iamshlomo commited on
Commit
0e027fa
·
verified ·
1 Parent(s): a7ed788

Upload generalization/mixing_20260324_1753/mixing_results.json with huggingface_hub

Browse files
generalization/mixing_20260324_1753/mixing_results.json ADDED
@@ -0,0 +1,581 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "run_id": "20260324_1753",
3
+ "max_iter": 500,
4
+ "eval_eps": 10,
5
+ "id_envs": [
6
+ "MiniHack-Room-Random-5x5-v0",
7
+ "MiniHack-Room-Random-15x15-v0",
8
+ "MiniHack-Corridor-R2-v0",
9
+ "MiniHack-MazeWalk-9x9-v0"
10
+ ],
11
+ "ood_envs": [
12
+ "MiniHack-Room-Dark-15x15-v0",
13
+ "MiniHack-Corridor-R5-v0",
14
+ "MiniHack-MazeWalk-45x19-v0"
15
+ ],
16
+ "oracle_fractions": [
17
+ 1.0,
18
+ 0.9,
19
+ 0.7,
20
+ 0.5,
21
+ 0.0
22
+ ],
23
+ "id_win_rates": [
24
+ 0.45,
25
+ 0.425,
26
+ 0.525,
27
+ 0.45,
28
+ 0.025
29
+ ],
30
+ "is_monotonic": false,
31
+ "baseline_id": 0.45,
32
+ "pure_rl_id": 0.025,
33
+ "mixing_details": {
34
+ "0.9": {
35
+ "final_id": 0.425,
36
+ "final_ood": 0.13333333333333333,
37
+ "history": {
38
+ "iter": [
39
+ 10.0,
40
+ 20.0,
41
+ 30.0,
42
+ 40.0,
43
+ 50.0,
44
+ 60.0,
45
+ 70.0,
46
+ 80.0,
47
+ 90.0,
48
+ 100.0,
49
+ 110.0,
50
+ 120.0,
51
+ 130.0,
52
+ 140.0,
53
+ 150.0,
54
+ 160.0,
55
+ 170.0,
56
+ 180.0,
57
+ 190.0,
58
+ 200.0,
59
+ 210.0,
60
+ 220.0,
61
+ 230.0,
62
+ 240.0,
63
+ 250.0,
64
+ 260.0,
65
+ 270.0,
66
+ 280.0,
67
+ 290.0,
68
+ 300.0,
69
+ 310.0,
70
+ 320.0,
71
+ 330.0,
72
+ 340.0,
73
+ 350.0,
74
+ 360.0,
75
+ 370.0,
76
+ 380.0,
77
+ 390.0,
78
+ 400.0,
79
+ 410.0,
80
+ 420.0,
81
+ 430.0,
82
+ 440.0,
83
+ 450.0,
84
+ 460.0,
85
+ 470.0,
86
+ 480.0,
87
+ 490.0,
88
+ 500.0
89
+ ],
90
+ "loss": [
91
+ 1.6752936661243438,
92
+ 0.5456457614898682,
93
+ 0.6338867455720901,
94
+ 0.5054281622171402,
95
+ 0.527776637673378,
96
+ 0.5110885083675385,
97
+ 0.579666268825531,
98
+ 0.5107972383499145,
99
+ 0.5372573375701905,
100
+ 0.4475385591387749,
101
+ 0.5456025630235672,
102
+ 0.4427765429019928,
103
+ 0.48316262364387513,
104
+ 0.48554708659648893,
105
+ 0.42708337903022764,
106
+ 0.4209475159645081,
107
+ 0.5221806704998017,
108
+ 0.3704051971435547,
109
+ 0.4507525831460953,
110
+ 0.47720264494419096,
111
+ 0.44097875952720644,
112
+ 0.47508017122745516,
113
+ 0.4531687557697296,
114
+ 0.44836609065532684,
115
+ 0.44673745334148407,
116
+ 0.3964699223637581,
117
+ 0.46190544664859773,
118
+ 0.42247503697872163,
119
+ 0.4005137324333191,
120
+ 0.4329902589321136,
121
+ 0.3522083267569542,
122
+ 0.45307801216840743,
123
+ 0.4122482717037201,
124
+ 0.43058977723121644,
125
+ 0.3758606493473053,
126
+ 0.43025832772254946,
127
+ 0.3754075258970261,
128
+ 0.4402051389217377,
129
+ 0.5073165476322175,
130
+ 0.4376820087432861,
131
+ 0.3616011530160904,
132
+ 0.4073281824588776,
133
+ 0.4005865752696991,
134
+ 0.45797563791275026,
135
+ 0.5316958755254746,
136
+ 0.4308278739452362,
137
+ 0.44909127950668337,
138
+ 0.43340407609939574,
139
+ 0.4299317389726639,
140
+ 0.5180808395147324
141
+ ],
142
+ "win_rate_episode": [
143
+ 0.7,
144
+ 0.6,
145
+ 0.3,
146
+ 0.7,
147
+ 0.4,
148
+ 0.4,
149
+ 0.6,
150
+ 0.6,
151
+ 0.6,
152
+ 0.2,
153
+ 0.5,
154
+ 0.5,
155
+ 0.5,
156
+ 0.6,
157
+ 0.8,
158
+ 0.3,
159
+ 0.7,
160
+ 0.5,
161
+ 0.5,
162
+ 0.5,
163
+ 0.7,
164
+ 0.4,
165
+ 0.5,
166
+ 0.8,
167
+ 0.7,
168
+ 0.7,
169
+ 0.5,
170
+ 0.1,
171
+ 0.8,
172
+ 0.5,
173
+ 0.4,
174
+ 0.4,
175
+ 0.6,
176
+ 0.6,
177
+ 0.6,
178
+ 0.4,
179
+ 0.8,
180
+ 0.6,
181
+ 0.8,
182
+ 0.7,
183
+ 0.4,
184
+ 0.6,
185
+ 0.5,
186
+ 0.5,
187
+ 0.3,
188
+ 0.7,
189
+ 0.7,
190
+ 0.6,
191
+ 0.5,
192
+ 0.1
193
+ ],
194
+ "id_winrate": [
195
+ 0.5249999999999999,
196
+ 0.425,
197
+ 0.4,
198
+ 0.6000000000000001,
199
+ 0.425
200
+ ],
201
+ "id_winrate_iter": [
202
+ 100.0,
203
+ 200.0,
204
+ 300.0,
205
+ 400.0,
206
+ 500.0
207
+ ],
208
+ "ood_winrate": [
209
+ 0.10000000000000002
210
+ ],
211
+ "ood_winrate_iter": [
212
+ 500.0
213
+ ]
214
+ }
215
+ },
216
+ "0.7": {
217
+ "final_id": 0.525,
218
+ "final_ood": 0.03333333333333333,
219
+ "history": {
220
+ "iter": [
221
+ 10.0,
222
+ 20.0,
223
+ 30.0,
224
+ 40.0,
225
+ 50.0,
226
+ 60.0,
227
+ 70.0,
228
+ 80.0,
229
+ 90.0,
230
+ 100.0,
231
+ 110.0,
232
+ 120.0,
233
+ 130.0,
234
+ 140.0,
235
+ 150.0,
236
+ 160.0,
237
+ 170.0,
238
+ 180.0,
239
+ 190.0,
240
+ 200.0,
241
+ 210.0,
242
+ 220.0,
243
+ 230.0,
244
+ 240.0,
245
+ 250.0,
246
+ 260.0,
247
+ 270.0,
248
+ 280.0,
249
+ 290.0,
250
+ 300.0,
251
+ 310.0,
252
+ 320.0,
253
+ 330.0,
254
+ 340.0,
255
+ 350.0,
256
+ 360.0,
257
+ 370.0,
258
+ 380.0,
259
+ 390.0,
260
+ 400.0,
261
+ 410.0,
262
+ 420.0,
263
+ 430.0,
264
+ 440.0,
265
+ 450.0,
266
+ 460.0,
267
+ 470.0,
268
+ 480.0,
269
+ 490.0,
270
+ 500.0
271
+ ],
272
+ "loss": [
273
+ 1.6250346541404723,
274
+ 0.7509583801031112,
275
+ 0.6045627593994141,
276
+ 0.5654109179973602,
277
+ 0.6625134199857712,
278
+ 0.4880312532186508,
279
+ 0.5012406677007675,
280
+ 0.6009388148784638,
281
+ 0.5926730215549469,
282
+ 0.6496199339628219,
283
+ 0.5335190117359161,
284
+ 0.4914944231510162,
285
+ 0.5975763022899627,
286
+ 0.5706877484917641,
287
+ 0.5325360238552094,
288
+ 0.5357880219817162,
289
+ 0.5610981732606888,
290
+ 0.5447167068719864,
291
+ 0.5535276144742965,
292
+ 0.4636155366897583,
293
+ 0.4978939801454544,
294
+ 0.4948939472436905,
295
+ 0.5450316801667213,
296
+ 0.5371656566858292,
297
+ 0.5236778199672699,
298
+ 0.5228035360574722,
299
+ 0.5021849542856216,
300
+ 0.47030882984399797,
301
+ 0.5468545973300933,
302
+ 0.41855976581573484,
303
+ 0.4042443186044693,
304
+ 0.45167635679244994,
305
+ 0.5066370971500873,
306
+ 0.45928401350975034,
307
+ 0.4441061705350876,
308
+ 0.4326098680496216,
309
+ 0.48207026720046997,
310
+ 0.5362424671649932,
311
+ 0.3166206330060959,
312
+ 0.41898477375507354,
313
+ 0.44150267243385316,
314
+ 0.46051640808582306,
315
+ 0.4345561623573303,
316
+ 0.4201108366250992,
317
+ 0.547324874997139,
318
+ 0.419434255361557,
319
+ 0.4851557433605194,
320
+ 0.46918579936027527,
321
+ 0.4255599558353424,
322
+ 0.484037908911705
323
+ ],
324
+ "win_rate_episode": [
325
+ 0.6,
326
+ 0.6,
327
+ 0.7,
328
+ 0.4,
329
+ 0.7,
330
+ 0.5,
331
+ 0.5,
332
+ 0.7,
333
+ 0.3,
334
+ 0.3,
335
+ 0.6,
336
+ 0.1,
337
+ 0.8,
338
+ 0.4,
339
+ 0.5,
340
+ 0.5,
341
+ 0.3,
342
+ 0.1,
343
+ 0.7,
344
+ 0.5,
345
+ 0.6,
346
+ 0.4,
347
+ 0.6,
348
+ 0.3,
349
+ 0.2,
350
+ 0.3,
351
+ 0.3,
352
+ 0.3,
353
+ 0.2,
354
+ 0.2,
355
+ 0.5,
356
+ 0.8,
357
+ 0.3,
358
+ 0.2,
359
+ 0.7,
360
+ 0.0,
361
+ 0.5,
362
+ 0.4,
363
+ 0.4,
364
+ 0.4,
365
+ 0.5,
366
+ 0.1,
367
+ 0.3,
368
+ 0.1,
369
+ 0.3,
370
+ 0.1,
371
+ 0.3,
372
+ 0.2,
373
+ 0.3,
374
+ 0.4
375
+ ],
376
+ "id_winrate": [
377
+ 0.47500000000000003,
378
+ 0.6,
379
+ 0.39999999999999997,
380
+ 0.39999999999999997,
381
+ 0.3
382
+ ],
383
+ "id_winrate_iter": [
384
+ 100.0,
385
+ 200.0,
386
+ 300.0,
387
+ 400.0,
388
+ 500.0
389
+ ],
390
+ "ood_winrate": [
391
+ 0.03333333333333333
392
+ ],
393
+ "ood_winrate_iter": [
394
+ 500.0
395
+ ]
396
+ }
397
+ },
398
+ "0.5": {
399
+ "final_id": 0.45,
400
+ "final_ood": 0.03333333333333333,
401
+ "history": {
402
+ "iter": [
403
+ 10.0,
404
+ 20.0,
405
+ 30.0,
406
+ 40.0,
407
+ 50.0,
408
+ 60.0,
409
+ 70.0,
410
+ 80.0,
411
+ 90.0,
412
+ 100.0,
413
+ 110.0,
414
+ 120.0,
415
+ 130.0,
416
+ 140.0,
417
+ 150.0,
418
+ 160.0,
419
+ 170.0,
420
+ 180.0,
421
+ 190.0,
422
+ 200.0,
423
+ 210.0,
424
+ 220.0,
425
+ 230.0,
426
+ 240.0,
427
+ 250.0,
428
+ 260.0,
429
+ 270.0,
430
+ 280.0,
431
+ 290.0,
432
+ 300.0,
433
+ 310.0,
434
+ 320.0,
435
+ 330.0,
436
+ 340.0,
437
+ 350.0,
438
+ 360.0,
439
+ 370.0,
440
+ 380.0,
441
+ 390.0,
442
+ 400.0,
443
+ 410.0,
444
+ 420.0,
445
+ 430.0,
446
+ 440.0,
447
+ 450.0,
448
+ 460.0,
449
+ 470.0,
450
+ 480.0,
451
+ 490.0,
452
+ 500.0
453
+ ],
454
+ "loss": [
455
+ 1.8360243797302247,
456
+ 0.8808628171682358,
457
+ 0.8936832666397094,
458
+ 0.8906782031059265,
459
+ 0.6086486503481865,
460
+ 0.5666145443916321,
461
+ 0.7302786946296692,
462
+ 0.7476702690124511,
463
+ 0.5891640901565551,
464
+ 0.5791785329580307,
465
+ 0.6193409830331802,
466
+ 0.5375891029834747,
467
+ 0.6533806771039963,
468
+ 0.5705962270498276,
469
+ 0.4899124324321747,
470
+ 0.484193828701973,
471
+ 0.7167756080627441,
472
+ 0.7172859489917756,
473
+ 0.6771132230758667,
474
+ 0.6476515769958496,
475
+ 0.5663518220186233,
476
+ 0.5441020250320434,
477
+ 0.5578588485717774,
478
+ 0.49196565300226214,
479
+ 0.6030937105417251,
480
+ 0.4176992058753967,
481
+ 0.4836134403944016,
482
+ 0.35279067903757094,
483
+ 0.575089767575264,
484
+ 0.48320265412330626,
485
+ 0.4918143391609192,
486
+ 0.5697446644306183,
487
+ 0.5520341217517852,
488
+ 0.45106160938739776,
489
+ 0.37911405861377717,
490
+ 0.4294638156890869,
491
+ 0.4478351235389709,
492
+ 0.4790546208620071,
493
+ 0.5499710202217102,
494
+ 0.4521573543548584,
495
+ 0.42517648339271547,
496
+ 0.4707114115357399,
497
+ 0.4371482700109482,
498
+ 0.5671361476182938,
499
+ 0.4924820393323898,
500
+ 0.37695282995700835,
501
+ 0.5502919137477875,
502
+ 0.36635997891426086,
503
+ 0.4519714146852493,
504
+ 0.5469056159257889
505
+ ],
506
+ "win_rate_episode": [
507
+ 0.5,
508
+ 0.2,
509
+ 0.5,
510
+ 0.4,
511
+ 0.5,
512
+ 0.5,
513
+ 0.5,
514
+ 0.3,
515
+ 0.3,
516
+ 0.1,
517
+ 0.4,
518
+ 0.6,
519
+ 0.2,
520
+ 0.5,
521
+ 0.5,
522
+ 0.1,
523
+ 0.7,
524
+ 0.3,
525
+ 0.2,
526
+ 0.4,
527
+ 0.0,
528
+ 0.5,
529
+ 0.4,
530
+ 0.5,
531
+ 0.5,
532
+ 0.2,
533
+ 0.3,
534
+ 0.3,
535
+ 0.4,
536
+ 0.3,
537
+ 0.3,
538
+ 0.2,
539
+ 0.3,
540
+ 0.4,
541
+ 0.1,
542
+ 0.5,
543
+ 0.3,
544
+ 0.4,
545
+ 0.4,
546
+ 0.2,
547
+ 0.3,
548
+ 0.6,
549
+ 0.3,
550
+ 0.5,
551
+ 0.3,
552
+ 0.3,
553
+ 0.7,
554
+ 0.1,
555
+ 0.3,
556
+ 0.3
557
+ ],
558
+ "id_winrate": [
559
+ 0.475,
560
+ 0.425,
561
+ 0.25,
562
+ 0.25,
563
+ 0.35
564
+ ],
565
+ "id_winrate_iter": [
566
+ 100.0,
567
+ 200.0,
568
+ 300.0,
569
+ 400.0,
570
+ 500.0
571
+ ],
572
+ "ood_winrate": [
573
+ 0.03333333333333333
574
+ ],
575
+ "ood_winrate_iter": [
576
+ 500.0
577
+ ]
578
+ }
579
+ }
580
+ }
581
+ }