225 +time/total (s) 24209.5 +Epoch -62 +------------------------------ ---------------- +2022-05-16 00:46:24.817255 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -61 finished +------------------------------ ---------------- +epoch -61 +replay_buffer/size 999047 +trainer/num train calls 940000 +trainer/QF1 Loss 1.11911 +trainer/QF2 Loss 1.2014 +trainer/Policy Loss 17.0714 +trainer/Q1 Predictions Mean -76.066 +trainer/Q1 Predictions Std 16.1796 +trainer/Q1 Predictions Max -1.85397 +trainer/Q1 Predictions Min -88.6416 +trainer/Q2 Predictions Mean -76.1099 +trainer/Q2 Predictions Std 16.101 +trainer/Q2 Predictions Max -3.4578 +trainer/Q2 Predictions Min -88.4164 +trainer/Q Targets Mean -76.2373 +trainer/Q Targets Std 16.1466 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0809 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00603441 +trainer/policy/mean Std 0.733634 +trainer/policy/mean Max 0.998526 +trainer/policy/mean Min -0.99766 +trainer/policy/std Mean 0.403849 +trainer/policy/std Std 0.0191981 +trainer/policy/std Max 0.425031 +trainer/policy/std Min 0.373538 +trainer/Advantage Weights Mean 3.75638 +trainer/Advantage Weights Std 16.3934 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.29668e-23 +trainer/Advantage Score Mean -0.323353 +trainer/Advantage Score Std 0.50337 +trainer/Advantage Score Max 1.00799 +trainer/Advantage Score Min -5.2128 +trainer/V1 Predictions Mean -75.9537 +trainer/V1 Predictions Std 16.3707 +trainer/V1 Predictions Max -0.179868 +trainer/V1 Predictions Min -87.8558 +trainer/VF Loss 0.0503959 +expl/num steps total 940000 +expl/num paths total 1294 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0504666 +expl/Actions Std 0.842074 +expl/Actions Max 2.21501 +expl/Actions Min -2.45049 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 845409 +eval/num paths total 945 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.162364 +eval/Actions Std 0.540383 +eval/Actions Max 0.998592 +eval/Actions Min -0.999651 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.81306e-06 +time/evaluation sampling (s) 3.17318 +time/exploration sampling (s) 4.34581 +time/logging (s) 0.0105182 +time/saving (s) 0.015164 +time/training (s) 15.8799 +time/epoch (s) 23.4246 +time/total (s) 24233 +Epoch -61 +------------------------------ ---------------- +2022-05-16 00:46:48.874964 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -60 finished +------------------------------ ---------------- +epoch -60 +replay_buffer/size 999047 +trainer/num train calls 941000 +trainer/QF1 Loss 0.631091 +trainer/QF2 Loss 0.688262 +trainer/Policy Loss 12.1248 +trainer/Q1 Predictions Mean -75.0539 +trainer/Q1 Predictions Std 17.8393 +trainer/Q1 Predictions Max -0.694901 +trainer/Q1 Predictions Min -88.3871 +trainer/Q2 Predictions Mean -75.037 +trainer/Q2 Predictions Std 17.7856 +trainer/Q2 Predictions Max -0.881512 +trainer/Q2 Predictions Min -88.0859 +trainer/Q Targets Mean -74.7256 +trainer/Q Targets Std 17.9348 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2511 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0107111 +trainer/policy/mean Std 0.730249 +trainer/policy/mean Max 0.998782 +trainer/policy/mean Min -0.999511 +trainer/policy/std Mean 0.405767 +trainer/policy/std Std 0.0199761 +trainer/policy/std Max 0.42809 +trainer/policy/std Min 0.375029 +trainer/Advantage Weights Mean 3.25917 +trainer/Advantage Weights Std 15.8697 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.63589e-32 +trainer/Advantage Score Mean -0.48785 +trainer/Advantage Score Std 0.702522 +trainer/Advantage Score Max 1.51191 +trainer/Advantage Score Min -7.21489 +trainer/V1 Predictions Mean -74.4328 +trainer/V1 Predictions Std 17.9423 +trainer/V1 Predictions Max -0.431998 +trainer/V1 Predictions Min -88.0528 +trainer/VF Loss 0.0918968 +expl/num steps total 941000 +expl/num paths total 1296 +expl/path length Mean 500 +expl/path length Std 156 +expl/path length Max 656 +expl/path length Min 344 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0254995 +expl/Actions Std 0.837128 +expl/Actions Max 2.35744 +expl/Actions Min -2.27944 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 846329 +eval/num paths total 946 +eval/path length Mean 920 +eval/path length Std 0 +eval/path length Max 920 +eval/path length Min 920 +eval/Rewards Mean 0.00108696 +eval/Rewards Std 0.0329511 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0622852 +eval/Actions Std 0.699671 +eval/Actions Max 0.999915 +eval/Actions Min -0.999995 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.95881e-06 +time/evaluation sampling (s) 3.84294 +time/exploration sampling (s) 4.657 +time/logging (s) 0.00753055 +time/saving (s) 0.0120354 +time/training (s) 15.5167 +time/epoch (s) 24.0362 +time/total (s) 24257 +Epoch -60 +------------------------------ ---------------- +2022-05-16 00:47:12.562143 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -59 finished +------------------------------ ---------------- +epoch -59 +replay_buffer/size 999047 +trainer/num train calls 942000 +trainer/QF1 Loss 0.788923 +trainer/QF2 Loss 0.866797 +trainer/Policy Loss 18.5351 +trainer/Q1 Predictions Mean -72.8854 +trainer/Q1 Predictions Std 20.1257 +trainer/Q1 Predictions Max -0.752632 +trainer/Q1 Predictions Min -88.6997 +trainer/Q2 Predictions Mean -72.9751 +trainer/Q2 Predictions Std 20.0418 +trainer/Q2 Predictions Max -0.489311 +trainer/Q2 Predictions Min -88.8717 +trainer/Q Targets Mean -72.7943 +trainer/Q Targets Std 20.2689 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9436 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0154909 +trainer/policy/mean Std 0.734138 +trainer/policy/mean Max 0.999107 +trainer/policy/mean Min -0.999241 +trainer/policy/std Mean 0.407379 +trainer/policy/std Std 0.0197556 +trainer/policy/std Max 0.430314 +trainer/policy/std Min 0.377508 +trainer/Advantage Weights Mean 4.93749 +trainer/Advantage Weights Std 19.6476 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.50877e-21 +trainer/Advantage Score Mean -0.471308 +trainer/Advantage Score Std 0.813473 +trainer/Advantage Score Max 1.87061 +trainer/Advantage Score Min -4.68483 +trainer/V1 Predictions Mean -72.5408 +trainer/V1 Predictions Std 20.3252 +trainer/V1 Predictions Max -0.278649 +trainer/V1 Predictions Min -87.7149 +trainer/VF Loss 0.121279 +expl/num steps total 942000 +expl/num paths total 1298 +expl/path length Mean 500 +expl/path length Std 108 +expl/path length Max 608 +expl/path length Min 392 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.012762 +expl/Actions Std 0.845792 +expl/Actions Max 2.51458 +expl/Actions Min -2.31662 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 847329 +eval/num paths total 947 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0356414 +eval/Actions Std 0.76344 +eval/Actions Max 0.999972 +eval/Actions Min -0.999867 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 1.02352e-05 +time/evaluation sampling (s) 3.10181 +time/exploration sampling (s) 5.03098 +time/logging (s) 0.0107254 +time/saving (s) 0.0146345 +time/training (s) 15.5173 +time/epoch (s) 23.6755 +time/total (s) 24280.7 +Epoch -59 +------------------------------ ---------------- +2022-05-16 00:47:36.524422 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -58 finished +------------------------------ ---------------- +epoch -58 +replay_buffer/size 999047 +trainer/num train calls 943000 +trainer/QF1 Loss 0.471966 +trainer/QF2 Loss 0.478529 +trainer/Policy Loss 10.1947 +trainer/Q1 Predictions Mean -74.8993 +trainer/Q1 Predictions Std 18.3692 +trainer/Q1 Predictions Max -2.10513 +trainer/Q1 Predictions Min -88.0308 +trainer/Q2 Predictions Mean -74.9936 +trainer/Q2 Predictions Std 18.3204 +trainer/Q2 Predictions Max -1.12461 +trainer/Q2 Predictions Min -88.3932 +trainer/Q Targets Mean -74.8603 +trainer/Q Targets Std 18.463 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.3467 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00712941 +trainer/policy/mean Std 0.73711 +trainer/policy/mean Max 0.999855 +trainer/policy/mean Min -0.999819 +trainer/policy/std Mean 0.409274 +trainer/policy/std Std 0.0195069 +trainer/policy/std Max 0.433822 +trainer/policy/std Min 0.380257 +trainer/Advantage Weights Mean 2.20886 +trainer/Advantage Weights Std 12.6786 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.31358e-12 +trainer/Advantage Score Mean -0.431738 +trainer/Advantage Score Std 0.477422 +trainer/Advantage Score Max 1.8759 +trainer/Advantage Score Min -2.6433 +trainer/V1 Predictions Mean -74.6875 +trainer/V1 Predictions Std 18.4126 +trainer/V1 Predictions Max -1.24799 +trainer/V1 Predictions Min -88.3231 +trainer/VF Loss 0.0615625 +expl/num steps total 943000 +expl/num paths total 1300 +expl/path length Mean 500 +expl/path length Std 258 +expl/path length Max 758 +expl/path length Min 242 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0480335 +expl/Actions Std 0.83625 +expl/Actions Max 2.24108 +expl/Actions Min -2.30369 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 848329 +eval/num paths total 948 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0685685 +eval/Actions Std 0.755011 +eval/Actions Max 0.999954 +eval/Actions Min -0.999925 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.72598e-06 +time/evaluation sampling (s) 3.35933 +time/exploration sampling (s) 4.22951 +time/logging (s) 0.0089772 +time/saving (s) 0.0138426 +time/training (s) 16.3292 +time/epoch (s) 23.9408 +time/total (s) 24304.6 +Epoch -58 +------------------------------ ---------------- +2022-05-16 00:47:59.962059 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -57 finished +------------------------------ ---------------- +epoch -57 +replay_buffer/size 999047 +trainer/num train calls 944000 +trainer/QF1 Loss 3.43325 +trainer/QF2 Loss 3.57054 +trainer/Policy Loss 11.2918 +trainer/Q1 Predictions Mean -73.9346 +trainer/Q1 Predictions Std 18.7266 +trainer/Q1 Predictions Max -1.48162 +trainer/Q1 Predictions Min -87.6099 +trainer/Q2 Predictions Mean -73.9969 +trainer/Q2 Predictions Std 18.6571 +trainer/Q2 Predictions Max -1.15237 +trainer/Q2 Predictions Min -87.8018 +trainer/Q Targets Mean -73.6613 +trainer/Q Targets Std 19.152 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6959 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0182299 +trainer/policy/mean Std 0.721185 +trainer/policy/mean Max 0.999978 +trainer/policy/mean Min -0.998771 +trainer/policy/std Mean 0.407402 +trainer/policy/std Std 0.0205456 +trainer/policy/std Max 0.430189 +trainer/policy/std Min 0.376387 +trainer/Advantage Weights Mean 3.94225 +trainer/Advantage Weights Std 16.0821 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.84467e-18 +trainer/Advantage Score Mean -0.412738 +trainer/Advantage Score Std 0.734043 +trainer/Advantage Score Max 1.52526 +trainer/Advantage Score Min -4.04011 +trainer/V1 Predictions Mean -73.5463 +trainer/V1 Predictions Std 19.0879 +trainer/V1 Predictions Max 0.0415294 +trainer/V1 Predictions Min -87.5466 +trainer/VF Loss 0.0895684 +expl/num steps total 944000 +expl/num paths total 1302 +expl/path length Mean 500 +expl/path length Std 178 +expl/path length Max 678 +expl/path length Min 322 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0472946 +expl/Actions Std 0.82671 +expl/Actions Max 2.4349 +expl/Actions Min -2.493 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 849024 +eval/num paths total 949 +eval/path length Mean 695 +eval/path length Std 0 +eval/path length Max 695 +eval/path length Min 695 +eval/Rewards Mean 0.00143885 +eval/Rewards Std 0.0379049 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0352857 +eval/Actions Std 0.71457 +eval/Actions Max 0.999793 +eval/Actions Min -0.99976 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.74181e-06 +time/evaluation sampling (s) 3.90281 +time/exploration sampling (s) 4.02999 +time/logging (s) 0.00877554 +time/saving (s) 0.0146481 +time/training (s) 15.4667 +time/epoch (s) 23.423 +time/total (s) 24328.1 +Epoch -57 +------------------------------ ---------------- +2022-05-16 00:48:23.812844 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -56 finished +------------------------------ ---------------- +epoch -56 +replay_buffer/size 999047 +trainer/num train calls 945000 +trainer/QF1 Loss 0.801614 +trainer/QF2 Loss 0.75965 +trainer/Policy Loss 18.0137 +trainer/Q1 Predictions Mean -74.9646 +trainer/Q1 Predictions Std 18.1928 +trainer/Q1 Predictions Max 0.862833 +trainer/Q1 Predictions Min -88.4334 +trainer/Q2 Predictions Mean -74.9186 +trainer/Q2 Predictions Std 18.1064 +trainer/Q2 Predictions Max -0.963069 +trainer/Q2 Predictions Min -88.3315 +trainer/Q Targets Mean -74.6441 +trainer/Q Targets Std 18.0143 +trainer/Q Targets Max -2.67173 +trainer/Q Targets Min -88.2606 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00150279 +trainer/policy/mean Std 0.721908 +trainer/policy/mean Max 0.999254 +trainer/policy/mean Min -0.99928 +trainer/policy/std Mean 0.406629 +trainer/policy/std Std 0.0192306 +trainer/policy/std Max 0.42941 +trainer/policy/std Min 0.377397 +trainer/Advantage Weights Mean 4.22592 +trainer/Advantage Weights Std 17.9007 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.75413e-19 +trainer/Advantage Score Mean -0.446339 +trainer/Advantage Score Std 0.759428 +trainer/Advantage Score Max 1.41246 +trainer/Advantage Score Min -4.31871 +trainer/V1 Predictions Mean -74.3718 +trainer/V1 Predictions Std 18.208 +trainer/V1 Predictions Max -2.12795 +trainer/V1 Predictions Min -88.133 +trainer/VF Loss 0.0969622 +expl/num steps total 945000 +expl/num paths total 1304 +expl/path length Mean 500 +expl/path length Std 299 +expl/path length Max 799 +expl/path length Min 201 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0275139 +expl/Actions Std 0.836127 +expl/Actions Max 2.30136 +expl/Actions Min -2.29707 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 850024 +eval/num paths total 950 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.308218 +eval/Actions Std 0.709105 +eval/Actions Max 0.999896 +eval/Actions Min -0.999149 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.79816e-06 +time/evaluation sampling (s) 3.54067 +time/exploration sampling (s) 4.93047 +time/logging (s) 0.0102147 +time/saving (s) 0.0147427 +time/training (s) 15.3432 +time/epoch (s) 23.8394 +time/total (s) 24351.9 +Epoch -56 +------------------------------ ---------------- +2022-05-16 00:48:47.230765 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -55 finished +------------------------------ ---------------- +epoch -55 +replay_buffer/size 999047 +trainer/num train calls 946000 +trainer/QF1 Loss 0.610131 +trainer/QF2 Loss 0.649902 +trainer/Policy Loss 7.6745 +trainer/Q1 Predictions Mean -75.2872 +trainer/Q1 Predictions Std 17.8997 +trainer/Q1 Predictions Max -1.34675 +trainer/Q1 Predictions Min -88.9118 +trainer/Q2 Predictions Mean -75.2654 +trainer/Q2 Predictions Std 17.9416 +trainer/Q2 Predictions Max -1.61538 +trainer/Q2 Predictions Min -88.7886 +trainer/Q Targets Mean -75.1714 +trainer/Q Targets Std 17.8018 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2905 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.012114 +trainer/policy/mean Std 0.742618 +trainer/policy/mean Max 0.999256 +trainer/policy/mean Min -0.998541 +trainer/policy/std Mean 0.408127 +trainer/policy/std Std 0.0191366 +trainer/policy/std Max 0.430666 +trainer/policy/std Min 0.381201 +trainer/Advantage Weights Mean 1.65929 +trainer/Advantage Weights Std 8.91798 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.46276e-21 +trainer/Advantage Score Mean -0.493526 +trainer/Advantage Score Std 0.630668 +trainer/Advantage Score Max 0.497482 +trainer/Advantage Score Min -4.7453 +trainer/V1 Predictions Mean -74.9192 +trainer/V1 Predictions Std 17.9156 +trainer/V1 Predictions Max -1.04041 +trainer/V1 Predictions Min -88.1636 +trainer/VF Loss 0.0681737 +expl/num steps total 946000 +expl/num paths total 1305 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.044861 +expl/Actions Std 0.841865 +expl/Actions Max 2.13997 +expl/Actions Min -2.54182 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 851024 +eval/num paths total 951 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.306915 +eval/Actions Std 0.549657 +eval/Actions Max 0.99971 +eval/Actions Min -0.999776 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.41585e-06 +time/evaluation sampling (s) 3.29459 +time/exploration sampling (s) 4.61246 +time/logging (s) 0.00692662 +time/saving (s) 0.0111228 +time/training (s) 15.4715 +time/epoch (s) 23.3966 +time/total (s) 24375.3 +Epoch -55 +------------------------------ ---------------- +2022-05-16 00:49:10.596667 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -54 finished +------------------------------ ---------------- +epoch -54 +replay_buffer/size 999047 +trainer/num train calls 947000 +trainer/QF1 Loss 0.658546 +trainer/QF2 Loss 0.554542 +trainer/Policy Loss 31.3672 +trainer/Q1 Predictions Mean -73.8064 +trainer/Q1 Predictions Std 18.3638 +trainer/Q1 Predictions Max -0.72188 +trainer/Q1 Predictions Min -87.5597 +trainer/Q2 Predictions Mean -73.8173 +trainer/Q2 Predictions Std 18.4664 +trainer/Q2 Predictions Max -1.48018 +trainer/Q2 Predictions Min -87.4811 +trainer/Q Targets Mean -74.0604 +trainer/Q Targets Std 18.7251 +trainer/Q Targets Max -1.57829 +trainer/Q Targets Min -87.844 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.022085 +trainer/policy/mean Std 0.732553 +trainer/policy/mean Max 0.999489 +trainer/policy/mean Min -0.999733 +trainer/policy/std Mean 0.407628 +trainer/policy/std Std 0.0188602 +trainer/policy/std Max 0.428072 +trainer/policy/std Min 0.378422 +trainer/Advantage Weights Mean 6.98987 +trainer/Advantage Weights Std 21.2458 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.51744e-15 +trainer/Advantage Score Mean -0.265759 +trainer/Advantage Score Std 0.593417 +trainer/Advantage Score Max 1.04936 +trainer/Advantage Score Min -3.41218 +trainer/V1 Predictions Mean -73.8354 +trainer/V1 Predictions Std 18.7858 +trainer/V1 Predictions Max -1.60991 +trainer/V1 Predictions Min -87.715 +trainer/VF Loss 0.06552 +expl/num steps total 947000 +expl/num paths total 1306 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0903596 +expl/Actions Std 0.905521 +expl/Actions Max 2.54487 +expl/Actions Min -2.59662 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 852024 +eval/num paths total 952 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0382022 +eval/Actions Std 0.729138 +eval/Actions Max 0.999753 +eval/Actions Min -0.999923 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.64523e-06 +time/evaluation sampling (s) 3.93078 +time/exploration sampling (s) 3.95422 +time/logging (s) 0.00728561 +time/saving (s) 0.0104859 +time/training (s) 15.4503 +time/epoch (s) 23.353 +time/total (s) 24398.7 +Epoch -54 +------------------------------ ---------------- +2022-05-16 00:49:34.438556 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -53 finished +------------------------------ ---------------- +epoch -53 +replay_buffer/size 999047 +trainer/num train calls 948000 +trainer/QF1 Loss 1.13144 +trainer/QF2 Loss 1.2361 +trainer/Policy Loss 11.8497 +trainer/Q1 Predictions Mean -74.4556 +trainer/Q1 Predictions Std 16.7538 +trainer/Q1 Predictions Max -4.40763 +trainer/Q1 Predictions Min -88.6563 +trainer/Q2 Predictions Mean -74.4247 +trainer/Q2 Predictions Std 16.6808 +trainer/Q2 Predictions Max -4.57526 +trainer/Q2 Predictions Min -88.4476 +trainer/Q Targets Mean -74.1275 +trainer/Q Targets Std 17.0054 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.3391 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0145611 +trainer/policy/mean Std 0.739661 +trainer/policy/mean Max 0.999285 +trainer/policy/mean Min -0.999242 +trainer/policy/std Mean 0.40628 +trainer/policy/std Std 0.0188142 +trainer/policy/std Max 0.42996 +trainer/policy/std Min 0.379511 +trainer/Advantage Weights Mean 3.16938 +trainer/Advantage Weights Std 14.8979 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.78393e-17 +trainer/Advantage Score Mean -0.517965 +trainer/Advantage Score Std 0.626734 +trainer/Advantage Score Max 1.88251 +trainer/Advantage Score Min -3.70919 +trainer/V1 Predictions Mean -73.8404 +trainer/V1 Predictions Std 17.0903 +trainer/V1 Predictions Max -3.50734 +trainer/V1 Predictions Min -88.279 +trainer/VF Loss 0.0908136 +expl/num steps total 948000 +expl/num paths total 1308 +expl/path length Mean 500 +expl/path length Std 469 +expl/path length Max 969 +expl/path length Min 31 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0429852 +expl/Actions Std 0.820634 +expl/Actions Max 2.14412 +expl/Actions Min -2.12716 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 852972 +eval/num paths total 953 +eval/path length Mean 948 +eval/path length Std 0 +eval/path length Max 948 +eval/path length Min 948 +eval/Rewards Mean 0.00105485 +eval/Rewards Std 0.0324614 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0400683 +eval/Actions Std 0.741569 +eval/Actions Max 0.999884 +eval/Actions Min -0.999625 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.99606e-06 +time/evaluation sampling (s) 3.66721 +time/exploration sampling (s) 4.66201 +time/logging (s) 0.00766975 +time/saving (s) 0.0120224 +time/training (s) 15.4798 +time/epoch (s) 23.8287 +time/total (s) 24422.5 +Epoch -53 +------------------------------ ---------------- +2022-05-16 00:49:57.733740 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -52 finished +------------------------------ ---------------- +epoch -52 +replay_buffer/size 999047 +trainer/num train calls 949000 +trainer/QF1 Loss 0.522614 +trainer/QF2 Loss 0.632139 +trainer/Policy Loss 7.42747 +trainer/Q1 Predictions Mean -76.8144 +trainer/Q1 Predictions Std 15.0119 +trainer/Q1 Predictions Max -1.99376 +trainer/Q1 Predictions Min -88.0371 +trainer/Q2 Predictions Mean -76.7933 +trainer/Q2 Predictions Std 14.9863 +trainer/Q2 Predictions Max -1.69844 +trainer/Q2 Predictions Min -88.074 +trainer/Q Targets Mean -76.8619 +trainer/Q Targets Std 15.0985 +trainer/Q Targets Max -2.9775 +trainer/Q Targets Min -88.3735 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0148309 +trainer/policy/mean Std 0.740986 +trainer/policy/mean Max 0.999988 +trainer/policy/mean Min -0.999397 +trainer/policy/std Mean 0.406453 +trainer/policy/std Std 0.017999 +trainer/policy/std Max 0.429958 +trainer/policy/std Min 0.37886 +trainer/Advantage Weights Mean 2.19836 +trainer/Advantage Weights Std 12.2761 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.9415e-16 +trainer/Advantage Score Mean -0.408371 +trainer/Advantage Score Std 0.581956 +trainer/Advantage Score Max 0.748642 +trainer/Advantage Score Min -3.61779 +trainer/V1 Predictions Mean -76.6165 +trainer/V1 Predictions Std 15.2225 +trainer/V1 Predictions Max -1.65197 +trainer/V1 Predictions Min -88.1313 +trainer/VF Loss 0.057472 +expl/num steps total 949000 +expl/num paths total 1309 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0163005 +expl/Actions Std 0.858923 +expl/Actions Max 2.41553 +expl/Actions Min -2.04138 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 853722 +eval/num paths total 954 +eval/path length Mean 750 +eval/path length Std 0 +eval/path length Max 750 +eval/path length Min 750 +eval/Rewards Mean 0.00133333 +eval/Rewards Std 0.0364905 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0351009 +eval/Actions Std 0.755269 +eval/Actions Max 0.99997 +eval/Actions Min -0.999605 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.06265e-06 +time/evaluation sampling (s) 3.41865 +time/exploration sampling (s) 4.59205 +time/logging (s) 0.00741712 +time/saving (s) 0.0133859 +time/training (s) 15.2486 +time/epoch (s) 23.2801 +time/total (s) 24445.8 +Epoch -52 +------------------------------ ---------------- +2022-05-16 00:50:20.842739 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -51 finished +------------------------------ ---------------- +epoch -51 +replay_buffer/size 999047 +trainer/num train calls 950000 +trainer/QF1 Loss 0.862806 +trainer/QF2 Loss 0.818397 +trainer/Policy Loss 2.61781 +trainer/Q1 Predictions Mean -75.2843 +trainer/Q1 Predictions Std 15.9984 +trainer/Q1 Predictions Max -1.71891 +trainer/Q1 Predictions Min -88.3927 +trainer/Q2 Predictions Mean -75.2916 +trainer/Q2 Predictions Std 16.0708 +trainer/Q2 Predictions Max -1.4022 +trainer/Q2 Predictions Min -88.4663 +trainer/Q Targets Mean -74.9752 +trainer/Q Targets Std 16.0716 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5174 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000273735 +trainer/policy/mean Std 0.736583 +trainer/policy/mean Max 0.999955 +trainer/policy/mean Min -0.999715 +trainer/policy/std Mean 0.407306 +trainer/policy/std Std 0.017808 +trainer/policy/std Max 0.426111 +trainer/policy/std Min 0.381591 +trainer/Advantage Weights Mean 1.1226 +trainer/Advantage Weights Std 8.11436 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.97062e-22 +trainer/Advantage Score Mean -0.721191 +trainer/Advantage Score Std 0.611734 +trainer/Advantage Score Max 0.596039 +trainer/Advantage Score Min -4.99785 +trainer/V1 Predictions Mean -74.715 +trainer/V1 Predictions Std 16.1335 +trainer/V1 Predictions Max -1.72536 +trainer/V1 Predictions Min -87.0693 +trainer/VF Loss 0.0924757 +expl/num steps total 950000 +expl/num paths total 1311 +expl/path length Mean 500 +expl/path length Std 176 +expl/path length Max 676 +expl/path length Min 324 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0310614 +expl/Actions Std 0.822279 +expl/Actions Max 2.33809 +expl/Actions Min -2.2446 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 854722 +eval/num paths total 955 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.109913 +eval/Actions Std 0.684276 +eval/Actions Max 0.999875 +eval/Actions Min -0.999646 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78186e-06 +time/evaluation sampling (s) 3.21444 +time/exploration sampling (s) 4.55033 +time/logging (s) 0.00702919 +time/saving (s) 0.0095943 +time/training (s) 15.3117 +time/epoch (s) 23.0931 +time/total (s) 24468.9 +Epoch -51 +------------------------------ ---------------- +2022-05-16 00:50:44.332689 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -50 finished +------------------------------ ---------------- +epoch -50 +replay_buffer/size 999047 +trainer/num train calls 951000 +trainer/QF1 Loss 1.7631 +trainer/QF2 Loss 1.31252 +trainer/Policy Loss 15.683 +trainer/Q1 Predictions Mean -72.3348 +trainer/Q1 Predictions Std 22.2431 +trainer/Q1 Predictions Max -0.221973 +trainer/Q1 Predictions Min -88.4469 +trainer/Q2 Predictions Mean -72.2416 +trainer/Q2 Predictions Std 22.2922 +trainer/Q2 Predictions Max -0.122528 +trainer/Q2 Predictions Min -88.7862 +trainer/Q Targets Mean -71.6379 +trainer/Q Targets Std 22.045 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8808 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00366963 +trainer/policy/mean Std 0.738222 +trainer/policy/mean Max 0.999741 +trainer/policy/mean Min -0.999816 +trainer/policy/std Mean 0.406485 +trainer/policy/std Std 0.0173397 +trainer/policy/std Max 0.424906 +trainer/policy/std Min 0.380463 +trainer/Advantage Weights Mean 2.61343 +trainer/Advantage Weights Std 15.3051 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.01058e-18 +trainer/Advantage Score Mean -0.730498 +trainer/Advantage Score Std 0.642122 +trainer/Advantage Score Max 1.49296 +trainer/Advantage Score Min -3.93658 +trainer/V1 Predictions Mean -71.4264 +trainer/V1 Predictions Std 22.1509 +trainer/V1 Predictions Max 0.432355 +trainer/V1 Predictions Min -87.801 +trainer/VF Loss 0.11821 +expl/num steps total 951000 +expl/num paths total 1313 +expl/path length Mean 500 +expl/path length Std 394 +expl/path length Max 894 +expl/path length Min 106 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0589675 +expl/Actions Std 0.815948 +expl/Actions Max 2.2363 +expl/Actions Min -2.1844 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 855722 +eval/num paths total 956 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.000671079 +eval/Actions Std 0.695606 +eval/Actions Max 0.999521 +eval/Actions Min -0.99969 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.12785e-06 +time/evaluation sampling (s) 3.85522 +time/exploration sampling (s) 4.2368 +time/logging (s) 0.00746347 +time/saving (s) 0.0193778 +time/training (s) 15.3581 +time/epoch (s) 23.477 +time/total (s) 24492.4 +Epoch -50 +------------------------------ ---------------- +2022-05-16 00:51:07.574086 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -49 finished +------------------------------ ---------------- +epoch -49 +replay_buffer/size 999047 +trainer/num train calls 952000 +trainer/QF1 Loss 0.535158 +trainer/QF2 Loss 0.657216 +trainer/Policy Loss 21.2748 +trainer/Q1 Predictions Mean -73.0478 +trainer/Q1 Predictions Std 19.9955 +trainer/Q1 Predictions Max -1.22939 +trainer/Q1 Predictions Min -87.4026 +trainer/Q2 Predictions Mean -73.1735 +trainer/Q2 Predictions Std 19.8714 +trainer/Q2 Predictions Max -1.77609 +trainer/Q2 Predictions Min -87.4227 +trainer/Q Targets Mean -73.2575 +trainer/Q Targets Std 20.0505 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6819 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0102291 +trainer/policy/mean Std 0.736985 +trainer/policy/mean Max 0.99977 +trainer/policy/mean Min -0.999732 +trainer/policy/std Mean 0.407652 +trainer/policy/std Std 0.0179747 +trainer/policy/std Max 0.42817 +trainer/policy/std Min 0.382498 +trainer/Advantage Weights Mean 4.12393 +trainer/Advantage Weights Std 17.4345 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.81613e-13 +trainer/Advantage Score Mean -0.371831 +trainer/Advantage Score Std 0.577918 +trainer/Advantage Score Max 0.897864 +trainer/Advantage Score Min -2.78774 +trainer/V1 Predictions Mean -72.9871 +trainer/V1 Predictions Std 20.1323 +trainer/V1 Predictions Max -0.741911 +trainer/V1 Predictions Min -87.6434 +trainer/VF Loss 0.0624479 +expl/num steps total 952000 +expl/num paths total 1314 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.153707 +expl/Actions Std 0.869383 +expl/Actions Max 2.28884 +expl/Actions Min -2.44006 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 856722 +eval/num paths total 957 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.268873 +eval/Actions Std 0.791839 +eval/Actions Max 0.999874 +eval/Actions Min -0.999653 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.96207e-06 +time/evaluation sampling (s) 3.47832 +time/exploration sampling (s) 4.37759 +time/logging (s) 0.00691364 +time/saving (s) 0.00978259 +time/training (s) 15.3543 +time/epoch (s) 23.2269 +time/total (s) 24515.6 +Epoch -49 +------------------------------ ---------------- +2022-05-16 00:51:30.754819 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -48 finished +------------------------------ ---------------- +epoch -48 +replay_buffer/size 999047 +trainer/num train calls 953000 +trainer/QF1 Loss 0.63607 +trainer/QF2 Loss 0.492677 +trainer/Policy Loss 14.4724 +trainer/Q1 Predictions Mean -73.3693 +trainer/Q1 Predictions Std 18.9802 +trainer/Q1 Predictions Max -0.981796 +trainer/Q1 Predictions Min -86.8152 +trainer/Q2 Predictions Mean -73.4 +trainer/Q2 Predictions Std 18.9939 +trainer/Q2 Predictions Max -0.477149 +trainer/Q2 Predictions Min -87.0662 +trainer/Q Targets Mean -73.4802 +trainer/Q Targets Std 19.0944 +trainer/Q Targets Max -2.07861 +trainer/Q Targets Min -86.9014 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000905905 +trainer/policy/mean Std 0.728696 +trainer/policy/mean Max 0.999199 +trainer/policy/mean Min -0.997767 +trainer/policy/std Mean 0.406093 +trainer/policy/std Std 0.0189846 +trainer/policy/std Max 0.4248 +trainer/policy/std Min 0.378759 +trainer/Advantage Weights Mean 4.53178 +trainer/Advantage Weights Std 18.9083 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.6919e-22 +trainer/Advantage Score Mean -0.433045 +trainer/Advantage Score Std 0.641914 +trainer/Advantage Score Max 1.26139 +trainer/Advantage Score Min -4.83856 +trainer/V1 Predictions Mean -73.1975 +trainer/V1 Predictions Std 19.2301 +trainer/V1 Predictions Max -0.439947 +trainer/V1 Predictions Min -86.7359 +trainer/VF Loss 0.0831646 +expl/num steps total 953000 +expl/num paths total 1316 +expl/path length Mean 500 +expl/path length Std 308 +expl/path length Max 808 +expl/path length Min 192 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0202 +expl/Actions Std 0.821975 +expl/Actions Max 2.13056 +expl/Actions Min -2.36129 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 857410 +eval/num paths total 958 +eval/path length Mean 688 +eval/path length Std 0 +eval/path length Max 688 +eval/path length Min 688 +eval/Rewards Mean 0.00145349 +eval/Rewards Std 0.0380969 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0364947 +eval/Actions Std 0.753408 +eval/Actions Max 0.999713 +eval/Actions Min -0.999563 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.06871e-06 +time/evaluation sampling (s) 4.01328 +time/exploration sampling (s) 4.42582 +time/logging (s) 0.00579656 +time/saving (s) 0.009212 +time/training (s) 14.712 +time/epoch (s) 23.1661 +time/total (s) 24538.8 +Epoch -48 +------------------------------ ---------------- +2022-05-16 00:51:54.782329 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -47 finished +------------------------------ ---------------- +epoch -47 +replay_buffer/size 999047 +trainer/num train calls 954000 +trainer/QF1 Loss 1.46206 +trainer/QF2 Loss 1.18082 +trainer/Policy Loss 15.2434 +trainer/Q1 Predictions Mean -74.1126 +trainer/Q1 Predictions Std 18.82 +trainer/Q1 Predictions Max -0.181445 +trainer/Q1 Predictions Min -88.0794 +trainer/Q2 Predictions Mean -74.049 +trainer/Q2 Predictions Std 18.78 +trainer/Q2 Predictions Max 0.261002 +trainer/Q2 Predictions Min -87.7512 +trainer/Q Targets Mean -74.3701 +trainer/Q Targets Std 18.5464 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9261 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0208926 +trainer/policy/mean Std 0.742476 +trainer/policy/mean Max 0.999863 +trainer/policy/mean Min -0.997857 +trainer/policy/std Mean 0.407121 +trainer/policy/std Std 0.018811 +trainer/policy/std Max 0.428384 +trainer/policy/std Min 0.375157 +trainer/Advantage Weights Mean 3.59901 +trainer/Advantage Weights Std 15.5433 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.40677e-20 +trainer/Advantage Score Mean -0.384188 +trainer/Advantage Score Std 0.617195 +trainer/Advantage Score Max 1.0717 +trainer/Advantage Score Min -4.51734 +trainer/V1 Predictions Mean -74.0338 +trainer/V1 Predictions Std 18.7968 +trainer/V1 Predictions Max 0.962439 +trainer/V1 Predictions Min -87.7335 +trainer/VF Loss 0.0672994 +expl/num steps total 954000 +expl/num paths total 1318 +expl/path length Mean 500 +expl/path length Std 283 +expl/path length Max 783 +expl/path length Min 217 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0285866 +expl/Actions Std 0.837088 +expl/Actions Max 2.21449 +expl/Actions Min -2.21804 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 857870 +eval/num paths total 959 +eval/path length Mean 460 +eval/path length Std 0 +eval/path length Max 460 +eval/path length Min 460 +eval/Rewards Mean 0.00217391 +eval/Rewards Std 0.0465745 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0109104 +eval/Actions Std 0.752777 +eval/Actions Max 0.999926 +eval/Actions Min -0.999986 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 5.06407e-06 +time/evaluation sampling (s) 3.99802 +time/exploration sampling (s) 5.05576 +time/logging (s) 0.00617699 +time/saving (s) 0.0126036 +time/training (s) 14.9443 +time/epoch (s) 24.0168 +time/total (s) 24562.8 +Epoch -47 +------------------------------ ---------------- +2022-05-16 00:52:17.789792 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -46 finished +------------------------------ ---------------- +epoch -46 +replay_buffer/size 999047 +trainer/num train calls 955000 +trainer/QF1 Loss 0.552103 +trainer/QF2 Loss 0.604016 +trainer/Policy Loss 11.5628 +trainer/Q1 Predictions Mean -74.7071 +trainer/Q1 Predictions Std 17.6725 +trainer/Q1 Predictions Max -0.627839 +trainer/Q1 Predictions Min -87.6 +trainer/Q2 Predictions Mean -74.6557 +trainer/Q2 Predictions Std 17.6447 +trainer/Q2 Predictions Max -1.23498 +trainer/Q2 Predictions Min -87.5741 +trainer/Q Targets Mean -74.7782 +trainer/Q Targets Std 17.8669 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5323 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0165268 +trainer/policy/mean Std 0.741901 +trainer/policy/mean Max 0.999877 +trainer/policy/mean Min -0.999691 +trainer/policy/std Mean 0.40846 +trainer/policy/std Std 0.0187745 +trainer/policy/std Max 0.428915 +trainer/policy/std Min 0.375927 +trainer/Advantage Weights Mean 2.68537 +trainer/Advantage Weights Std 14.0253 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.76092e-17 +trainer/Advantage Score Mean -0.406712 +trainer/Advantage Score Std 0.620551 +trainer/Advantage Score Max 1.90556 +trainer/Advantage Score Min -3.85781 +trainer/V1 Predictions Mean -74.5296 +trainer/V1 Predictions Std 17.8688 +trainer/V1 Predictions Max -0.684949 +trainer/V1 Predictions Min -87.3563 +trainer/VF Loss 0.0837483 +expl/num steps total 955000 +expl/num paths total 1320 +expl/path length Mean 500 +expl/path length Std 422 +expl/path length Max 922 +expl/path length Min 78 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00848664 +expl/Actions Std 0.819805 +expl/Actions Max 2.77344 +expl/Actions Min -2.1356 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 858870 +eval/num paths total 960 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.195856 +eval/Actions Std 0.707152 +eval/Actions Max 0.999575 +eval/Actions Min -0.999702 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.31691e-06 +time/evaluation sampling (s) 2.94468 +time/exploration sampling (s) 5.08716 +time/logging (s) 0.00742008 +time/saving (s) 0.00905141 +time/training (s) 14.9459 +time/epoch (s) 22.9942 +time/total (s) 24585.8 +Epoch -46 +------------------------------ ---------------- +2022-05-16 00:52:41.351653 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -45 finished +------------------------------ ---------------- +epoch -45 +replay_buffer/size 999047 +trainer/num train calls 956000 +trainer/QF1 Loss 0.462422 +trainer/QF2 Loss 0.473177 +trainer/Policy Loss 9.69602 +trainer/Q1 Predictions Mean -75.3869 +trainer/Q1 Predictions Std 16.0918 +trainer/Q1 Predictions Max -0.230908 +trainer/Q1 Predictions Min -88.1041 +trainer/Q2 Predictions Mean -75.4039 +trainer/Q2 Predictions Std 16.1271 +trainer/Q2 Predictions Max -1.01509 +trainer/Q2 Predictions Min -88.001 +trainer/Q Targets Mean -75.3023 +trainer/Q Targets Std 16.3494 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9491 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0088218 +trainer/policy/mean Std 0.745079 +trainer/policy/mean Max 0.999551 +trainer/policy/mean Min -0.999886 +trainer/policy/std Mean 0.409199 +trainer/policy/std Std 0.0189632 +trainer/policy/std Max 0.431907 +trainer/policy/std Min 0.376905 +trainer/Advantage Weights Mean 3.3238 +trainer/Advantage Weights Std 15.4226 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.38637e-20 +trainer/Advantage Score Mean -0.381685 +trainer/Advantage Score Std 0.63706 +trainer/Advantage Score Max 1.05055 +trainer/Advantage Score Min -4.39251 +trainer/V1 Predictions Mean -75.121 +trainer/V1 Predictions Std 16.2846 +trainer/V1 Predictions Max -0.0654895 +trainer/V1 Predictions Min -87.8248 +trainer/VF Loss 0.0683829 +expl/num steps total 956000 +expl/num paths total 1321 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.365509 +expl/Actions Std 0.833167 +expl/Actions Max 2.13172 +expl/Actions Min -2.30398 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 859419 +eval/num paths total 961 +eval/path length Mean 549 +eval/path length Std 0 +eval/path length Max 549 +eval/path length Min 549 +eval/Rewards Mean 0.00182149 +eval/Rewards Std 0.0426401 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00752827 +eval/Actions Std 0.744327 +eval/Actions Max 0.999816 +eval/Actions Min -0.999875 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.67988e-06 +time/evaluation sampling (s) 3.55815 +time/exploration sampling (s) 4.20168 +time/logging (s) 0.0109145 +time/saving (s) 0.0163004 +time/training (s) 15.7655 +time/epoch (s) 23.5526 +time/total (s) 24609.3 +Epoch -45 +------------------------------ ---------------- +2022-05-16 00:53:04.805525 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -44 finished +------------------------------ ---------------- +epoch -44 +replay_buffer/size 999047 +trainer/num train calls 957000 +trainer/QF1 Loss 0.436409 +trainer/QF2 Loss 0.414996 +trainer/Policy Loss 9.12046 +trainer/Q1 Predictions Mean -75.113 +trainer/Q1 Predictions Std 16.9294 +trainer/Q1 Predictions Max -1.52172 +trainer/Q1 Predictions Min -88.6549 +trainer/Q2 Predictions Mean -75.0367 +trainer/Q2 Predictions Std 16.904 +trainer/Q2 Predictions Max -2.03647 +trainer/Q2 Predictions Min -88.7322 +trainer/Q Targets Mean -75.0966 +trainer/Q Targets Std 16.9574 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.077 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00201216 +trainer/policy/mean Std 0.73497 +trainer/policy/mean Max 0.999331 +trainer/policy/mean Min -0.999875 +trainer/policy/std Mean 0.40678 +trainer/policy/std Std 0.0196482 +trainer/policy/std Max 0.431723 +trainer/policy/std Min 0.375473 +trainer/Advantage Weights Mean 2.23394 +trainer/Advantage Weights Std 12.7102 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.03708e-18 +trainer/Advantage Score Mean -0.466447 +trainer/Advantage Score Std 0.579286 +trainer/Advantage Score Max 0.946228 +trainer/Advantage Score Min -4.03356 +trainer/V1 Predictions Mean -74.846 +trainer/V1 Predictions Std 17.1044 +trainer/V1 Predictions Max -0.704882 +trainer/V1 Predictions Min -88.0532 +trainer/VF Loss 0.0645956 +expl/num steps total 957000 +expl/num paths total 1323 +expl/path length Mean 500 +expl/path length Std 196 +expl/path length Max 696 +expl/path length Min 304 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0407063 +expl/Actions Std 0.832755 +expl/Actions Max 2.43675 +expl/Actions Min -2.28215 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 860419 +eval/num paths total 962 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.270927 +eval/Actions Std 0.637231 +eval/Actions Max 0.999902 +eval/Actions Min -0.999399 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.69968e-06 +time/evaluation sampling (s) 3.70827 +time/exploration sampling (s) 4.74922 +time/logging (s) 0.00714778 +time/saving (s) 0.00967104 +time/training (s) 14.9584 +time/epoch (s) 23.4327 +time/total (s) 24632.8 +Epoch -44 +------------------------------ ---------------- +2022-05-16 00:53:27.850323 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -43 finished +------------------------------ ---------------- +epoch -43 +replay_buffer/size 999047 +trainer/num train calls 958000 +trainer/QF1 Loss 0.832426 +trainer/QF2 Loss 0.860971 +trainer/Policy Loss 24.812 +trainer/Q1 Predictions Mean -74.7552 +trainer/Q1 Predictions Std 18.6383 +trainer/Q1 Predictions Max -0.193135 +trainer/Q1 Predictions Min -88.5685 +trainer/Q2 Predictions Mean -74.7816 +trainer/Q2 Predictions Std 18.6169 +trainer/Q2 Predictions Max 0.511463 +trainer/Q2 Predictions Min -88.6746 +trainer/Q Targets Mean -74.3477 +trainer/Q Targets Std 18.6584 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5654 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0155105 +trainer/policy/mean Std 0.732061 +trainer/policy/mean Max 0.999288 +trainer/policy/mean Min -0.999457 +trainer/policy/std Mean 0.406136 +trainer/policy/std Std 0.0192517 +trainer/policy/std Max 0.427981 +trainer/policy/std Min 0.377373 +trainer/Advantage Weights Mean 3.06915 +trainer/Advantage Weights Std 16.3115 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.98647e-14 +trainer/Advantage Score Mean -0.510191 +trainer/Advantage Score Std 0.569295 +trainer/Advantage Score Max 1.15088 +trainer/Advantage Score Min -3.06295 +trainer/V1 Predictions Mean -74.1038 +trainer/V1 Predictions Std 18.8071 +trainer/V1 Predictions Max 0.778273 +trainer/V1 Predictions Min -87.4499 +trainer/VF Loss 0.0721229 +expl/num steps total 958000 +expl/num paths total 1325 +expl/path length Mean 500 +expl/path length Std 396 +expl/path length Max 896 +expl/path length Min 104 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0311067 +expl/Actions Std 0.824004 +expl/Actions Max 2.25612 +expl/Actions Min -2.42221 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 861419 +eval/num paths total 963 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.160869 +eval/Actions Std 0.746241 +eval/Actions Max 0.999955 +eval/Actions Min -0.999594 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.22377e-06 +time/evaluation sampling (s) 3.66391 +time/exploration sampling (s) 4.12506 +time/logging (s) 0.0071178 +time/saving (s) 0.0122815 +time/training (s) 15.2236 +time/epoch (s) 23.032 +time/total (s) 24655.8 +Epoch -43 +------------------------------ ---------------- +2022-05-16 00:53:50.400475 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -42 finished +------------------------------ ---------------- +epoch -42 +replay_buffer/size 999047 +trainer/num train calls 959000 +trainer/QF1 Loss 0.739282 +trainer/QF2 Loss 0.780203 +trainer/Policy Loss 4.5512 +trainer/Q1 Predictions Mean -77.2043 +trainer/Q1 Predictions Std 15.0783 +trainer/Q1 Predictions Max -0.207107 +trainer/Q1 Predictions Min -88.9263 +trainer/Q2 Predictions Mean -77.1729 +trainer/Q2 Predictions Std 15.1761 +trainer/Q2 Predictions Max 0.332019 +trainer/Q2 Predictions Min -88.7607 +trainer/Q Targets Mean -76.888 +trainer/Q Targets Std 14.6971 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1677 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0113496 +trainer/policy/mean Std 0.739065 +trainer/policy/mean Max 0.999907 +trainer/policy/mean Min -0.998649 +trainer/policy/std Mean 0.405808 +trainer/policy/std Std 0.0180201 +trainer/policy/std Max 0.423842 +trainer/policy/std Min 0.378179 +trainer/Advantage Weights Mean 1.42046 +trainer/Advantage Weights Std 9.881 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.072e-24 +trainer/Advantage Score Mean -0.524446 +trainer/Advantage Score Std 0.620743 +trainer/Advantage Score Max 1.12292 +trainer/Advantage Score Min -5.51925 +trainer/V1 Predictions Mean -76.529 +trainer/V1 Predictions Std 15.0499 +trainer/V1 Predictions Max -0.313897 +trainer/V1 Predictions Min -88.016 +trainer/VF Loss 0.0724835 +expl/num steps total 959000 +expl/num paths total 1327 +expl/path length Mean 500 +expl/path length Std 238 +expl/path length Max 738 +expl/path length Min 262 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.024582 +expl/Actions Std 0.843893 +expl/Actions Max 2.12344 +expl/Actions Min -2.27026 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 862419 +eval/num paths total 964 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.167021 +eval/Actions Std 0.669839 +eval/Actions Max 0.999877 +eval/Actions Min -0.999908 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.05008e-06 +time/evaluation sampling (s) 3.9981 +time/exploration sampling (s) 3.85643 +time/logging (s) 0.00694288 +time/saving (s) 0.00950369 +time/training (s) 14.6661 +time/epoch (s) 22.5371 +time/total (s) 24678.4 +Epoch -42 +------------------------------ ---------------- +2022-05-16 00:54:14.114127 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -41 finished +------------------------------ ---------------- +epoch -41 +replay_buffer/size 999047 +trainer/num train calls 960000 +trainer/QF1 Loss 1.40301 +trainer/QF2 Loss 1.19743 +trainer/Policy Loss 13.9562 +trainer/Q1 Predictions Mean -76.228 +trainer/Q1 Predictions Std 15.8045 +trainer/Q1 Predictions Max -0.362362 +trainer/Q1 Predictions Min -88.3967 +trainer/Q2 Predictions Mean -76.1757 +trainer/Q2 Predictions Std 15.7974 +trainer/Q2 Predictions Max 0.139023 +trainer/Q2 Predictions Min -88.4794 +trainer/Q Targets Mean -75.9964 +trainer/Q Targets Std 15.4933 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8103 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -3.28748e-05 +trainer/policy/mean Std 0.722746 +trainer/policy/mean Max 0.999341 +trainer/policy/mean Min -0.997459 +trainer/policy/std Mean 0.407159 +trainer/policy/std Std 0.0179527 +trainer/policy/std Max 0.42697 +trainer/policy/std Min 0.379133 +trainer/Advantage Weights Mean 3.90628 +trainer/Advantage Weights Std 17.3414 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.23345e-15 +trainer/Advantage Score Mean -0.431881 +trainer/Advantage Score Std 0.515243 +trainer/Advantage Score Max 1.74781 +trainer/Advantage Score Min -3.30958 +trainer/V1 Predictions Mean -75.7946 +trainer/V1 Predictions Std 15.5942 +trainer/V1 Predictions Max 0.101734 +trainer/V1 Predictions Min -87.697 +trainer/VF Loss 0.0692954 +expl/num steps total 960000 +expl/num paths total 1328 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0262441 +expl/Actions Std 0.815367 +expl/Actions Max 2.27513 +expl/Actions Min -2.19086 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 863419 +eval/num paths total 965 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0629438 +eval/Actions Std 0.733256 +eval/Actions Max 0.999742 +eval/Actions Min -0.999792 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.82796e-06 +time/evaluation sampling (s) 3.74987 +time/exploration sampling (s) 5.00292 +time/logging (s) 0.00704316 +time/saving (s) 0.0090956 +time/training (s) 14.932 +time/epoch (s) 23.7009 +time/total (s) 24702.1 +Epoch -41 +------------------------------ ---------------- +2022-05-16 00:54:37.092817 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -40 finished +------------------------------ ---------------- +epoch -40 +replay_buffer/size 999047 +trainer/num train calls 961000 +trainer/QF1 Loss 0.766379 +trainer/QF2 Loss 0.785636 +trainer/Policy Loss 8.18531 +trainer/Q1 Predictions Mean -76.0182 +trainer/Q1 Predictions Std 15.9191 +trainer/Q1 Predictions Max -3.04427 +trainer/Q1 Predictions Min -87.5034 +trainer/Q2 Predictions Mean -76.0994 +trainer/Q2 Predictions Std 15.8594 +trainer/Q2 Predictions Max -3.65403 +trainer/Q2 Predictions Min -87.6822 +trainer/Q Targets Mean -75.6491 +trainer/Q Targets Std 15.9949 +trainer/Q Targets Max -1.79027 +trainer/Q Targets Min -87.3883 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000851852 +trainer/policy/mean Std 0.733347 +trainer/policy/mean Max 0.999655 +trainer/policy/mean Min -0.999056 +trainer/policy/std Mean 0.406322 +trainer/policy/std Std 0.0179909 +trainer/policy/std Max 0.426878 +trainer/policy/std Min 0.378645 +trainer/Advantage Weights Mean 2.96505 +trainer/Advantage Weights Std 15.3406 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.97145e-17 +trainer/Advantage Score Mean -0.476083 +trainer/Advantage Score Std 0.627786 +trainer/Advantage Score Max 1.63679 +trainer/Advantage Score Min -3.77648 +trainer/V1 Predictions Mean -75.4231 +trainer/V1 Predictions Std 16.1683 +trainer/V1 Predictions Max -0.961577 +trainer/V1 Predictions Min -87.0959 +trainer/VF Loss 0.0785223 +expl/num steps total 961000 +expl/num paths total 1329 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.103914 +expl/Actions Std 0.844359 +expl/Actions Max 2.15593 +expl/Actions Min -2.14795 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 864419 +eval/num paths total 966 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.158837 +eval/Actions Std 0.706641 +eval/Actions Max 0.999873 +eval/Actions Min -0.999699 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.12276e-06 +time/evaluation sampling (s) 2.94741 +time/exploration sampling (s) 5.1534 +time/logging (s) 0.00734279 +time/saving (s) 0.0104982 +time/training (s) 14.8475 +time/epoch (s) 22.9662 +time/total (s) 24725 +Epoch -40 +------------------------------ ---------------- +2022-05-16 00:55:00.040286 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -39 finished +------------------------------ ---------------- +epoch -39 +replay_buffer/size 999047 +trainer/num train calls 962000 +trainer/QF1 Loss 5.1767 +trainer/QF2 Loss 5.18146 +trainer/Policy Loss 33.2189 +trainer/Q1 Predictions Mean -72.9325 +trainer/Q1 Predictions Std 18.9052 +trainer/Q1 Predictions Max -0.167482 +trainer/Q1 Predictions Min -87.6749 +trainer/Q2 Predictions Mean -72.888 +trainer/Q2 Predictions Std 18.9486 +trainer/Q2 Predictions Max 0.470095 +trainer/Q2 Predictions Min -88.3038 +trainer/Q Targets Mean -73.1349 +trainer/Q Targets Std 19.0075 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5211 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -7.82637e-05 +trainer/policy/mean Std 0.724988 +trainer/policy/mean Max 0.998865 +trainer/policy/mean Min -0.999449 +trainer/policy/std Mean 0.406457 +trainer/policy/std Std 0.0186665 +trainer/policy/std Max 0.425946 +trainer/policy/std Min 0.377945 +trainer/Advantage Weights Mean 8.00967 +trainer/Advantage Weights Std 22.0089 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.01217e-16 +trainer/Advantage Score Mean -0.209492 +trainer/Advantage Score Std 0.56923 +trainer/Advantage Score Max 1.60445 +trainer/Advantage Score Min -3.52295 +trainer/V1 Predictions Mean -72.9779 +trainer/V1 Predictions Std 18.9777 +trainer/V1 Predictions Max 0.241906 +trainer/V1 Predictions Min -87.4767 +trainer/VF Loss 0.0734203 +expl/num steps total 962000 +expl/num paths total 1331 +expl/path length Mean 500 +expl/path length Std 367 +expl/path length Max 867 +expl/path length Min 133 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0372959 +expl/Actions Std 0.828545 +expl/Actions Max 2.31795 +expl/Actions Min -2.32053 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 865419 +eval/num paths total 967 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0978841 +eval/Actions Std 0.659358 +eval/Actions Max 0.999838 +eval/Actions Min -0.999683 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.70922e-06 +time/evaluation sampling (s) 3.43612 +time/exploration sampling (s) 4.28441 +time/logging (s) 0.00675824 +time/saving (s) 0.00990541 +time/training (s) 15.1968 +time/epoch (s) 22.934 +time/total (s) 24748 +Epoch -39 +------------------------------ ---------------- +2022-05-16 00:55:23.235882 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -38 finished +------------------------------ ---------------- +epoch -38 +replay_buffer/size 999047 +trainer/num train calls 963000 +trainer/QF1 Loss 1.23632 +trainer/QF2 Loss 1.09981 +trainer/Policy Loss 29.0116 +trainer/Q1 Predictions Mean -72.9056 +trainer/Q1 Predictions Std 20.5075 +trainer/Q1 Predictions Max -1.56142 +trainer/Q1 Predictions Min -87.9188 +trainer/Q2 Predictions Mean -72.9029 +trainer/Q2 Predictions Std 20.4724 +trainer/Q2 Predictions Max -1.66372 +trainer/Q2 Predictions Min -87.9399 +trainer/Q Targets Mean -73.0221 +trainer/Q Targets Std 20.4965 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2107 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0094276 +trainer/policy/mean Std 0.732739 +trainer/policy/mean Max 0.999884 +trainer/policy/mean Min -0.998734 +trainer/policy/std Mean 0.405396 +trainer/policy/std Std 0.0185684 +trainer/policy/std Max 0.424304 +trainer/policy/std Min 0.37641 +trainer/Advantage Weights Mean 5.21027 +trainer/Advantage Weights Std 19.3442 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 8.75616e-17 +trainer/Advantage Score Mean -0.363368 +trainer/Advantage Score Std 0.69244 +trainer/Advantage Score Max 1.47596 +trainer/Advantage Score Min -3.69742 +trainer/V1 Predictions Mean -72.834 +trainer/V1 Predictions Std 20.5137 +trainer/V1 Predictions Max -2.69124 +trainer/V1 Predictions Min -88.0917 +trainer/VF Loss 0.0906229 +expl/num steps total 963000 +expl/num paths total 1332 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.319784 +expl/Actions Std 0.856578 +expl/Actions Max 2.13975 +expl/Actions Min -2.35657 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 866419 +eval/num paths total 968 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.318981 +eval/Actions Std 0.719377 +eval/Actions Max 0.999714 +eval/Actions Min -0.999869 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.77581e-06 +time/evaluation sampling (s) 3.97259 +time/exploration sampling (s) 4.44306 +time/logging (s) 0.00655652 +time/saving (s) 0.00909333 +time/training (s) 14.7517 +time/epoch (s) 23.183 +time/total (s) 24771.2 +Epoch -38 +------------------------------ ---------------- +2022-05-16 00:55:46.161659 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -37 finished +------------------------------ ---------------- +epoch -37 +replay_buffer/size 999047 +trainer/num train calls 964000 +trainer/QF1 Loss 6.61235 +trainer/QF2 Loss 6.54272 +trainer/Policy Loss 15.9974 +trainer/Q1 Predictions Mean -74.1661 +trainer/Q1 Predictions Std 18.9867 +trainer/Q1 Predictions Max -1.71572 +trainer/Q1 Predictions Min -87.6718 +trainer/Q2 Predictions Mean -74.1122 +trainer/Q2 Predictions Std 19.0102 +trainer/Q2 Predictions Max -1.57376 +trainer/Q2 Predictions Min -87.5296 +trainer/Q Targets Mean -73.8169 +trainer/Q Targets Std 19.3802 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9424 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.022246 +trainer/policy/mean Std 0.728086 +trainer/policy/mean Max 0.998634 +trainer/policy/mean Min -0.999648 +trainer/policy/std Mean 0.405523 +trainer/policy/std Std 0.0192175 +trainer/policy/std Max 0.42747 +trainer/policy/std Min 0.377691 +trainer/Advantage Weights Mean 4.28369 +trainer/Advantage Weights Std 18.2783 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.49944e-14 +trainer/Advantage Score Mean -0.427892 +trainer/Advantage Score Std 0.60008 +trainer/Advantage Score Max 0.881409 +trainer/Advantage Score Min -3.13201 +trainer/V1 Predictions Mean -73.7045 +trainer/V1 Predictions Std 19.3418 +trainer/V1 Predictions Max -0.264517 +trainer/V1 Predictions Min -87.4372 +trainer/VF Loss 0.0669422 +expl/num steps total 964000 +expl/num paths total 1333 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0376611 +expl/Actions Std 0.921067 +expl/Actions Max 2.36391 +expl/Actions Min -2.49563 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 866858 +eval/num paths total 969 +eval/path length Mean 439 +eval/path length Std 0 +eval/path length Max 439 +eval/path length Min 439 +eval/Rewards Mean 0.0022779 +eval/Rewards Std 0.047673 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00921744 +eval/Actions Std 0.746856 +eval/Actions Max 0.999896 +eval/Actions Min -0.999988 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 7.289e-06 +time/evaluation sampling (s) 3.19665 +time/exploration sampling (s) 5.16304 +time/logging (s) 0.00636848 +time/saving (s) 0.0125607 +time/training (s) 14.5348 +time/epoch (s) 22.9135 +time/total (s) 24794.1 +Epoch -37 +------------------------------ ---------------- +2022-05-16 00:56:08.918019 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -36 finished +------------------------------ ---------------- +epoch -36 +replay_buffer/size 999047 +trainer/num train calls 965000 +trainer/QF1 Loss 0.633085 +trainer/QF2 Loss 0.676144 +trainer/Policy Loss 7.08237 +trainer/Q1 Predictions Mean -74.0786 +trainer/Q1 Predictions Std 18.9808 +trainer/Q1 Predictions Max -1.07269 +trainer/Q1 Predictions Min -88.2438 +trainer/Q2 Predictions Mean -74.09 +trainer/Q2 Predictions Std 18.8397 +trainer/Q2 Predictions Max -0.980369 +trainer/Q2 Predictions Min -87.9327 +trainer/Q Targets Mean -73.8301 +trainer/Q Targets Std 19.2339 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1585 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00173945 +trainer/policy/mean Std 0.727494 +trainer/policy/mean Max 0.998505 +trainer/policy/mean Min -0.999495 +trainer/policy/std Mean 0.406537 +trainer/policy/std Std 0.0206576 +trainer/policy/std Max 0.430226 +trainer/policy/std Min 0.378466 +trainer/Advantage Weights Mean 1.96346 +trainer/Advantage Weights Std 11.0274 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.67734e-23 +trainer/Advantage Score Mean -0.426449 +trainer/Advantage Score Std 0.688663 +trainer/Advantage Score Max 0.962853 +trainer/Advantage Score Min -5.24422 +trainer/V1 Predictions Mean -73.6491 +trainer/V1 Predictions Std 19.2462 +trainer/V1 Predictions Max -0.639178 +trainer/V1 Predictions Min -88.0252 +trainer/VF Loss 0.0734474 +expl/num steps total 965000 +expl/num paths total 1335 +expl/path length Mean 500 +expl/path length Std 457 +expl/path length Max 957 +expl/path length Min 43 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.034088 +expl/Actions Std 0.829492 +expl/Actions Max 2.43525 +expl/Actions Min -2.25558 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 867858 +eval/num paths total 970 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.124138 +eval/Actions Std 0.730018 +eval/Actions Max 0.999844 +eval/Actions Min -0.99944 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.57604e-06 +time/evaluation sampling (s) 3.53806 +time/exploration sampling (s) 4.71094 +time/logging (s) 0.00825763 +time/saving (s) 0.0129692 +time/training (s) 14.4731 +time/epoch (s) 22.7433 +time/total (s) 24816.8 +Epoch -36 +------------------------------ ---------------- +2022-05-16 00:56:31.911223 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -35 finished +------------------------------ ---------------- +epoch -35 +replay_buffer/size 999047 +trainer/num train calls 966000 +trainer/QF1 Loss 0.504769 +trainer/QF2 Loss 0.609385 +trainer/Policy Loss 13.7918 +trainer/Q1 Predictions Mean -75.2224 +trainer/Q1 Predictions Std 17.7434 +trainer/Q1 Predictions Max -0.787835 +trainer/Q1 Predictions Min -88.6646 +trainer/Q2 Predictions Mean -75.1413 +trainer/Q2 Predictions Std 17.8078 +trainer/Q2 Predictions Max -1.79029 +trainer/Q2 Predictions Min -88.5471 +trainer/Q Targets Mean -75.2385 +trainer/Q Targets Std 17.625 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.4257 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.031456 +trainer/policy/mean Std 0.735442 +trainer/policy/mean Max 0.999387 +trainer/policy/mean Min -0.999578 +trainer/policy/std Mean 0.406967 +trainer/policy/std Std 0.019727 +trainer/policy/std Max 0.429 +trainer/policy/std Min 0.380727 +trainer/Advantage Weights Mean 4.38233 +trainer/Advantage Weights Std 16.786 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.69098e-15 +trainer/Advantage Score Mean -0.290177 +trainer/Advantage Score Std 0.549593 +trainer/Advantage Score Max 0.694057 +trainer/Advantage Score Min -3.2638 +trainer/V1 Predictions Mean -74.941 +trainer/V1 Predictions Std 17.8974 +trainer/V1 Predictions Max -0.869233 +trainer/V1 Predictions Min -88.1665 +trainer/VF Loss 0.0515247 +expl/num steps total 966000 +expl/num paths total 1337 +expl/path length Mean 500 +expl/path length Std 423 +expl/path length Max 923 +expl/path length Min 77 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0413046 +expl/Actions Std 0.836672 +expl/Actions Max 2.28464 +expl/Actions Min -2.40674 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 868858 +eval/num paths total 971 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.00518055 +eval/Actions Std 0.740741 +eval/Actions Max 0.999852 +eval/Actions Min -0.99997 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.80514e-06 +time/evaluation sampling (s) 3.15323 +time/exploration sampling (s) 5.18165 +time/logging (s) 0.00669675 +time/saving (s) 0.0112162 +time/training (s) 14.6236 +time/epoch (s) 22.9764 +time/total (s) 24839.8 +Epoch -35 +------------------------------ ---------------- +2022-05-16 00:56:54.713261 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -34 finished +------------------------------ ---------------- +epoch -34 +replay_buffer/size 999047 +trainer/num train calls 967000 +trainer/QF1 Loss 0.770522 +trainer/QF2 Loss 0.506296 +trainer/Policy Loss 30.4714 +trainer/Q1 Predictions Mean -75.6462 +trainer/Q1 Predictions Std 16.2326 +trainer/Q1 Predictions Max -1.2072 +trainer/Q1 Predictions Min -87.5314 +trainer/Q2 Predictions Mean -75.6633 +trainer/Q2 Predictions Std 16.317 +trainer/Q2 Predictions Max -0.989014 +trainer/Q2 Predictions Min -87.8517 +trainer/Q Targets Mean -75.5882 +trainer/Q Targets Std 16.4267 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7395 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00668989 +trainer/policy/mean Std 0.734762 +trainer/policy/mean Max 0.998956 +trainer/policy/mean Min -0.998877 +trainer/policy/std Mean 0.406368 +trainer/policy/std Std 0.0191433 +trainer/policy/std Max 0.428154 +trainer/policy/std Min 0.378095 +trainer/Advantage Weights Mean 6.0897 +trainer/Advantage Weights Std 19.9047 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.76226e-23 +trainer/Advantage Score Mean -0.302517 +trainer/Advantage Score Std 0.753595 +trainer/Advantage Score Max 3.17412 +trainer/Advantage Score Min -5.13987 +trainer/V1 Predictions Mean -75.3296 +trainer/V1 Predictions Std 16.6107 +trainer/V1 Predictions Max 0.32352 +trainer/V1 Predictions Min -87.6408 +trainer/VF Loss 0.124433 +expl/num steps total 967000 +expl/num paths total 1339 +expl/path length Mean 500 +expl/path length Std 391 +expl/path length Max 891 +expl/path length Min 109 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0400218 +expl/Actions Std 0.816013 +expl/Actions Max 2.18935 +expl/Actions Min -2.14672 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 869548 +eval/num paths total 972 +eval/path length Mean 690 +eval/path length Std 0 +eval/path length Max 690 +eval/path length Min 690 +eval/Rewards Mean 0.00144928 +eval/Rewards Std 0.0380418 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.03028 +eval/Actions Std 0.756293 +eval/Actions Max 0.999751 +eval/Actions Min -0.999527 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.63937e-06 +time/evaluation sampling (s) 3.08143 +time/exploration sampling (s) 4.58842 +time/logging (s) 0.0072959 +time/saving (s) 0.0127712 +time/training (s) 15.098 +time/epoch (s) 22.788 +time/total (s) 24862.6 +Epoch -34 +------------------------------ ---------------- +2022-05-16 00:57:17.461451 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -33 finished +------------------------------ ---------------- +epoch -33 +replay_buffer/size 999047 +trainer/num train calls 968000 +trainer/QF1 Loss 1.1093 +trainer/QF2 Loss 1.14353 +trainer/Policy Loss 17.2946 +trainer/Q1 Predictions Mean -75.617 +trainer/Q1 Predictions Std 17.0161 +trainer/Q1 Predictions Max -0.683138 +trainer/Q1 Predictions Min -87.9989 +trainer/Q2 Predictions Mean -75.6699 +trainer/Q2 Predictions Std 17.0354 +trainer/Q2 Predictions Max 0.565551 +trainer/Q2 Predictions Min -88.1651 +trainer/Q Targets Mean -75.6927 +trainer/Q Targets Std 16.8804 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5906 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00320252 +trainer/policy/mean Std 0.740627 +trainer/policy/mean Max 0.999417 +trainer/policy/mean Min -0.999257 +trainer/policy/std Mean 0.406119 +trainer/policy/std Std 0.0185466 +trainer/policy/std Max 0.424956 +trainer/policy/std Min 0.376504 +trainer/Advantage Weights Mean 4.90417 +trainer/Advantage Weights Std 19.4975 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.64746e-11 +trainer/Advantage Score Mean -0.455834 +trainer/Advantage Score Std 0.540212 +trainer/Advantage Score Max 1.52721 +trainer/Advantage Score Min -2.37921 +trainer/V1 Predictions Mean -75.4245 +trainer/V1 Predictions Std 16.9691 +trainer/V1 Predictions Max -0.712535 +trainer/V1 Predictions Min -87.4955 +trainer/VF Loss 0.071791 +expl/num steps total 968000 +expl/num paths total 1341 +expl/path length Mean 500 +expl/path length Std 317 +expl/path length Max 817 +expl/path length Min 183 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0406094 +expl/Actions Std 0.824042 +expl/Actions Max 2.33205 +expl/Actions Min -2.31081 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 870548 +eval/num paths total 973 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.141883 +eval/Actions Std 0.772147 +eval/Actions Max 0.999798 +eval/Actions Min -0.999698 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78512e-06 +time/evaluation sampling (s) 3.58954 +time/exploration sampling (s) 4.20786 +time/logging (s) 0.00720379 +time/saving (s) 0.0100379 +time/training (s) 14.9182 +time/epoch (s) 22.7329 +time/total (s) 24885.3 +Epoch -33 +------------------------------ ---------------- +2022-05-16 00:57:40.450040 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -32 finished +------------------------------ ---------------- +epoch -32 +replay_buffer/size 999047 +trainer/num train calls 969000 +trainer/QF1 Loss 0.587121 +trainer/QF2 Loss 0.608961 +trainer/Policy Loss 33.0249 +trainer/Q1 Predictions Mean -73.6803 +trainer/Q1 Predictions Std 19.1852 +trainer/Q1 Predictions Max -0.165802 +trainer/Q1 Predictions Min -87.2027 +trainer/Q2 Predictions Mean -73.6804 +trainer/Q2 Predictions Std 19.1746 +trainer/Q2 Predictions Max 0.310674 +trainer/Q2 Predictions Min -87.8591 +trainer/Q Targets Mean -73.9318 +trainer/Q Targets Std 19.1643 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6403 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0157442 +trainer/policy/mean Std 0.714264 +trainer/policy/mean Max 0.999703 +trainer/policy/mean Min -0.999902 +trainer/policy/std Mean 0.406749 +trainer/policy/std Std 0.0193312 +trainer/policy/std Max 0.428197 +trainer/policy/std Min 0.376011 +trainer/Advantage Weights Mean 8.02818 +trainer/Advantage Weights Std 21.981 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.88842e-13 +trainer/Advantage Score Mean -0.187681 +trainer/Advantage Score Std 0.580512 +trainer/Advantage Score Max 1.62029 +trainer/Advantage Score Min -2.92979 +trainer/V1 Predictions Mean -73.7114 +trainer/V1 Predictions Std 19.2201 +trainer/V1 Predictions Max 0.209233 +trainer/V1 Predictions Min -87.4213 +trainer/VF Loss 0.079527 +expl/num steps total 969000 +expl/num paths total 1343 +expl/path length Mean 500 +expl/path length Std 72 +expl/path length Max 572 +expl/path length Min 428 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0599921 +expl/Actions Std 0.83503 +expl/Actions Max 2.57628 +expl/Actions Min -2.44576 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 871258 +eval/num paths total 974 +eval/path length Mean 710 +eval/path length Std 0 +eval/path length Max 710 +eval/path length Min 710 +eval/Rewards Mean 0.00140845 +eval/Rewards Std 0.0375029 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0263794 +eval/Actions Std 0.720132 +eval/Actions Max 0.999888 +eval/Actions Min -0.999648 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.20608e-06 +time/evaluation sampling (s) 3.43714 +time/exploration sampling (s) 4.44732 +time/logging (s) 0.00650958 +time/saving (s) 0.0117338 +time/training (s) 15.0722 +time/epoch (s) 22.9749 +time/total (s) 24908.3 +Epoch -32 +------------------------------ ---------------- +2022-05-16 00:58:03.281622 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -31 finished +------------------------------ ---------------- +epoch -31 +replay_buffer/size 999047 +trainer/num train calls 970000 +trainer/QF1 Loss 0.839615 +trainer/QF2 Loss 0.649487 +trainer/Policy Loss 15.3238 +trainer/Q1 Predictions Mean -74.364 +trainer/Q1 Predictions Std 18.4639 +trainer/Q1 Predictions Max -0.1945 +trainer/Q1 Predictions Min -88.8461 +trainer/Q2 Predictions Mean -74.2743 +trainer/Q2 Predictions Std 18.483 +trainer/Q2 Predictions Max -0.407675 +trainer/Q2 Predictions Min -88.645 +trainer/Q Targets Mean -74.015 +trainer/Q Targets Std 18.6725 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.3891 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0347322 +trainer/policy/mean Std 0.732157 +trainer/policy/mean Max 0.999083 +trainer/policy/mean Min -0.999501 +trainer/policy/std Mean 0.4055 +trainer/policy/std Std 0.0191943 +trainer/policy/std Max 0.427322 +trainer/policy/std Min 0.376516 +trainer/Advantage Weights Mean 3.6099 +trainer/Advantage Weights Std 16.1858 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.29333e-27 +trainer/Advantage Score Mean -0.525644 +trainer/Advantage Score Std 0.816476 +trainer/Advantage Score Max 1.67199 +trainer/Advantage Score Min -6.05033 +trainer/V1 Predictions Mean -73.7506 +trainer/V1 Predictions Std 18.7415 +trainer/V1 Predictions Max 0.447015 +trainer/V1 Predictions Min -88.2271 +trainer/VF Loss 0.113997 +expl/num steps total 970000 +expl/num paths total 1345 +expl/path length Mean 500 +expl/path length Std 213 +expl/path length Max 713 +expl/path length Min 287 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0215464 +expl/Actions Std 0.818555 +expl/Actions Max 2.37116 +expl/Actions Min -2.03128 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 872258 +eval/num paths total 975 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.122065 +eval/Actions Std 0.656932 +eval/Actions Max 0.995045 +eval/Actions Min -0.989839 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.75718e-06 +time/evaluation sampling (s) 4.12122 +time/exploration sampling (s) 4.03537 +time/logging (s) 0.00974905 +time/saving (s) 0.0114854 +time/training (s) 14.6423 +time/epoch (s) 22.8201 +time/total (s) 24931.1 +Epoch -31 +------------------------------ ---------------- +2022-05-16 00:58:27.345583 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -30 finished +------------------------------ ---------------- +epoch -30 +replay_buffer/size 999047 +trainer/num train calls 971000 +trainer/QF1 Loss 1.73322 +trainer/QF2 Loss 1.97708 +trainer/Policy Loss 28.352 +trainer/Q1 Predictions Mean -72.9709 +trainer/Q1 Predictions Std 20.1628 +trainer/Q1 Predictions Max -0.263302 +trainer/Q1 Predictions Min -87.7781 +trainer/Q2 Predictions Mean -72.9275 +trainer/Q2 Predictions Std 20.0331 +trainer/Q2 Predictions Max 0.0423762 +trainer/Q2 Predictions Min -87.8166 +trainer/Q Targets Mean -73.0152 +trainer/Q Targets Std 20.4404 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8687 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0105792 +trainer/policy/mean Std 0.723065 +trainer/policy/mean Max 0.999391 +trainer/policy/mean Min -0.99986 +trainer/policy/std Mean 0.406276 +trainer/policy/std Std 0.0196552 +trainer/policy/std Max 0.427203 +trainer/policy/std Min 0.375101 +trainer/Advantage Weights Mean 6.36298 +trainer/Advantage Weights Std 18.7718 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.72464e-15 +trainer/Advantage Score Mean -0.239445 +trainer/Advantage Score Std 0.596726 +trainer/Advantage Score Max 1.85814 +trainer/Advantage Score Min -3.39938 +trainer/V1 Predictions Mean -72.8707 +trainer/V1 Predictions Std 20.3396 +trainer/V1 Predictions Max -0.0338242 +trainer/V1 Predictions Min -87.9085 +trainer/VF Loss 0.074872 +expl/num steps total 971000 +expl/num paths total 1346 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.124419 +expl/Actions Std 0.826042 +expl/Actions Max 2.28218 +expl/Actions Min -2.74319 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 872933 +eval/num paths total 976 +eval/path length Mean 675 +eval/path length Std 0 +eval/path length Max 675 +eval/path length Min 675 +eval/Rewards Mean 0.00148148 +eval/Rewards Std 0.0384615 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0230145 +eval/Actions Std 0.735867 +eval/Actions Max 0.999975 +eval/Actions Min -0.999523 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.54903e-06 +time/evaluation sampling (s) 4.01469 +time/exploration sampling (s) 5.49255 +time/logging (s) 0.00615819 +time/saving (s) 0.0101581 +time/training (s) 14.5238 +time/epoch (s) 24.0473 +time/total (s) 24955.2 +Epoch -30 +------------------------------ ---------------- +2022-05-16 00:58:51.600381 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -29 finished +------------------------------ ---------------- +epoch -29 +replay_buffer/size 999047 +trainer/num train calls 972000 +trainer/QF1 Loss 0.607804 +trainer/QF2 Loss 0.599626 +trainer/Policy Loss 12.0319 +trainer/Q1 Predictions Mean -73.5704 +trainer/Q1 Predictions Std 19.3263 +trainer/Q1 Predictions Max -0.611713 +trainer/Q1 Predictions Min -87.6462 +trainer/Q2 Predictions Mean -73.5758 +trainer/Q2 Predictions Std 19.2813 +trainer/Q2 Predictions Max -0.45683 +trainer/Q2 Predictions Min -87.6406 +trainer/Q Targets Mean -73.4488 +trainer/Q Targets Std 19.5297 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9222 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00946103 +trainer/policy/mean Std 0.733672 +trainer/policy/mean Max 0.99979 +trainer/policy/mean Min -0.998976 +trainer/policy/std Mean 0.409055 +trainer/policy/std Std 0.0190468 +trainer/policy/std Max 0.429197 +trainer/policy/std Min 0.3787 +trainer/Advantage Weights Mean 3.87699 +trainer/Advantage Weights Std 16.7661 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.21463e-19 +trainer/Advantage Score Mean -0.455245 +trainer/Advantage Score Std 0.71629 +trainer/Advantage Score Max 1.72657 +trainer/Advantage Score Min -4.25814 +trainer/V1 Predictions Mean -73.2301 +trainer/V1 Predictions Std 19.5828 +trainer/V1 Predictions Max 0.254282 +trainer/V1 Predictions Min -87.8149 +trainer/VF Loss 0.0975301 +expl/num steps total 972000 +expl/num paths total 1348 +expl/path length Mean 500 +expl/path length Std 176 +expl/path length Max 676 +expl/path length Min 324 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0423102 +expl/Actions Std 0.82085 +expl/Actions Max 2.45444 +expl/Actions Min -2.25141 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 873403 +eval/num paths total 977 +eval/path length Mean 470 +eval/path length Std 0 +eval/path length Max 470 +eval/path length Min 470 +eval/Rewards Mean 0.00212766 +eval/Rewards Std 0.0460775 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.000277173 +eval/Actions Std 0.747334 +eval/Actions Max 0.999596 +eval/Actions Min -0.999641 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.06312e-06 +time/evaluation sampling (s) 3.70157 +time/exploration sampling (s) 5.498 +time/logging (s) 0.00639031 +time/saving (s) 0.0123599 +time/training (s) 15.0255 +time/epoch (s) 24.2438 +time/total (s) 24979.4 +Epoch -29 +------------------------------ ---------------- +2022-05-16 00:59:13.987085 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -28 finished +------------------------------ ---------------- +epoch -28 +replay_buffer/size 999047 +trainer/num train calls 973000 +trainer/QF1 Loss 1.22479 +trainer/QF2 Loss 1.21985 +trainer/Policy Loss 15.6644 +trainer/Q1 Predictions Mean -72.5455 +trainer/Q1 Predictions Std 20.5769 +trainer/Q1 Predictions Max -0.987967 +trainer/Q1 Predictions Min -87.3958 +trainer/Q2 Predictions Mean -72.457 +trainer/Q2 Predictions Std 20.6042 +trainer/Q2 Predictions Max -1.35853 +trainer/Q2 Predictions Min -87.4955 +trainer/Q Targets Mean -72.4898 +trainer/Q Targets Std 20.5117 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3865 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0331313 +trainer/policy/mean Std 0.728991 +trainer/policy/mean Max 0.999196 +trainer/policy/mean Min -0.999747 +trainer/policy/std Mean 0.406918 +trainer/policy/std Std 0.0194906 +trainer/policy/std Max 0.427603 +trainer/policy/std Min 0.37781 +trainer/Advantage Weights Mean 3.85525 +trainer/Advantage Weights Std 15.5725 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.47382e-17 +trainer/Advantage Score Mean -0.365721 +trainer/Advantage Score Std 0.600982 +trainer/Advantage Score Max 1.93783 +trainer/Advantage Score Min -3.82382 +trainer/V1 Predictions Mean -72.1896 +trainer/V1 Predictions Std 20.5917 +trainer/V1 Predictions Max -0.450598 +trainer/V1 Predictions Min -87.3955 +trainer/VF Loss 0.0740439 +expl/num steps total 973000 +expl/num paths total 1349 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0251876 +expl/Actions Std 0.840478 +expl/Actions Max 2.16477 +expl/Actions Min -2.32176 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 874014 +eval/num paths total 978 +eval/path length Mean 611 +eval/path length Std 0 +eval/path length Max 611 +eval/path length Min 611 +eval/Rewards Mean 0.00163666 +eval/Rewards Std 0.0404226 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0275309 +eval/Actions Std 0.755695 +eval/Actions Max 0.999681 +eval/Actions Min -0.999878 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.39676e-06 +time/evaluation sampling (s) 3.04647 +time/exploration sampling (s) 4.88091 +time/logging (s) 0.00648663 +time/saving (s) 0.0118919 +time/training (s) 14.4302 +time/epoch (s) 22.376 +time/total (s) 25001.8 +Epoch -28 +------------------------------ ---------------- +2022-05-16 00:59:37.291274 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -27 finished +------------------------------ ---------------- +epoch -27 +replay_buffer/size 999047 +trainer/num train calls 974000 +trainer/QF1 Loss 0.580688 +trainer/QF2 Loss 0.62862 +trainer/Policy Loss 31.4703 +trainer/Q1 Predictions Mean -74.3223 +trainer/Q1 Predictions Std 17.344 +trainer/Q1 Predictions Max -2.66128 +trainer/Q1 Predictions Min -87.6976 +trainer/Q2 Predictions Mean -74.2795 +trainer/Q2 Predictions Std 17.3827 +trainer/Q2 Predictions Max -2.4171 +trainer/Q2 Predictions Min -87.558 +trainer/Q Targets Mean -74.3947 +trainer/Q Targets Std 17.3802 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7879 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0109127 +trainer/policy/mean Std 0.749569 +trainer/policy/mean Max 0.999726 +trainer/policy/mean Min -0.999601 +trainer/policy/std Mean 0.406278 +trainer/policy/std Std 0.0192427 +trainer/policy/std Max 0.426917 +trainer/policy/std Min 0.37543 +trainer/Advantage Weights Mean 7.46688 +trainer/Advantage Weights Std 22.9672 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.66144e-15 +trainer/Advantage Score Mean -0.215505 +trainer/Advantage Score Std 0.562779 +trainer/Advantage Score Max 2.02993 +trainer/Advantage Score Min -3.26424 +trainer/V1 Predictions Mean -74.2342 +trainer/V1 Predictions Std 17.3235 +trainer/V1 Predictions Max -1.11383 +trainer/V1 Predictions Min -87.6554 +trainer/VF Loss 0.0876673 +expl/num steps total 974000 +expl/num paths total 1351 +expl/path length Mean 500 +expl/path length Std 135 +expl/path length Max 635 +expl/path length Min 365 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0260609 +expl/Actions Std 0.835111 +expl/Actions Max 2.28508 +expl/Actions Min -2.28772 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 874420 +eval/num paths total 979 +eval/path length Mean 406 +eval/path length Std 0 +eval/path length Max 406 +eval/path length Min 406 +eval/Rewards Mean 0.00246305 +eval/Rewards Std 0.049568 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00647777 +eval/Actions Std 0.761902 +eval/Actions Max 0.999805 +eval/Actions Min -0.999813 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 3.25264e-06 +time/evaluation sampling (s) 3.54609 +time/exploration sampling (s) 4.8263 +time/logging (s) 0.00652325 +time/saving (s) 0.0128505 +time/training (s) 14.8977 +time/epoch (s) 23.2894 +time/total (s) 25025.1 +Epoch -27 +------------------------------ ---------------- +2022-05-16 01:00:00.428148 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -26 finished +------------------------------ ---------------- +epoch -26 +replay_buffer/size 999047 +trainer/num train calls 975000 +trainer/QF1 Loss 0.421093 +trainer/QF2 Loss 0.467054 +trainer/Policy Loss 18.5866 +trainer/Q1 Predictions Mean -76.5984 +trainer/Q1 Predictions Std 15.8003 +trainer/Q1 Predictions Max -0.732868 +trainer/Q1 Predictions Min -87.7926 +trainer/Q2 Predictions Mean -76.6116 +trainer/Q2 Predictions Std 15.7649 +trainer/Q2 Predictions Max -0.63792 +trainer/Q2 Predictions Min -87.7336 +trainer/Q Targets Mean -76.6035 +trainer/Q Targets Std 15.8306 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9532 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0294215 +trainer/policy/mean Std 0.745559 +trainer/policy/mean Max 0.999931 +trainer/policy/mean Min -0.99965 +trainer/policy/std Mean 0.407414 +trainer/policy/std Std 0.0190966 +trainer/policy/std Max 0.429672 +trainer/policy/std Min 0.379954 +trainer/Advantage Weights Mean 5.26354 +trainer/Advantage Weights Std 19.8648 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.20365e-13 +trainer/Advantage Score Mean -0.315931 +trainer/Advantage Score Std 0.482856 +trainer/Advantage Score Max 1.24256 +trainer/Advantage Score Min -2.97483 +trainer/V1 Predictions Mean -76.3778 +trainer/V1 Predictions Std 15.8354 +trainer/V1 Predictions Max 1.25657 +trainer/V1 Predictions Min -87.6697 +trainer/VF Loss 0.0534286 +expl/num steps total 975000 +expl/num paths total 1353 +expl/path length Mean 500 +expl/path length Std 163 +expl/path length Max 663 +expl/path length Min 337 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0237007 +expl/Actions Std 0.832786 +expl/Actions Max 2.2925 +expl/Actions Min -2.37117 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 875016 +eval/num paths total 980 +eval/path length Mean 596 +eval/path length Std 0 +eval/path length Max 596 +eval/path length Min 596 +eval/Rewards Mean 0.00167785 +eval/Rewards Std 0.0409272 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0369106 +eval/Actions Std 0.749163 +eval/Actions Max 0.999786 +eval/Actions Min -0.999791 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.96999e-06 +time/evaluation sampling (s) 3.43245 +time/exploration sampling (s) 4.51265 +time/logging (s) 0.00560998 +time/saving (s) 0.00898578 +time/training (s) 15.1612 +time/epoch (s) 23.1209 +time/total (s) 25048.2 +Epoch -26 +------------------------------ ---------------- +2022-05-16 01:00:23.526037 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -25 finished +------------------------------ ---------------- +epoch -25 +replay_buffer/size 999047 +trainer/num train calls 976000 +trainer/QF1 Loss 0.979478 +trainer/QF2 Loss 0.829396 +trainer/Policy Loss 43.976 +trainer/Q1 Predictions Mean -74.6728 +trainer/Q1 Predictions Std 18.6143 +trainer/Q1 Predictions Max -0.195382 +trainer/Q1 Predictions Min -88.2064 +trainer/Q2 Predictions Mean -74.6425 +trainer/Q2 Predictions Std 18.5656 +trainer/Q2 Predictions Max 0.36794 +trainer/Q2 Predictions Min -87.7673 +trainer/Q Targets Mean -75.0906 +trainer/Q Targets Std 18.6954 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.6012 +trainer/rewards Mean -0.984375 +trainer/rewards Std 0.12402 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.015625 +trainer/terminals Std 0.12402 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0335596 +trainer/policy/mean Std 0.747022 +trainer/policy/mean Max 0.999447 +trainer/policy/mean Min -0.999209 +trainer/policy/std Mean 0.406825 +trainer/policy/std Std 0.0196123 +trainer/policy/std Max 0.42841 +trainer/policy/std Min 0.376424 +trainer/Advantage Weights Mean 11.7849 +trainer/Advantage Weights Std 26.6136 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.04161e-11 +trainer/Advantage Score Mean -0.112214 +trainer/Advantage Score Std 0.483876 +trainer/Advantage Score Max 1.00906 +trainer/Advantage Score Min -2.52877 +trainer/V1 Predictions Mean -74.9159 +trainer/V1 Predictions Std 18.6924 +trainer/V1 Predictions Max 0.603866 +trainer/V1 Predictions Min -88.5387 +trainer/VF Loss 0.0648438 +expl/num steps total 976000 +expl/num paths total 1355 +expl/path length Mean 500 +expl/path length Std 215 +expl/path length Max 715 +expl/path length Min 285 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0177455 +expl/Actions Std 0.842685 +expl/Actions Max 2.27016 +expl/Actions Min -2.41087 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 875573 +eval/num paths total 981 +eval/path length Mean 557 +eval/path length Std 0 +eval/path length Max 557 +eval/path length Min 557 +eval/Rewards Mean 0.00179533 +eval/Rewards Std 0.0423333 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0490431 +eval/Actions Std 0.748449 +eval/Actions Max 0.999942 +eval/Actions Min -0.999579 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.54717e-06 +time/evaluation sampling (s) 4.11212 +time/exploration sampling (s) 4.09488 +time/logging (s) 0.0055849 +time/saving (s) 0.00926272 +time/training (s) 14.8655 +time/epoch (s) 23.0873 +time/total (s) 25071.3 +Epoch -25 +------------------------------ ---------------- +2022-05-16 01:00:47.268087 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -24 finished +------------------------------ ---------------- +epoch -24 +replay_buffer/size 999047 +trainer/num train calls 977000 +trainer/QF1 Loss 0.47591 +trainer/QF2 Loss 0.447403 +trainer/Policy Loss 10.2498 +trainer/Q1 Predictions Mean -75.9554 +trainer/Q1 Predictions Std 16.2051 +trainer/Q1 Predictions Max -0.876129 +trainer/Q1 Predictions Min -88.0549 +trainer/Q2 Predictions Mean -75.893 +trainer/Q2 Predictions Std 16.1721 +trainer/Q2 Predictions Max -0.650806 +trainer/Q2 Predictions Min -87.8923 +trainer/Q Targets Mean -75.8665 +trainer/Q Targets Std 16.4105 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2932 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.036434 +trainer/policy/mean Std 0.736753 +trainer/policy/mean Max 0.999758 +trainer/policy/mean Min -0.999328 +trainer/policy/std Mean 0.408 +trainer/policy/std Std 0.019584 +trainer/policy/std Max 0.431354 +trainer/policy/std Min 0.377743 +trainer/Advantage Weights Mean 3.69675 +trainer/Advantage Weights Std 16.0313 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 6.27008e-21 +trainer/Advantage Score Mean -0.401458 +trainer/Advantage Score Std 0.603221 +trainer/Advantage Score Max 0.970291 +trainer/Advantage Score Min -4.65185 +trainer/V1 Predictions Mean -75.6943 +trainer/V1 Predictions Std 16.4194 +trainer/V1 Predictions Max -0.459142 +trainer/V1 Predictions Min -87.929 +trainer/VF Loss 0.0649323 +expl/num steps total 977000 +expl/num paths total 1357 +expl/path length Mean 500 +expl/path length Std 56 +expl/path length Max 556 +expl/path length Min 444 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0208264 +expl/Actions Std 0.840133 +expl/Actions Max 2.37634 +expl/Actions Min -2.1466 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 876573 +eval/num paths total 982 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0959576 +eval/Actions Std 0.719834 +eval/Actions Max 0.999894 +eval/Actions Min -0.999954 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 7.74488e-06 +time/evaluation sampling (s) 4.15583 +time/exploration sampling (s) 5.24988 +time/logging (s) 0.00735261 +time/saving (s) 0.00977851 +time/training (s) 14.3106 +time/epoch (s) 23.7334 +time/total (s) 25095.1 +Epoch -24 +------------------------------ ---------------- +2022-05-16 01:01:10.855122 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -23 finished +------------------------------ ---------------- +epoch -23 +replay_buffer/size 999047 +trainer/num train calls 978000 +trainer/QF1 Loss 1.05033 +trainer/QF2 Loss 1.0017 +trainer/Policy Loss 2.36765 +trainer/Q1 Predictions Mean -76.6687 +trainer/Q1 Predictions Std 16.0769 +trainer/Q1 Predictions Max -0.906437 +trainer/Q1 Predictions Min -88.5541 +trainer/Q2 Predictions Mean -76.6266 +trainer/Q2 Predictions Std 16.1075 +trainer/Q2 Predictions Max -0.226195 +trainer/Q2 Predictions Min -88.4908 +trainer/Q Targets Mean -75.9985 +trainer/Q Targets Std 16.2861 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8821 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00855966 +trainer/policy/mean Std 0.7388 +trainer/policy/mean Max 0.998943 +trainer/policy/mean Min -0.99976 +trainer/policy/std Mean 0.409348 +trainer/policy/std Std 0.0196898 +trainer/policy/std Max 0.432374 +trainer/policy/std Min 0.380147 +trainer/Advantage Weights Mean 0.895101 +trainer/Advantage Weights Std 8.80762 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.79325e-26 +trainer/Advantage Score Mean -0.689176 +trainer/Advantage Score Std 0.688083 +trainer/Advantage Score Max 1.01421 +trainer/Advantage Score Min -5.92832 +trainer/V1 Predictions Mean -75.7415 +trainer/V1 Predictions Std 16.4405 +trainer/V1 Predictions Max 0.20959 +trainer/V1 Predictions Min -87.7639 +trainer/VF Loss 0.0990129 +expl/num steps total 978000 +expl/num paths total 1359 +expl/path length Mean 500 +expl/path length Std 269 +expl/path length Max 769 +expl/path length Min 231 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0111744 +expl/Actions Std 0.832387 +expl/Actions Max 2.41131 +expl/Actions Min -2.0857 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 877573 +eval/num paths total 983 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0953603 +eval/Actions Std 0.749823 +eval/Actions Max 0.999873 +eval/Actions Min -0.99943 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.65799e-06 +time/evaluation sampling (s) 3.607 +time/exploration sampling (s) 5.3206 +time/logging (s) 0.00734825 +time/saving (s) 0.0109594 +time/training (s) 14.6284 +time/epoch (s) 23.5743 +time/total (s) 25118.7 +Epoch -23 +------------------------------ ---------------- +2022-05-16 01:01:34.146575 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -22 finished +------------------------------ ---------------- +epoch -22 +replay_buffer/size 999047 +trainer/num train calls 979000 +trainer/QF1 Loss 0.626168 +trainer/QF2 Loss 0.55023 +trainer/Policy Loss 17.8812 +trainer/Q1 Predictions Mean -75.3204 +trainer/Q1 Predictions Std 17.7165 +trainer/Q1 Predictions Max -1.22928 +trainer/Q1 Predictions Min -88.2945 +trainer/Q2 Predictions Mean -75.309 +trainer/Q2 Predictions Std 17.6877 +trainer/Q2 Predictions Max -0.454589 +trainer/Q2 Predictions Min -88.789 +trainer/Q Targets Mean -75.5381 +trainer/Q Targets Std 17.6307 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.4284 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00185981 +trainer/policy/mean Std 0.735794 +trainer/policy/mean Max 0.999381 +trainer/policy/mean Min -0.997321 +trainer/policy/std Mean 0.40724 +trainer/policy/std Std 0.0199881 +trainer/policy/std Max 0.42806 +trainer/policy/std Min 0.375201 +trainer/Advantage Weights Mean 4.66522 +trainer/Advantage Weights Std 17.3534 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.30826e-11 +trainer/Advantage Score Mean -0.271158 +trainer/Advantage Score Std 0.484511 +trainer/Advantage Score Max 1.32251 +trainer/Advantage Score Min -2.50597 +trainer/V1 Predictions Mean -75.3168 +trainer/V1 Predictions Std 17.7337 +trainer/V1 Predictions Max -1.95952 +trainer/V1 Predictions Min -88.3146 +trainer/VF Loss 0.0502611 +expl/num steps total 979000 +expl/num paths total 1361 +expl/path length Mean 500 +expl/path length Std 321 +expl/path length Max 821 +expl/path length Min 179 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0124628 +expl/Actions Std 0.832074 +expl/Actions Max 2.22683 +expl/Actions Min -2.31485 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 878573 +eval/num paths total 984 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.182045 +eval/Actions Std 0.714682 +eval/Actions Max 0.999602 +eval/Actions Min -0.999088 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.26502e-06 +time/evaluation sampling (s) 3.47873 +time/exploration sampling (s) 4.84062 +time/logging (s) 0.00720652 +time/saving (s) 0.0101937 +time/training (s) 14.941 +time/epoch (s) 23.2778 +time/total (s) 25141.9 +Epoch -22 +------------------------------ ---------------- +2022-05-16 01:01:56.545230 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -21 finished +------------------------------ ---------------- +epoch -21 +replay_buffer/size 999047 +trainer/num train calls 980000 +trainer/QF1 Loss 0.663044 +trainer/QF2 Loss 0.673979 +trainer/Policy Loss 20.6768 +trainer/Q1 Predictions Mean -75.9493 +trainer/Q1 Predictions Std 16.1805 +trainer/Q1 Predictions Max -4.96686 +trainer/Q1 Predictions Min -88.2888 +trainer/Q2 Predictions Mean -75.9355 +trainer/Q2 Predictions Std 16.2367 +trainer/Q2 Predictions Max -4.82277 +trainer/Q2 Predictions Min -88.3747 +trainer/Q Targets Mean -76.2113 +trainer/Q Targets Std 16.3153 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6741 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0195709 +trainer/policy/mean Std 0.737658 +trainer/policy/mean Max 0.999178 +trainer/policy/mean Min -0.999351 +trainer/policy/std Mean 0.406218 +trainer/policy/std Std 0.0200501 +trainer/policy/std Max 0.427818 +trainer/policy/std Min 0.373673 +trainer/Advantage Weights Mean 4.98787 +trainer/Advantage Weights Std 18.375 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.30441e-18 +trainer/Advantage Score Mean -0.269039 +trainer/Advantage Score Std 0.590369 +trainer/Advantage Score Max 1.64917 +trainer/Advantage Score Min -4.06117 +trainer/V1 Predictions Mean -75.9371 +trainer/V1 Predictions Std 16.5413 +trainer/V1 Predictions Max -1.39066 +trainer/V1 Predictions Min -87.8835 +trainer/VF Loss 0.0690698 +expl/num steps total 980000 +expl/num paths total 1363 +expl/path length Mean 500 +expl/path length Std 274 +expl/path length Max 774 +expl/path length Min 226 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.00839956 +expl/Actions Std 0.829512 +expl/Actions Max 2.37265 +expl/Actions Min -2.20851 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 879573 +eval/num paths total 985 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.00140325 +eval/Actions Std 0.717333 +eval/Actions Max 0.999984 +eval/Actions Min -0.999935 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 3.32482e-06 +time/evaluation sampling (s) 3.6085 +time/exploration sampling (s) 4.4403 +time/logging (s) 0.00900034 +time/saving (s) 0.0142125 +time/training (s) 14.3155 +time/epoch (s) 22.3875 +time/total (s) 25164.3 +Epoch -21 +------------------------------ ---------------- +2022-05-16 01:02:19.897996 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -20 finished +------------------------------ ---------------- +epoch -20 +replay_buffer/size 999047 +trainer/num train calls 981000 +trainer/QF1 Loss 1.24603 +trainer/QF2 Loss 1.26171 +trainer/Policy Loss 13.1765 +trainer/Q1 Predictions Mean -76.4203 +trainer/Q1 Predictions Std 15.9942 +trainer/Q1 Predictions Max -1.92545 +trainer/Q1 Predictions Min -88.6397 +trainer/Q2 Predictions Mean -76.421 +trainer/Q2 Predictions Std 16.049 +trainer/Q2 Predictions Max -1.74269 +trainer/Q2 Predictions Min -88.6084 +trainer/Q Targets Mean -76.0756 +trainer/Q Targets Std 16.4853 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.2736 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0157064 +trainer/policy/mean Std 0.746483 +trainer/policy/mean Max 0.999853 +trainer/policy/mean Min -0.998686 +trainer/policy/std Mean 0.404548 +trainer/policy/std Std 0.0199867 +trainer/policy/std Max 0.426283 +trainer/policy/std Min 0.373237 +trainer/Advantage Weights Mean 5.22442 +trainer/Advantage Weights Std 18.2066 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.57108e-19 +trainer/Advantage Score Mean -0.326305 +trainer/Advantage Score Std 0.637578 +trainer/Advantage Score Max 1.46828 +trainer/Advantage Score Min -4.22294 +trainer/V1 Predictions Mean -75.9071 +trainer/V1 Predictions Std 16.4674 +trainer/V1 Predictions Max -1.51968 +trainer/V1 Predictions Min -88.1553 +trainer/VF Loss 0.0840854 +expl/num steps total 981000 +expl/num paths total 1364 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0402165 +expl/Actions Std 0.848422 +expl/Actions Max 2.28955 +expl/Actions Min -2.24397 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 880573 +eval/num paths total 986 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0701127 +eval/Actions Std 0.725552 +eval/Actions Max 0.999986 +eval/Actions Min -0.999575 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.78885e-06 +time/evaluation sampling (s) 4.06468 +time/exploration sampling (s) 4.2995 +time/logging (s) 0.00803944 +time/saving (s) 0.0128489 +time/training (s) 14.9514 +time/epoch (s) 23.3365 +time/total (s) 25187.7 +Epoch -20 +------------------------------ ---------------- +2022-05-16 01:02:42.229302 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -19 finished +------------------------------ ---------------- +epoch -19 +replay_buffer/size 999047 +trainer/num train calls 982000 +trainer/QF1 Loss 0.7728 +trainer/QF2 Loss 0.8447 +trainer/Policy Loss 15.6995 +trainer/Q1 Predictions Mean -74.5864 +trainer/Q1 Predictions Std 17.4696 +trainer/Q1 Predictions Max -1.29815 +trainer/Q1 Predictions Min -87.8963 +trainer/Q2 Predictions Mean -74.5817 +trainer/Q2 Predictions Std 17.4582 +trainer/Q2 Predictions Max -1.12232 +trainer/Q2 Predictions Min -87.7199 +trainer/Q Targets Mean -74.7562 +trainer/Q Targets Std 17.435 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.7638 +trainer/rewards Mean -0.988281 +trainer/rewards Std 0.107617 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0117188 +trainer/terminals Std 0.107617 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0022361 +trainer/policy/mean Std 0.726454 +trainer/policy/mean Max 0.998949 +trainer/policy/mean Min -0.998456 +trainer/policy/std Mean 0.405253 +trainer/policy/std Std 0.0189593 +trainer/policy/std Max 0.425402 +trainer/policy/std Min 0.374729 +trainer/Advantage Weights Mean 5.55817 +trainer/Advantage Weights Std 19.4543 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.51141e-20 +trainer/Advantage Score Mean -0.270055 +trainer/Advantage Score Std 0.560079 +trainer/Advantage Score Max 1.15659 +trainer/Advantage Score Min -4.56387 +trainer/V1 Predictions Mean -74.5418 +trainer/V1 Predictions Std 17.5036 +trainer/V1 Predictions Max -1.50922 +trainer/V1 Predictions Min -87.6643 +trainer/VF Loss 0.0621125 +expl/num steps total 982000 +expl/num paths total 1365 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.017674 +expl/Actions Std 0.83318 +expl/Actions Max 2.41783 +expl/Actions Min -2.32819 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 881097 +eval/num paths total 987 +eval/path length Mean 524 +eval/path length Std 0 +eval/path length Max 524 +eval/path length Min 524 +eval/Rewards Mean 0.0019084 +eval/Rewards Std 0.0436435 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0135836 +eval/Actions Std 0.741197 +eval/Actions Max 0.999397 +eval/Actions Min -0.99997 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.54391e-06 +time/evaluation sampling (s) 3.32596 +time/exploration sampling (s) 4.65678 +time/logging (s) 0.00606281 +time/saving (s) 0.0156676 +time/training (s) 14.3097 +time/epoch (s) 22.3142 +time/total (s) 25210 +Epoch -19 +------------------------------ ---------------- +2022-05-16 01:03:05.297251 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -18 finished +------------------------------ ---------------- +epoch -18 +replay_buffer/size 999047 +trainer/num train calls 983000 +trainer/QF1 Loss 0.615021 +trainer/QF2 Loss 0.664551 +trainer/Policy Loss 6.37366 +trainer/Q1 Predictions Mean -74.3788 +trainer/Q1 Predictions Std 18.2847 +trainer/Q1 Predictions Max -0.200057 +trainer/Q1 Predictions Min -88.6981 +trainer/Q2 Predictions Mean -74.3595 +trainer/Q2 Predictions Std 18.2056 +trainer/Q2 Predictions Max 0.291498 +trainer/Q2 Predictions Min -88.5711 +trainer/Q Targets Mean -74.4347 +trainer/Q Targets Std 18.3462 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.6053 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0001075 +trainer/policy/mean Std 0.734235 +trainer/policy/mean Max 0.999082 +trainer/policy/mean Min -0.999038 +trainer/policy/std Mean 0.406387 +trainer/policy/std Std 0.0197194 +trainer/policy/std Max 0.427014 +trainer/policy/std Min 0.372973 +trainer/Advantage Weights Mean 2.15609 +trainer/Advantage Weights Std 11.78 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.55976e-17 +trainer/Advantage Score Mean -0.390258 +trainer/Advantage Score Std 0.530068 +trainer/Advantage Score Max 2.09966 +trainer/Advantage Score Min -3.78743 +trainer/V1 Predictions Mean -74.2256 +trainer/V1 Predictions Std 18.3803 +trainer/V1 Predictions Max 0.167558 +trainer/V1 Predictions Min -88.6298 +trainer/VF Loss 0.0614349 +expl/num steps total 983000 +expl/num paths total 1367 +expl/path length Mean 500 +expl/path length Std 228 +expl/path length Max 728 +expl/path length Min 272 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0239858 +expl/Actions Std 0.82821 +expl/Actions Max 2.12928 +expl/Actions Min -2.1698 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 882097 +eval/num paths total 988 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0982666 +eval/Actions Std 0.739527 +eval/Actions Max 0.999874 +eval/Actions Min -0.999952 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.99793e-06 +time/evaluation sampling (s) 3.67584 +time/exploration sampling (s) 4.70975 +time/logging (s) 0.00836988 +time/saving (s) 0.0130989 +time/training (s) 14.6484 +time/epoch (s) 23.0554 +time/total (s) 25233 +Epoch -18 +------------------------------ ---------------- +2022-05-16 01:03:28.200150 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -17 finished +------------------------------ ---------------- +epoch -17 +replay_buffer/size 999047 +trainer/num train calls 984000 +trainer/QF1 Loss 3.8246 +trainer/QF2 Loss 3.64908 +trainer/Policy Loss 25.5377 +trainer/Q1 Predictions Mean -70.9539 +trainer/Q1 Predictions Std 20.5975 +trainer/Q1 Predictions Max -0.575455 +trainer/Q1 Predictions Min -87.9808 +trainer/Q2 Predictions Mean -71.0221 +trainer/Q2 Predictions Std 20.4135 +trainer/Q2 Predictions Max -0.407493 +trainer/Q2 Predictions Min -88.141 +trainer/Q Targets Mean -71.1446 +trainer/Q Targets Std 20.5815 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.4497 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0218955 +trainer/policy/mean Std 0.733015 +trainer/policy/mean Max 0.999044 +trainer/policy/mean Min -0.999186 +trainer/policy/std Mean 0.407678 +trainer/policy/std Std 0.0192654 +trainer/policy/std Max 0.430271 +trainer/policy/std Min 0.377127 +trainer/Advantage Weights Mean 6.54064 +trainer/Advantage Weights Std 22.2037 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 2.5095e-14 +trainer/Advantage Score Mean -0.353861 +trainer/Advantage Score Std 0.640285 +trainer/Advantage Score Max 2.50111 +trainer/Advantage Score Min -3.13161 +trainer/V1 Predictions Mean -70.9807 +trainer/V1 Predictions Std 20.6013 +trainer/V1 Predictions Max -0.679344 +trainer/V1 Predictions Min -88.2832 +trainer/VF Loss 0.114982 +expl/num steps total 984000 +expl/num paths total 1369 +expl/path length Mean 500 +expl/path length Std 158 +expl/path length Max 658 +expl/path length Min 342 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0362865 +expl/Actions Std 0.826945 +expl/Actions Max 2.40698 +expl/Actions Min -2.27228 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 882908 +eval/num paths total 989 +eval/path length Mean 811 +eval/path length Std 0 +eval/path length Max 811 +eval/path length Min 811 +eval/Rewards Mean 0.00123305 +eval/Rewards Std 0.0350931 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0282944 +eval/Actions Std 0.737996 +eval/Actions Max 0.99997 +eval/Actions Min -0.999427 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.40002e-06 +time/evaluation sampling (s) 3.02138 +time/exploration sampling (s) 5.1391 +time/logging (s) 0.00644831 +time/saving (s) 0.00935184 +time/training (s) 14.7079 +time/epoch (s) 22.8841 +time/total (s) 25255.9 +Epoch -17 +------------------------------ ---------------- +2022-05-16 01:03:50.133019 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -16 finished +------------------------------ ---------------- +epoch -16 +replay_buffer/size 999047 +trainer/num train calls 985000 +trainer/QF1 Loss 0.660964 +trainer/QF2 Loss 0.726054 +trainer/Policy Loss 8.08923 +trainer/Q1 Predictions Mean -73.1981 +trainer/Q1 Predictions Std 18.7691 +trainer/Q1 Predictions Max -0.455701 +trainer/Q1 Predictions Min -87.3364 +trainer/Q2 Predictions Mean -73.24 +trainer/Q2 Predictions Std 18.7045 +trainer/Q2 Predictions Max -0.316185 +trainer/Q2 Predictions Min -87.3176 +trainer/Q Targets Mean -73.2156 +trainer/Q Targets Std 18.7927 +trainer/Q Targets Max 0.323147 +trainer/Q Targets Min -87.6321 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0121246 +trainer/policy/mean Std 0.730337 +trainer/policy/mean Max 0.999237 +trainer/policy/mean Min -0.998974 +trainer/policy/std Mean 0.406083 +trainer/policy/std Std 0.0194918 +trainer/policy/std Max 0.427342 +trainer/policy/std Min 0.375866 +trainer/Advantage Weights Mean 2.28313 +trainer/Advantage Weights Std 12.6966 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.27105e-24 +trainer/Advantage Score Mean -0.645491 +trainer/Advantage Score Std 0.698029 +trainer/Advantage Score Max 1.00031 +trainer/Advantage Score Min -5.32781 +trainer/V1 Predictions Mean -72.8912 +trainer/V1 Predictions Std 18.9709 +trainer/V1 Predictions Max 0.444111 +trainer/V1 Predictions Min -87.6348 +trainer/VF Loss 0.0987963 +expl/num steps total 985000 +expl/num paths total 1371 +expl/path length Mean 500 +expl/path length Std 245 +expl/path length Max 745 +expl/path length Min 255 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 3.44658e-05 +expl/Actions Std 0.835245 +expl/Actions Max 2.34928 +expl/Actions Min -2.34175 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 883904 +eval/num paths total 991 +eval/path length Mean 498 +eval/path length Std 28 +eval/path length Max 526 +eval/path length Min 470 +eval/Rewards Mean 0.00200803 +eval/Rewards Std 0.0447661 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0235162 +eval/Actions Std 0.736128 +eval/Actions Max 0.999942 +eval/Actions Min -0.999947 +eval/Num Paths 2 +eval/Average Returns 1 +time/data storing (s) 2.55182e-06 +time/evaluation sampling (s) 3.40532 +time/exploration sampling (s) 3.86035 +time/logging (s) 0.00870148 +time/saving (s) 0.0165748 +time/training (s) 14.6318 +time/epoch (s) 21.9227 +time/total (s) 25277.9 +Epoch -16 +------------------------------ ---------------- +2022-05-16 01:04:14.080319 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -15 finished +------------------------------ ---------------- +epoch -15 +replay_buffer/size 999047 +trainer/num train calls 986000 +trainer/QF1 Loss 0.440842 +trainer/QF2 Loss 0.539794 +trainer/Policy Loss 14.6986 +trainer/Q1 Predictions Mean -74.9322 +trainer/Q1 Predictions Std 18.6904 +trainer/Q1 Predictions Max -0.644341 +trainer/Q1 Predictions Min -88.0598 +trainer/Q2 Predictions Mean -74.9089 +trainer/Q2 Predictions Std 18.765 +trainer/Q2 Predictions Max 0.503906 +trainer/Q2 Predictions Min -87.9185 +trainer/Q Targets Mean -75.149 +trainer/Q Targets Std 18.8677 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9751 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00746011 +trainer/policy/mean Std 0.739728 +trainer/policy/mean Max 0.999282 +trainer/policy/mean Min -0.999974 +trainer/policy/std Mean 0.406092 +trainer/policy/std Std 0.0205395 +trainer/policy/std Max 0.430197 +trainer/policy/std Min 0.376759 +trainer/Advantage Weights Mean 5.07234 +trainer/Advantage Weights Std 18.368 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.72995e-13 +trainer/Advantage Score Mean -0.29026 +trainer/Advantage Score Std 0.562421 +trainer/Advantage Score Max 1.95183 +trainer/Advantage Score Min -2.86172 +trainer/V1 Predictions Mean -74.9341 +trainer/V1 Predictions Std 18.9243 +trainer/V1 Predictions Max -0.467122 +trainer/V1 Predictions Min -87.9951 +trainer/VF Loss 0.082243 +expl/num steps total 986000 +expl/num paths total 1372 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.463631 +expl/Actions Std 0.748284 +expl/Actions Max 2.31945 +expl/Actions Min -2.25877 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 884904 +eval/num paths total 992 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.0201171 +eval/Actions Std 0.717187 +eval/Actions Max 0.999493 +eval/Actions Min -0.999608 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.7162e-06 +time/evaluation sampling (s) 3.64955 +time/exploration sampling (s) 5.24074 +time/logging (s) 0.00663562 +time/saving (s) 0.00958753 +time/training (s) 15.0208 +time/epoch (s) 23.9273 +time/total (s) 25301.8 +Epoch -15 +------------------------------ ---------------- +2022-05-16 01:04:37.518127 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -14 finished +------------------------------ ---------------- +epoch -14 +replay_buffer/size 999047 +trainer/num train calls 987000 +trainer/QF1 Loss 0.664646 +trainer/QF2 Loss 0.547465 +trainer/Policy Loss 6.3787 +trainer/Q1 Predictions Mean -73.7418 +trainer/Q1 Predictions Std 18.7731 +trainer/Q1 Predictions Max -1.34615 +trainer/Q1 Predictions Min -87.9883 +trainer/Q2 Predictions Mean -73.7013 +trainer/Q2 Predictions Std 18.8832 +trainer/Q2 Predictions Max -1.05307 +trainer/Q2 Predictions Min -88.0661 +trainer/Q Targets Mean -73.6888 +trainer/Q Targets Std 18.8583 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8555 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0053318 +trainer/policy/mean Std 0.743999 +trainer/policy/mean Max 0.999737 +trainer/policy/mean Min -0.999586 +trainer/policy/std Mean 0.407481 +trainer/policy/std Std 0.0203456 +trainer/policy/std Max 0.430537 +trainer/policy/std Min 0.379886 +trainer/Advantage Weights Mean 3.55297 +trainer/Advantage Weights Std 16.2012 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.46361e-27 +trainer/Advantage Score Mean -0.496985 +trainer/Advantage Score Std 0.762747 +trainer/Advantage Score Max 1.3077 +trainer/Advantage Score Min -6.04717 +trainer/V1 Predictions Mean -73.4667 +trainer/V1 Predictions Std 19.1229 +trainer/V1 Predictions Max -0.36533 +trainer/V1 Predictions Min -87.835 +trainer/VF Loss 0.0996739 +expl/num steps total 987000 +expl/num paths total 1374 +expl/path length Mean 500 +expl/path length Std 234 +expl/path length Max 734 +expl/path length Min 266 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0061272 +expl/Actions Std 0.820419 +expl/Actions Max 2.08394 +expl/Actions Min -2.35158 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 885591 +eval/num paths total 993 +eval/path length Mean 687 +eval/path length Std 0 +eval/path length Max 687 +eval/path length Min 687 +eval/Rewards Mean 0.0014556 +eval/Rewards Std 0.0381246 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0444736 +eval/Actions Std 0.748885 +eval/Actions Max 0.999771 +eval/Actions Min -0.999728 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 8.54582e-06 +time/evaluation sampling (s) 3.70144 +time/exploration sampling (s) 5.15551 +time/logging (s) 0.00703163 +time/saving (s) 0.0125305 +time/training (s) 14.5476 +time/epoch (s) 23.4241 +time/total (s) 25325.2 +Epoch -14 +------------------------------ ---------------- +2022-05-16 01:04:59.905746 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -13 finished +------------------------------ ---------------- +epoch -13 +replay_buffer/size 999047 +trainer/num train calls 988000 +trainer/QF1 Loss 0.749611 +trainer/QF2 Loss 0.657941 +trainer/Policy Loss 27.5836 +trainer/Q1 Predictions Mean -75.6252 +trainer/Q1 Predictions Std 16.7642 +trainer/Q1 Predictions Max -1.54889 +trainer/Q1 Predictions Min -88.1247 +trainer/Q2 Predictions Mean -75.6632 +trainer/Q2 Predictions Std 16.8443 +trainer/Q2 Predictions Max -1.39579 +trainer/Q2 Predictions Min -87.9299 +trainer/Q Targets Mean -75.7491 +trainer/Q Targets Std 17.1828 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.0519 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00302017 +trainer/policy/mean Std 0.718848 +trainer/policy/mean Max 0.999992 +trainer/policy/mean Min -0.999108 +trainer/policy/std Mean 0.406023 +trainer/policy/std Std 0.019911 +trainer/policy/std Max 0.429673 +trainer/policy/std Min 0.377735 +trainer/Advantage Weights Mean 6.11727 +trainer/Advantage Weights Std 20.085 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 9.2242e-18 +trainer/Advantage Score Mean -0.293796 +trainer/Advantage Score Std 0.651879 +trainer/Advantage Score Max 0.993362 +trainer/Advantage Score Min -3.92247 +trainer/V1 Predictions Mean -75.5499 +trainer/V1 Predictions Std 17.1538 +trainer/V1 Predictions Max -0.0404242 +trainer/V1 Predictions Min -87.9319 +trainer/VF Loss 0.0730356 +expl/num steps total 988000 +expl/num paths total 1376 +expl/path length Mean 500 +expl/path length Std 188 +expl/path length Max 688 +expl/path length Min 312 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0159877 +expl/Actions Std 0.856015 +expl/Actions Max 2.30448 +expl/Actions Min -2.36663 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 886591 +eval/num paths total 994 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.156579 +eval/Actions Std 0.718176 +eval/Actions Max 0.999491 +eval/Actions Min -0.999912 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.90805e-06 +time/evaluation sampling (s) 3.10536 +time/exploration sampling (s) 4.80046 +time/logging (s) 0.00642498 +time/saving (s) 0.00911526 +time/training (s) 14.4506 +time/epoch (s) 22.372 +time/total (s) 25347.6 +Epoch -13 +------------------------------ ---------------- +2022-05-16 01:05:22.667229 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -12 finished +------------------------------ ---------------- +epoch -12 +replay_buffer/size 999047 +trainer/num train calls 989000 +trainer/QF1 Loss 0.642826 +trainer/QF2 Loss 0.575781 +trainer/Policy Loss 29.4797 +trainer/Q1 Predictions Mean -75.2612 +trainer/Q1 Predictions Std 16.2092 +trainer/Q1 Predictions Max -8.43746 +trainer/Q1 Predictions Min -87.8879 +trainer/Q2 Predictions Mean -75.2413 +trainer/Q2 Predictions Std 16.2734 +trainer/Q2 Predictions Max -8.22391 +trainer/Q2 Predictions Min -87.8253 +trainer/Q Targets Mean -75.1735 +trainer/Q Targets Std 16.4165 +trainer/Q Targets Max -9.25154 +trainer/Q Targets Min -87.8252 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0175626 +trainer/policy/mean Std 0.735569 +trainer/policy/mean Max 0.999734 +trainer/policy/mean Min -0.999767 +trainer/policy/std Mean 0.406537 +trainer/policy/std Std 0.0185359 +trainer/policy/std Max 0.429924 +trainer/policy/std Min 0.378594 +trainer/Advantage Weights Mean 7.01544 +trainer/Advantage Weights Std 22.5697 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.28577e-24 +trainer/Advantage Score Mean -0.312549 +trainer/Advantage Score Std 0.707031 +trainer/Advantage Score Max 1.26096 +trainer/Advantage Score Min -5.32761 +trainer/V1 Predictions Mean -74.9355 +trainer/V1 Predictions Std 16.549 +trainer/V1 Predictions Max -8.65677 +trainer/V1 Predictions Min -87.6336 +trainer/VF Loss 0.0925052 +expl/num steps total 989000 +expl/num paths total 1378 +expl/path length Mean 500 +expl/path length Std 118 +expl/path length Max 618 +expl/path length Min 382 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0300286 +expl/Actions Std 0.829336 +expl/Actions Max 2.32221 +expl/Actions Min -2.39645 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 887569 +eval/num paths total 996 +eval/path length Mean 489 +eval/path length Std 43 +eval/path length Max 532 +eval/path length Min 446 +eval/Rewards Mean 0.00204499 +eval/Rewards Std 0.0451753 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0271128 +eval/Actions Std 0.748922 +eval/Actions Max 0.999975 +eval/Actions Min -0.999655 +eval/Num Paths 2 +eval/Average Returns 1 +time/data storing (s) 2.51876e-06 +time/evaluation sampling (s) 3.75075 +time/exploration sampling (s) 3.9981 +time/logging (s) 0.00664554 +time/saving (s) 0.00939259 +time/training (s) 14.9846 +time/epoch (s) 22.7495 +time/total (s) 25370.4 +Epoch -12 +------------------------------ ---------------- +2022-05-16 01:05:45.343546 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -11 finished +------------------------------ ---------------- +epoch -11 +replay_buffer/size 999047 +trainer/num train calls 990000 +trainer/QF1 Loss 0.503397 +trainer/QF2 Loss 0.588896 +trainer/Policy Loss 37.3559 +trainer/Q1 Predictions Mean -76.629 +trainer/Q1 Predictions Std 15.5874 +trainer/Q1 Predictions Max -2.12099 +trainer/Q1 Predictions Min -87.6525 +trainer/Q2 Predictions Mean -76.6108 +trainer/Q2 Predictions Std 15.6228 +trainer/Q2 Predictions Max -1.12843 +trainer/Q2 Predictions Min -87.8118 +trainer/Q Targets Mean -76.6508 +trainer/Q Targets Std 15.4318 +trainer/Q Targets Max -2.43526 +trainer/Q Targets Min -87.4266 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.00437085 +trainer/policy/mean Std 0.72342 +trainer/policy/mean Max 0.999701 +trainer/policy/mean Min -0.999588 +trainer/policy/std Mean 0.407999 +trainer/policy/std Std 0.018055 +trainer/policy/std Max 0.430188 +trainer/policy/std Min 0.380256 +trainer/Advantage Weights Mean 7.35345 +trainer/Advantage Weights Std 22.2156 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.20686e-15 +trainer/Advantage Score Mean -0.245443 +trainer/Advantage Score Std 0.612088 +trainer/Advantage Score Max 2.45097 +trainer/Advantage Score Min -3.33735 +trainer/V1 Predictions Mean -76.4177 +trainer/V1 Predictions Std 15.5916 +trainer/V1 Predictions Max -0.429566 +trainer/V1 Predictions Min -87.2938 +trainer/VF Loss 0.106512 +expl/num steps total 990000 +expl/num paths total 1380 +expl/path length Mean 500 +expl/path length Std 348 +expl/path length Max 848 +expl/path length Min 152 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0238659 +expl/Actions Std 0.816379 +expl/Actions Max 2.34244 +expl/Actions Min -2.22157 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 888504 +eval/num paths total 997 +eval/path length Mean 935 +eval/path length Std 0 +eval/path length Max 935 +eval/path length Min 935 +eval/Rewards Mean 0.00106952 +eval/Rewards Std 0.032686 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0342479 +eval/Actions Std 0.720838 +eval/Actions Max 0.999839 +eval/Actions Min -0.999897 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 8.51089e-06 +time/evaluation sampling (s) 3.63628 +time/exploration sampling (s) 4.22491 +time/logging (s) 0.00708603 +time/saving (s) 0.0100779 +time/training (s) 14.7852 +time/epoch (s) 22.6636 +time/total (s) 25393 +Epoch -11 +------------------------------ ---------------- +2022-05-16 01:06:07.857129 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -10 finished +------------------------------ ---------------- +epoch -10 +replay_buffer/size 999047 +trainer/num train calls 991000 +trainer/QF1 Loss 0.723616 +trainer/QF2 Loss 0.749507 +trainer/Policy Loss 26.9509 +trainer/Q1 Predictions Mean -74.4334 +trainer/Q1 Predictions Std 18.4368 +trainer/Q1 Predictions Max -0.570928 +trainer/Q1 Predictions Min -88.0923 +trainer/Q2 Predictions Mean -74.4436 +trainer/Q2 Predictions Std 18.4526 +trainer/Q2 Predictions Max -1.49343 +trainer/Q2 Predictions Min -87.9681 +trainer/Q Targets Mean -74.5975 +trainer/Q Targets Std 18.2147 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.8034 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000254478 +trainer/policy/mean Std 0.727819 +trainer/policy/mean Max 0.999576 +trainer/policy/mean Min -0.999787 +trainer/policy/std Mean 0.406538 +trainer/policy/std Std 0.0172911 +trainer/policy/std Max 0.427134 +trainer/policy/std Min 0.37859 +trainer/Advantage Weights Mean 6.83307 +trainer/Advantage Weights Std 23.05 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.19172e-14 +trainer/Advantage Score Mean -0.266111 +trainer/Advantage Score Std 0.568171 +trainer/Advantage Score Max 1.59573 +trainer/Advantage Score Min -3.20608 +trainer/V1 Predictions Mean -74.2248 +trainer/V1 Predictions Std 18.6332 +trainer/V1 Predictions Max -1.06726 +trainer/V1 Predictions Min -87.679 +trainer/VF Loss 0.0678476 +expl/num steps total 991000 +expl/num paths total 1381 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean 0.0941151 +expl/Actions Std 0.845692 +expl/Actions Max 2.12105 +expl/Actions Min -2.2625 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 889504 +eval/num paths total 998 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0542579 +eval/Actions Std 0.75282 +eval/Actions Max 0.999972 +eval/Actions Min -0.999427 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.4219e-06 +time/evaluation sampling (s) 3.40328 +time/exploration sampling (s) 4.88753 +time/logging (s) 0.00633526 +time/saving (s) 0.00894888 +time/training (s) 14.1933 +time/epoch (s) 22.4994 +time/total (s) 25415.5 +Epoch -10 +------------------------------ ---------------- +2022-05-16 01:06:30.389448 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -9 finished +------------------------------ ---------------- +epoch -9 +replay_buffer/size 999047 +trainer/num train calls 992000 +trainer/QF1 Loss 0.839141 +trainer/QF2 Loss 0.735697 +trainer/Policy Loss 16.4228 +trainer/Q1 Predictions Mean -75.4479 +trainer/Q1 Predictions Std 17.1875 +trainer/Q1 Predictions Max -1.10071 +trainer/Q1 Predictions Min -88.1424 +trainer/Q2 Predictions Mean -75.3704 +trainer/Q2 Predictions Std 17.1774 +trainer/Q2 Predictions Max -0.660926 +trainer/Q2 Predictions Min -87.9349 +trainer/Q Targets Mean -75.2276 +trainer/Q Targets Std 17.2588 +trainer/Q Targets Max -1.6238 +trainer/Q Targets Min -88.1507 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0183349 +trainer/policy/mean Std 0.738658 +trainer/policy/mean Max 0.999728 +trainer/policy/mean Min -0.999837 +trainer/policy/std Mean 0.405689 +trainer/policy/std Std 0.0177227 +trainer/policy/std Max 0.424727 +trainer/policy/std Min 0.37597 +trainer/Advantage Weights Mean 3.36725 +trainer/Advantage Weights Std 15.1527 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 4.31655e-12 +trainer/Advantage Score Mean -0.431231 +trainer/Advantage Score Std 0.531099 +trainer/Advantage Score Max 0.743961 +trainer/Advantage Score Min -2.61686 +trainer/V1 Predictions Mean -75.0225 +trainer/V1 Predictions Std 17.3698 +trainer/V1 Predictions Max 0.888904 +trainer/V1 Predictions Min -88.0227 +trainer/VF Loss 0.0554051 +expl/num steps total 992000 +expl/num paths total 1383 +expl/path length Mean 500 +expl/path length Std 132 +expl/path length Max 632 +expl/path length Min 368 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean -0.0271694 +expl/Actions Std 0.85859 +expl/Actions Max 2.46988 +expl/Actions Min -2.38114 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 890063 +eval/num paths total 999 +eval/path length Mean 559 +eval/path length Std 0 +eval/path length Max 559 +eval/path length Min 559 +eval/Rewards Mean 0.00178891 +eval/Rewards Std 0.0422576 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0200983 +eval/Actions Std 0.741049 +eval/Actions Max 0.999686 +eval/Actions Min -0.999604 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.64728e-06 +time/evaluation sampling (s) 3.35359 +time/exploration sampling (s) 5.0473 +time/logging (s) 0.00596239 +time/saving (s) 0.010222 +time/training (s) 14.1022 +time/epoch (s) 22.5193 +time/total (s) 25438 +Epoch -9 +------------------------------ ---------------- +2022-05-16 01:06:52.392288 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -8 finished +------------------------------ ---------------- +epoch -8 +replay_buffer/size 999047 +trainer/num train calls 993000 +trainer/QF1 Loss 1.15454 +trainer/QF2 Loss 0.951733 +trainer/Policy Loss 16.2303 +trainer/Q1 Predictions Mean -72.8023 +trainer/Q1 Predictions Std 18.7711 +trainer/Q1 Predictions Max -2.45662 +trainer/Q1 Predictions Min -87.5185 +trainer/Q2 Predictions Mean -72.8115 +trainer/Q2 Predictions Std 18.7291 +trainer/Q2 Predictions Max -2.58632 +trainer/Q2 Predictions Min -87.7916 +trainer/Q Targets Mean -72.785 +trainer/Q Targets Std 18.9702 +trainer/Q Targets Max 0.313375 +trainer/Q Targets Min -87.5742 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0362904 +trainer/policy/mean Std 0.737432 +trainer/policy/mean Max 0.999702 +trainer/policy/mean Min -0.999147 +trainer/policy/std Mean 0.407723 +trainer/policy/std Std 0.0185312 +trainer/policy/std Max 0.427079 +trainer/policy/std Min 0.380038 +trainer/Advantage Weights Mean 3.06952 +trainer/Advantage Weights Std 12.5226 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.11734e-20 +trainer/Advantage Score Mean -0.497879 +trainer/Advantage Score Std 0.788932 +trainer/Advantage Score Max 0.584709 +trainer/Advantage Score Min -4.44191 +trainer/V1 Predictions Mean -72.4667 +trainer/V1 Predictions Std 19.3029 +trainer/V1 Predictions Max 0.675925 +trainer/V1 Predictions Min -87.4999 +trainer/VF Loss 0.0954896 +expl/num steps total 993000 +expl/num paths total 1384 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0437581 +expl/Actions Std 0.851348 +expl/Actions Max 2.60517 +expl/Actions Min -2.17451 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 890714 +eval/num paths total 1000 +eval/path length Mean 651 +eval/path length Std 0 +eval/path length Max 651 +eval/path length Min 651 +eval/Rewards Mean 0.0015361 +eval/Rewards Std 0.039163 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0312559 +eval/Actions Std 0.742424 +eval/Actions Max 0.999898 +eval/Actions Min -0.999508 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.81958e-06 +time/evaluation sampling (s) 3.50095 +time/exploration sampling (s) 4.148 +time/logging (s) 0.00627275 +time/saving (s) 0.0119474 +time/training (s) 14.3252 +time/epoch (s) 21.9923 +time/total (s) 25460 +Epoch -8 +------------------------------ ---------------- +2022-05-16 01:07:15.518785 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -7 finished +------------------------------ ---------------- +epoch -7 +replay_buffer/size 999047 +trainer/num train calls 994000 +trainer/QF1 Loss 3.34169 +trainer/QF2 Loss 3.34296 +trainer/Policy Loss 37.35 +trainer/Q1 Predictions Mean -73.5982 +trainer/Q1 Predictions Std 19.4635 +trainer/Q1 Predictions Max -0.156766 +trainer/Q1 Predictions Min -87.258 +trainer/Q2 Predictions Mean -73.7009 +trainer/Q2 Predictions Std 19.3541 +trainer/Q2 Predictions Max 0.155639 +trainer/Q2 Predictions Min -87.3515 +trainer/Q Targets Mean -73.7976 +trainer/Q Targets Std 19.3544 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.6507 +trainer/rewards Mean -0.980469 +trainer/rewards Std 0.138383 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0195312 +trainer/terminals Std 0.138383 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0255817 +trainer/policy/mean Std 0.744533 +trainer/policy/mean Max 0.999791 +trainer/policy/mean Min -0.99945 +trainer/policy/std Mean 0.407426 +trainer/policy/std Std 0.0185717 +trainer/policy/std Max 0.426685 +trainer/policy/std Min 0.377808 +trainer/Advantage Weights Mean 6.88669 +trainer/Advantage Weights Std 22.4429 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.68342e-14 +trainer/Advantage Score Mean -0.284812 +trainer/Advantage Score Std 0.616262 +trainer/Advantage Score Max 2.14082 +trainer/Advantage Score Min -3.09323 +trainer/V1 Predictions Mean -73.6547 +trainer/V1 Predictions Std 19.3996 +trainer/V1 Predictions Max 0.00160795 +trainer/V1 Predictions Min -87.6556 +trainer/VF Loss 0.0988741 +expl/num steps total 994000 +expl/num paths total 1386 +expl/path length Mean 500 +expl/path length Std 419 +expl/path length Max 919 +expl/path length Min 81 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0227917 +expl/Actions Std 0.829744 +expl/Actions Max 2.07786 +expl/Actions Min -2.28643 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 891714 +eval/num paths total 1001 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean -0.439125 +eval/Actions Std 0.717964 +eval/Actions Max 0.99992 +eval/Actions Min -0.999743 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 8.63429e-06 +time/evaluation sampling (s) 4.04102 +time/exploration sampling (s) 3.97528 +time/logging (s) 0.00667289 +time/saving (s) 0.00927067 +time/training (s) 15.083 +time/epoch (s) 23.1152 +time/total (s) 25483.2 +Epoch -7 +------------------------------ ---------------- +2022-05-16 01:07:38.980942 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -6 finished +------------------------------ ---------------- +epoch -6 +replay_buffer/size 999047 +trainer/num train calls 995000 +trainer/QF1 Loss 0.801906 +trainer/QF2 Loss 0.873683 +trainer/Policy Loss 19.2294 +trainer/Q1 Predictions Mean -74.0554 +trainer/Q1 Predictions Std 19.4936 +trainer/Q1 Predictions Max -0.306028 +trainer/Q1 Predictions Min -87.7301 +trainer/Q2 Predictions Mean -74.047 +trainer/Q2 Predictions Std 19.3683 +trainer/Q2 Predictions Max -0.424517 +trainer/Q2 Predictions Min -87.4989 +trainer/Q Targets Mean -73.9219 +trainer/Q Targets Std 19.4672 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.3997 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.0283017 +trainer/policy/mean Std 0.74733 +trainer/policy/mean Max 0.999888 +trainer/policy/mean Min -0.999117 +trainer/policy/std Mean 0.40772 +trainer/policy/std Std 0.0184333 +trainer/policy/std Max 0.427849 +trainer/policy/std Min 0.374771 +trainer/Advantage Weights Mean 4.064 +trainer/Advantage Weights Std 17.6278 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.8485e-21 +trainer/Advantage Score Mean -0.396795 +trainer/Advantage Score Std 0.582774 +trainer/Advantage Score Max 1.5798 +trainer/Advantage Score Min -4.6294 +trainer/V1 Predictions Mean -73.68 +trainer/V1 Predictions Std 19.6443 +trainer/V1 Predictions Max -0.14518 +trainer/V1 Predictions Min -87.2699 +trainer/VF Loss 0.0745776 +expl/num steps total 995000 +expl/num paths total 1388 +expl/path length Mean 500 +expl/path length Std 426 +expl/path length Max 926 +expl/path length Min 74 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0251927 +expl/Actions Std 0.827306 +expl/Actions Max 2.43951 +expl/Actions Min -2.27928 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 892312 +eval/num paths total 1002 +eval/path length Mean 598 +eval/path length Std 0 +eval/path length Max 598 +eval/path length Min 598 +eval/Rewards Mean 0.00167224 +eval/Rewards Std 0.0408588 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0342012 +eval/Actions Std 0.761776 +eval/Actions Max 0.999822 +eval/Actions Min -0.99974 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.69711e-06 +time/evaluation sampling (s) 3.69137 +time/exploration sampling (s) 5.26068 +time/logging (s) 0.00587841 +time/saving (s) 0.0090014 +time/training (s) 14.4825 +time/epoch (s) 23.4494 +time/total (s) 25506.6 +Epoch -6 +------------------------------ ---------------- +2022-05-16 01:08:02.271698 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -5 finished +------------------------------ ---------------- +epoch -5 +replay_buffer/size 999047 +trainer/num train calls 996000 +trainer/QF1 Loss 1.06913 +trainer/QF2 Loss 1.10614 +trainer/Policy Loss 13.78 +trainer/Q1 Predictions Mean -74.9754 +trainer/Q1 Predictions Std 16.4888 +trainer/Q1 Predictions Max -0.342365 +trainer/Q1 Predictions Min -88.0885 +trainer/Q2 Predictions Mean -75.03 +trainer/Q2 Predictions Std 16.4309 +trainer/Q2 Predictions Max 0.38321 +trainer/Q2 Predictions Min -88.0747 +trainer/Q Targets Mean -74.9989 +trainer/Q Targets Std 16.6181 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.1035 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0178617 +trainer/policy/mean Std 0.725028 +trainer/policy/mean Max 0.999715 +trainer/policy/mean Min -0.99848 +trainer/policy/std Mean 0.405947 +trainer/policy/std Std 0.0185427 +trainer/policy/std Max 0.429274 +trainer/policy/std Min 0.377818 +trainer/Advantage Weights Mean 3.7948 +trainer/Advantage Weights Std 16.7474 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.80666e-25 +trainer/Advantage Score Mean -0.411526 +trainer/Advantage Score Std 0.755158 +trainer/Advantage Score Max 0.962915 +trainer/Advantage Score Min -5.62279 +trainer/V1 Predictions Mean -74.6996 +trainer/V1 Predictions Std 16.7156 +trainer/V1 Predictions Max -0.193144 +trainer/V1 Predictions Min -87.9542 +trainer/VF Loss 0.0852556 +expl/num steps total 996000 +expl/num paths total 1390 +expl/path length Mean 500 +expl/path length Std 215 +expl/path length Max 715 +expl/path length Min 285 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0184928 +expl/Actions Std 0.835902 +expl/Actions Max 2.2177 +expl/Actions Min -2.34526 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 893312 +eval/num paths total 1003 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.0514423 +eval/Actions Std 0.707065 +eval/Actions Max 0.999871 +eval/Actions Min -0.999824 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.63471e-06 +time/evaluation sampling (s) 3.50177 +time/exploration sampling (s) 5.22658 +time/logging (s) 0.0074154 +time/saving (s) 0.0119854 +time/training (s) 14.5337 +time/epoch (s) 23.2814 +time/total (s) 25529.9 +Epoch -5 +------------------------------ ---------------- +2022-05-16 01:08:25.023349 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -4 finished +------------------------------ ---------------- +epoch -4 +replay_buffer/size 999047 +trainer/num train calls 997000 +trainer/QF1 Loss 0.576221 +trainer/QF2 Loss 0.522699 +trainer/Policy Loss 23.3642 +trainer/Q1 Predictions Mean -74.9043 +trainer/Q1 Predictions Std 16.182 +trainer/Q1 Predictions Max -2.1302 +trainer/Q1 Predictions Min -87.4093 +trainer/Q2 Predictions Mean -74.9408 +trainer/Q2 Predictions Std 16.1616 +trainer/Q2 Predictions Max -2.09307 +trainer/Q2 Predictions Min -87.53 +trainer/Q Targets Mean -75.1354 +trainer/Q Targets Std 16.1593 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.5512 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.0133936 +trainer/policy/mean Std 0.730527 +trainer/policy/mean Max 0.999306 +trainer/policy/mean Min -0.998746 +trainer/policy/std Mean 0.405883 +trainer/policy/std Std 0.020117 +trainer/policy/std Max 0.429432 +trainer/policy/std Min 0.372549 +trainer/Advantage Weights Mean 4.83944 +trainer/Advantage Weights Std 17.1502 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 3.05592e-12 +trainer/Advantage Score Mean -0.246944 +trainer/Advantage Score Std 0.525904 +trainer/Advantage Score Max 1.65254 +trainer/Advantage Score Min -2.65139 +trainer/V1 Predictions Mean -74.9107 +trainer/V1 Predictions Std 16.2255 +trainer/V1 Predictions Max -2.34146 +trainer/V1 Predictions Min -87.4268 +trainer/VF Loss 0.0546508 +expl/num steps total 997000 +expl/num paths total 1391 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.0385997 +expl/Actions Std 0.852396 +expl/Actions Max 2.26946 +expl/Actions Min -2.48938 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 894312 +eval/num paths total 1004 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.349011 +eval/Actions Std 0.723782 +eval/Actions Max 0.998153 +eval/Actions Min -0.99784 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 2.61934e-06 +time/evaluation sampling (s) 3.22351 +time/exploration sampling (s) 4.94539 +time/logging (s) 0.00805841 +time/saving (s) 0.0169409 +time/training (s) 14.5433 +time/epoch (s) 22.7372 +time/total (s) 25552.6 +Epoch -4 +------------------------------ ---------------- +2022-05-16 01:08:48.037468 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -3 finished +------------------------------ ---------------- +epoch -3 +replay_buffer/size 999047 +trainer/num train calls 998000 +trainer/QF1 Loss 0.791971 +trainer/QF2 Loss 0.865396 +trainer/Policy Loss 18.7461 +trainer/Q1 Predictions Mean -74.0249 +trainer/Q1 Predictions Std 19.8157 +trainer/Q1 Predictions Max -0.907003 +trainer/Q1 Predictions Min -88.3185 +trainer/Q2 Predictions Mean -73.8825 +trainer/Q2 Predictions Std 19.9694 +trainer/Q2 Predictions Max -1.28001 +trainer/Q2 Predictions Min -88.1567 +trainer/Q Targets Mean -73.8668 +trainer/Q Targets Std 19.5685 +trainer/Q Targets Max -2.53503 +trainer/Q Targets Min -87.9076 +trainer/rewards Mean -1 +trainer/rewards Std 0 +trainer/rewards Max -1 +trainer/rewards Min -1 +trainer/terminals Mean 0 +trainer/terminals Std 0 +trainer/terminals Max 0 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.000853657 +trainer/policy/mean Std 0.727032 +trainer/policy/mean Max 0.999937 +trainer/policy/mean Min -0.999988 +trainer/policy/std Mean 0.407 +trainer/policy/std Std 0.0198934 +trainer/policy/std Max 0.4291 +trainer/policy/std Min 0.378454 +trainer/Advantage Weights Mean 3.07852 +trainer/Advantage Weights Std 14.5015 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 1.88782e-17 +trainer/Advantage Score Mean -0.515229 +trainer/Advantage Score Std 0.674824 +trainer/Advantage Score Max 0.885471 +trainer/Advantage Score Min -3.85085 +trainer/V1 Predictions Mean -73.565 +trainer/V1 Predictions Std 19.9382 +trainer/V1 Predictions Max -0.312843 +trainer/V1 Predictions Min -87.7811 +trainer/VF Loss 0.0812043 +expl/num steps total 998000 +expl/num paths total 1393 +expl/path length Mean 500 +expl/path length Std 164 +expl/path length Max 664 +expl/path length Min 336 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0215674 +expl/Actions Std 0.818143 +expl/Actions Max 2.34685 +expl/Actions Min -2.27377 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 894864 +eval/num paths total 1005 +eval/path length Mean 552 +eval/path length Std 0 +eval/path length Max 552 +eval/path length Min 552 +eval/Rewards Mean 0.00181159 +eval/Rewards Std 0.0425243 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.0356896 +eval/Actions Std 0.737398 +eval/Actions Max 0.99966 +eval/Actions Min -0.999404 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.80701e-06 +time/evaluation sampling (s) 3.07414 +time/exploration sampling (s) 4.81914 +time/logging (s) 0.00845252 +time/saving (s) 0.0132646 +time/training (s) 15.0842 +time/epoch (s) 22.9992 +time/total (s) 25575.6 +Epoch -3 +------------------------------ ---------------- +2022-05-16 01:09:10.830931 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -2 finished +------------------------------ ---------------- +epoch -2 +replay_buffer/size 999047 +trainer/num train calls 999000 +trainer/QF1 Loss 0.678634 +trainer/QF2 Loss 0.624753 +trainer/Policy Loss 23.1502 +trainer/Q1 Predictions Mean -75.7188 +trainer/Q1 Predictions Std 17.0643 +trainer/Q1 Predictions Max -0.811782 +trainer/Q1 Predictions Min -88.4484 +trainer/Q2 Predictions Mean -75.7029 +trainer/Q2 Predictions Std 17.0366 +trainer/Q2 Predictions Max -1.11902 +trainer/Q2 Predictions Min -88.2297 +trainer/Q Targets Mean -75.7751 +trainer/Q Targets Std 17.0334 +trainer/Q Targets Max 0 +trainer/Q Targets Min -88.3758 +trainer/rewards Mean -0.996094 +trainer/rewards Std 0.0623778 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.00390625 +trainer/terminals Std 0.0623778 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean 0.00626569 +trainer/policy/mean Std 0.726659 +trainer/policy/mean Max 0.998328 +trainer/policy/mean Min -0.999313 +trainer/policy/std Mean 0.405148 +trainer/policy/std Std 0.019479 +trainer/policy/std Max 0.427338 +trainer/policy/std Min 0.375731 +trainer/Advantage Weights Mean 5.27514 +trainer/Advantage Weights Std 19.2901 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 7.83523e-11 +trainer/Advantage Score Mean -0.248799 +trainer/Advantage Score Std 0.432744 +trainer/Advantage Score Max 1.20284 +trainer/Advantage Score Min -2.32698 +trainer/V1 Predictions Mean -75.5726 +trainer/V1 Predictions Std 17.0346 +trainer/V1 Predictions Max 0.754585 +trainer/V1 Predictions Min -88.2372 +trainer/VF Loss 0.0458559 +expl/num steps total 999000 +expl/num paths total 1395 +expl/path length Mean 500 +expl/path length Std 192 +expl/path length Max 692 +expl/path length Min 308 +expl/Rewards Mean 0.001 +expl/Rewards Std 0.031607 +expl/Rewards Max 1 +expl/Rewards Min 0 +expl/Returns Mean 0.5 +expl/Returns Std 0.5 +expl/Returns Max 1 +expl/Returns Min 0 +expl/Actions Mean 0.0304918 +expl/Actions Std 0.832034 +expl/Actions Max 2.177 +expl/Actions Min -2.20643 +expl/Num Paths 2 +expl/Average Returns 0.5 +eval/num steps total 895864 +eval/num paths total 1006 +eval/path length Mean 1000 +eval/path length Std 0 +eval/path length Max 1000 +eval/path length Min 1000 +eval/Rewards Mean 0 +eval/Rewards Std 0 +eval/Rewards Max 0 +eval/Rewards Min 0 +eval/Returns Mean 0 +eval/Returns Std 0 +eval/Returns Max 0 +eval/Returns Min 0 +eval/Actions Mean 0.258356 +eval/Actions Std 0.52873 +eval/Actions Max 0.999803 +eval/Actions Min -0.999886 +eval/Num Paths 1 +eval/Average Returns 0 +time/data storing (s) 9.53209e-06 +time/evaluation sampling (s) 3.16076 +time/exploration sampling (s) 4.70958 +time/logging (s) 0.00651298 +time/saving (s) 0.00977038 +time/training (s) 14.8897 +time/epoch (s) 22.7763 +time/total (s) 25598.4 +Epoch -2 +------------------------------ ---------------- +2022-05-16 01:09:32.819754 PDT | [05-15-iql-normalized-antmaze-large-diverse-v0_2022_05_15_18_02_29_0000--s-0] Epoch -1 finished +------------------------------ ---------------- +epoch -1 +replay_buffer/size 999047 +trainer/num train calls 1e+06 +trainer/QF1 Loss 0.547929 +trainer/QF2 Loss 0.509117 +trainer/Policy Loss 23.0539 +trainer/Q1 Predictions Mean -74.9732 +trainer/Q1 Predictions Std 17.2498 +trainer/Q1 Predictions Max -1.60261 +trainer/Q1 Predictions Min -88.025 +trainer/Q2 Predictions Mean -74.9458 +trainer/Q2 Predictions Std 17.2044 +trainer/Q2 Predictions Max -2.21049 +trainer/Q2 Predictions Min -87.9659 +trainer/Q Targets Mean -75.1234 +trainer/Q Targets Std 17.1892 +trainer/Q Targets Max 0 +trainer/Q Targets Min -87.9642 +trainer/rewards Mean -0.992188 +trainer/rewards Std 0.0880424 +trainer/rewards Max 0 +trainer/rewards Min -1 +trainer/terminals Mean 0.0078125 +trainer/terminals Std 0.0880424 +trainer/terminals Max 1 +trainer/terminals Min 0 +trainer/replay_buffer_len 999047 +trainer/policy/mean Mean -0.019199 +trainer/policy/mean Std 0.721023 +trainer/policy/mean Max 0.999982 +trainer/policy/mean Min -0.999858 +trainer/policy/std Mean 0.405561 +trainer/policy/std Std 0.0180295 +trainer/policy/std Max 0.425023 +trainer/policy/std Min 0.378818 +trainer/Advantage Weights Mean 6.44321 +trainer/Advantage Weights Std 20.8195 +trainer/Advantage Weights Max 100 +trainer/Advantage Weights Min 5.56008e-20 +trainer/Advantage Score Mean -0.320989 +trainer/Advantage Score Std 0.71055 +trainer/Advantage Score Max 3.23609 +trainer/Advantage Score Min -4.43361 +trainer/V1 Predictions Mean -74.8177 +trainer/V1 Predictions Std 17.4273 +trainer/V1 Predictions Max -1.00982 +trainer/V1 Predictions Min -87.9727 +trainer/VF Loss 0.115571 +expl/num steps total 1e+06 +expl/num paths total 1396 +expl/path length Mean 1000 +expl/path length Std 0 +expl/path length Max 1000 +expl/path length Min 1000 +expl/Rewards Mean 0 +expl/Rewards Std 0 +expl/Rewards Max 0 +expl/Rewards Min 0 +expl/Returns Mean 0 +expl/Returns Std 0 +expl/Returns Max 0 +expl/Returns Min 0 +expl/Actions Mean -0.480857 +expl/Actions Std 0.794139 +expl/Actions Max 2.69121 +expl/Actions Min -2.51516 +expl/Num Paths 1 +expl/Average Returns 0 +eval/num steps total 896470 +eval/num paths total 1007 +eval/path length Mean 606 +eval/path length Std 0 +eval/path length Max 606 +eval/path length Min 606 +eval/Rewards Mean 0.00165017 +eval/Rewards Std 0.0405887 +eval/Rewards Max 1 +eval/Rewards Min 0 +eval/Returns Mean 1 +eval/Returns Std 0 +eval/Returns Max 1 +eval/Returns Min 1 +eval/Actions Mean 0.00869274 +eval/Actions Std 0.73599 +eval/Actions Max 0.999682 +eval/Actions Min -0.999966 +eval/Num Paths 1 +eval/Average Returns 1 +time/data storing (s) 2.85823e-06 +time/evaluation sampling (s) 3.4206 +time/exploration sampling (s) 4.13799 +time/logging (s) 0.00577966 +time/saving (s) 0.0089013 +time/training (s) 14.4023 +time/epoch (s) 21.9755 +time/total (s) 25620.4 +Epoch -1 +------------------------------ ----------------