End of training

all_results.json

@@ -1,17 +1,17 @@
 {
-    "epoch": 7.0,
-    "eval_accuracy": 0.6666666666666666,
-    "eval_auc_class_0": 0.625,
-    "eval_auc_class_1": 0.75,
-    "eval_auc_class_2": 0.875,
-    "eval_auc_overall": 0.75,
-    "eval_loss": 0.9276485443115234,
-    "eval_runtime": 0.7838,
-    "eval_samples_per_second": 7.655,
-    "eval_steps_per_second": 1.276,
-    "total_flos": 8226380544399360.0,
-    "train_loss": 0.14502547468457902,
-    "train_runtime": 156.7549,
-    "train_samples_per_second": 0.67,
-    "train_steps_per_second": 0.045
+    "epoch": 6.885245901639344,
+    "eval_accuracy": 0.8772656654583117,
+    "eval_auc_class_0": 0.9495477284481044,
+    "eval_auc_class_1": 0.9596716610509716,
+    "eval_auc_class_2": 0.9626115194580721,
+    "eval_auc_overall": 0.9572769696523826,
+    "eval_loss": 0.30786964297294617,
+    "eval_runtime": 208.4141,
+    "eval_samples_per_second": 9.265,
+    "eval_steps_per_second": 0.077,
+    "total_flos": 4.1785312376666235e+18,
+    "train_loss": 0.38806929134187246,
+    "train_runtime": 12116.6555,
+    "train_samples_per_second": 4.47,
+    "train_steps_per_second": 0.009
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 7.0,
-    "eval_accuracy": 0.6666666666666666,
-    "eval_auc_class_0": 0.625,
-    "eval_auc_class_1": 0.75,
-    "eval_auc_class_2": 0.875,
-    "eval_auc_overall": 0.75,
-    "eval_loss": 0.9276485443115234,
-    "eval_runtime": 0.7838,
-    "eval_samples_per_second": 7.655,
-    "eval_steps_per_second": 1.276
+    "epoch": 6.885245901639344,
+    "eval_accuracy": 0.8772656654583117,
+    "eval_auc_class_0": 0.9495477284481044,
+    "eval_auc_class_1": 0.9596716610509716,
+    "eval_auc_class_2": 0.9626115194580721,
+    "eval_auc_overall": 0.9572769696523826,
+    "eval_loss": 0.30786964297294617,
+    "eval_runtime": 208.4141,
+    "eval_samples_per_second": 9.265,
+    "eval_steps_per_second": 0.077
 }

runs/Sep03_08-52-31_data-science-wbi/events.out.tfevents.1725365893.data-science-wbi

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:238dbe3ab42a9310ed1dec662a4d08a096f94596be2838d7688dda72f753c9db
+size 87

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 7.0,
-    "total_flos": 8226380544399360.0,
-    "train_loss": 0.14502547468457902,
-    "train_runtime": 156.7549,
-    "train_samples_per_second": 0.67,
-    "train_steps_per_second": 0.045
+    "epoch": 6.885245901639344,
+    "total_flos": 4.1785312376666235e+18,
+    "train_loss": 0.38806929134187246,
+    "train_runtime": 12116.6555,
+    "train_samples_per_second": 4.47,
+    "train_steps_per_second": 0.009
 }

trainer_state.json

@@ -1,207 +1,277 @@
 {
-  "best_metric": 0.6666666666666666,
-  "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-cons_Classification_base_V10/checkpoint-2",
-  "epoch": 7.0,
+  "best_metric": 0.8772656654583117,
+  "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-cons_Classification_base_V10/checkpoint-105",
+  "epoch": 6.885245901639344,
   "eval_steps": 500,
-  "global_step": 7,
+  "global_step": 105,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.0,
-      "step": 1,
-      "train_accuracy": 0.6,
-      "train_auc_class_0": 0.36,
-      "train_auc_class_1": 0.8,
-      "train_auc_class_2": 0.8400000000000001,
-      "train_auc_overall": 0.6666666666666666,
-      "train_loss": 0.9687549471855164,
-      "train_runtime": 1.665,
-      "train_samples_per_second": 9.009,
-      "train_steps_per_second": 0.601
-    },
-    {
-      "epoch": 1.0,
-      "eval_accuracy": 0.5,
-      "eval_auc_class_0": 0.625,
-      "eval_auc_class_1": 0.75,
-      "eval_auc_class_2": 0.75,
-      "eval_auc_overall": 0.7083333333333334,
-      "eval_loss": 1.028936505317688,
-      "eval_runtime": 0.6838,
-      "eval_samples_per_second": 8.774,
-      "eval_steps_per_second": 1.462,
-      "step": 1
-    },
-    {
-      "epoch": 2.0,
-      "step": 2,
-      "train_accuracy": 0.8,
-      "train_auc_class_0": 0.98,
-      "train_auc_class_1": 1.0,
-      "train_auc_class_2": 1.0,
-      "train_auc_overall": 0.9933333333333333,
-      "train_loss": 0.6606505513191223,
-      "train_runtime": 1.7751,
-      "train_samples_per_second": 8.45,
-      "train_steps_per_second": 0.563
-    },
-    {
-      "epoch": 2.0,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_auc_class_0": 0.625,
-      "eval_auc_class_1": 0.75,
-      "eval_auc_class_2": 0.875,
-      "eval_auc_overall": 0.75,
-      "eval_loss": 0.9276485443115234,
-      "eval_runtime": 0.7125,
-      "eval_samples_per_second": 8.421,
-      "eval_steps_per_second": 1.404,
-      "step": 2
-    },
-    {
-      "epoch": 3.0,
-      "step": 3,
-      "train_accuracy": 1.0,
-      "train_auc_class_0": 1.0,
-      "train_auc_class_1": 1.0,
-      "train_auc_class_2": 1.0,
-      "train_auc_overall": 1.0,
-      "train_loss": 0.5248541831970215,
-      "train_runtime": 1.627,
-      "train_samples_per_second": 9.219,
-      "train_steps_per_second": 0.615
-    },
-    {
-      "epoch": 3.0,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_auc_class_0": 0.625,
-      "eval_auc_class_1": 1.0,
-      "eval_auc_class_2": 0.875,
-      "eval_auc_overall": 0.8333333333333334,
-      "eval_loss": 0.8520984053611755,
-      "eval_runtime": 0.7245,
-      "eval_samples_per_second": 8.281,
-      "eval_steps_per_second": 1.38,
-      "step": 3
+      "epoch": 0.6557377049180327,
+      "grad_norm": 4.642137050628662,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": 0.7975,
+      "step": 10
+    },
+    {
+      "epoch": 0.9836065573770492,
+      "step": 15,
+      "train_accuracy": 0.8178880703114902,
+      "train_auc_class_0": 0.8980438574304173,
+      "train_auc_class_1": 0.9139826205363317,
+      "train_auc_class_2": 0.8919882412630196,
+      "train_auc_overall": 0.9013382397432562,
+      "train_loss": 0.46714112162590027,
+      "train_runtime": 689.8789,
+      "train_samples_per_second": 11.215,
+      "train_steps_per_second": 0.088
+    },
+    {
+      "epoch": 0.9836065573770492,
+      "eval_accuracy": 0.8524080787156914,
+      "eval_auc_class_0": 0.9181721665274296,
+      "eval_auc_class_1": 0.9324510717614166,
+      "eval_auc_class_2": 0.9299151491872829,
+      "eval_auc_overall": 0.9268461291587097,
+      "eval_loss": 0.4173153042793274,
+      "eval_runtime": 342.4588,
+      "eval_samples_per_second": 5.639,
+      "eval_steps_per_second": 0.047,
+      "step": 15
+    },
+    {
+      "epoch": 1.3114754098360657,
+      "grad_norm": 2.1029186248779297,
+      "learning_rate": 4.5212765957446815e-05,
+      "loss": 0.5023,
+      "step": 20
+    },
+    {
+      "epoch": 1.9672131147540983,
+      "grad_norm": 1.8800333738327026,
+      "learning_rate": 3.9893617021276594e-05,
+      "loss": 0.4094,
+      "step": 30
+    },
+    {
+      "epoch": 1.9672131147540983,
+      "step": 30,
+      "train_accuracy": 0.8503295851105079,
+      "train_auc_class_0": 0.9326458902508805,
+      "train_auc_class_1": 0.9445850486238231,
+      "train_auc_class_2": 0.9445771168575358,
+      "train_auc_overall": 0.9406026852440799,
+      "train_loss": 0.3614955544471741,
+      "train_runtime": 693.803,
+      "train_samples_per_second": 11.152,
+      "train_steps_per_second": 0.088
+    },
+    {
+      "epoch": 1.9672131147540983,
+      "eval_accuracy": 0.849818746763335,
+      "eval_auc_class_0": 0.9336946035724231,
+      "eval_auc_class_1": 0.9488809233636819,
+      "eval_auc_class_2": 0.9534673603715279,
+      "eval_auc_overall": 0.9453476291025443,
+      "eval_loss": 0.3673703968524933,
+      "eval_runtime": 266.4299,
+      "eval_samples_per_second": 7.248,
+      "eval_steps_per_second": 0.06,
+      "step": 30
+    },
+    {
+      "epoch": 2.6229508196721314,
+      "grad_norm": 2.8551852703094482,
+      "learning_rate": 3.4574468085106386e-05,
+      "loss": 0.3672,
+      "step": 40
+    },
+    {
+      "epoch": 2.9508196721311473,
+      "step": 45,
+      "train_accuracy": 0.864417732971436,
+      "train_auc_class_0": 0.9455359129962017,
+      "train_auc_class_1": 0.9545585538591823,
+      "train_auc_class_2": 0.9593322426521459,
+      "train_auc_overall": 0.9531422365025098,
+      "train_loss": 0.32408276200294495,
+      "train_runtime": 685.3567,
+      "train_samples_per_second": 11.289,
+      "train_steps_per_second": 0.089
+    },
+    {
+      "epoch": 2.9508196721311473,
+      "eval_accuracy": 0.8726048679440704,
+      "eval_auc_class_0": 0.9429575923936826,
+      "eval_auc_class_1": 0.9536497239945516,
+      "eval_auc_class_2": 0.9603724007891477,
+      "eval_auc_overall": 0.9523265723924607,
+      "eval_loss": 0.3238150477409363,
+      "eval_runtime": 204.7854,
+      "eval_samples_per_second": 9.429,
+      "eval_steps_per_second": 0.078,
+      "step": 45
+    },
+    {
+      "epoch": 3.278688524590164,
+      "grad_norm": 2.7802083492279053,
+      "learning_rate": 2.925531914893617e-05,
+      "loss": 0.3328,
+      "step": 50
+    },
+    {
+      "epoch": 3.9344262295081966,
+      "grad_norm": 2.6021904945373535,
+      "learning_rate": 2.393617021276596e-05,
+      "loss": 0.3237,
+      "step": 60
     },
     {
       "epoch": 4.0,
-      "step": 4,
-      "train_accuracy": 1.0,
-      "train_auc_class_0": 1.0,
-      "train_auc_class_1": 1.0,
-      "train_auc_class_2": 1.0,
-      "train_auc_overall": 1.0,
-      "train_loss": 0.36024707555770874,
-      "train_runtime": 1.7829,
-      "train_samples_per_second": 8.413,
-      "train_steps_per_second": 0.561
+      "step": 61,
+      "train_accuracy": 0.8729481711257593,
+      "train_auc_class_0": 0.9494105682362736,
+      "train_auc_class_1": 0.960331196185584,
+      "train_auc_class_2": 0.9563328693868581,
+      "train_auc_overall": 0.955358211269572,
+      "train_loss": 0.31062838435173035,
+      "train_runtime": 681.8052,
+      "train_samples_per_second": 11.348,
+      "train_steps_per_second": 0.089
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_auc_class_0": 0.875,
-      "eval_auc_class_1": 1.0,
-      "eval_auc_class_2": 1.0,
-      "eval_auc_overall": 0.9583333333333334,
-      "eval_loss": 0.7887496948242188,
-      "eval_runtime": 0.8054,
-      "eval_samples_per_second": 7.45,
-      "eval_steps_per_second": 1.242,
-      "step": 4
-    },
-    {
-      "epoch": 5.0,
-      "step": 5,
-      "train_accuracy": 1.0,
-      "train_auc_class_0": 1.0,
-      "train_auc_class_1": 1.0,
-      "train_auc_class_2": 1.0,
-      "train_auc_overall": 1.0,
-      "train_loss": 0.31448933482170105,
-      "train_runtime": 1.9887,
-      "train_samples_per_second": 7.543,
-      "train_steps_per_second": 0.503
-    },
-    {
-      "epoch": 5.0,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_auc_class_0": 1.0,
-      "eval_auc_class_1": 1.0,
-      "eval_auc_class_2": 1.0,
-      "eval_auc_overall": 1.0,
-      "eval_loss": 0.7302556037902832,
-      "eval_runtime": 0.719,
-      "eval_samples_per_second": 8.345,
-      "eval_steps_per_second": 1.391,
-      "step": 5
-    },
-    {
-      "epoch": 6.0,
-      "step": 6,
-      "train_accuracy": 0.9333333333333333,
-      "train_auc_class_0": 1.0,
-      "train_auc_class_1": 1.0,
-      "train_auc_class_2": 1.0,
-      "train_auc_overall": 1.0,
-      "train_loss": 0.31753572821617126,
-      "train_runtime": 1.8048,
-      "train_samples_per_second": 8.311,
-      "train_steps_per_second": 0.554
-    },
-    {
-      "epoch": 6.0,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_auc_class_0": 1.0,
-      "eval_auc_class_1": 1.0,
-      "eval_auc_class_2": 1.0,
-      "eval_auc_overall": 1.0,
-      "eval_loss": 0.688157320022583,
-      "eval_runtime": 0.719,
-      "eval_samples_per_second": 8.345,
-      "eval_steps_per_second": 1.391,
-      "step": 6
-    },
-    {
-      "epoch": 7.0,
-      "step": 7,
-      "train_accuracy": 1.0,
-      "train_auc_class_0": 1.0,
-      "train_auc_class_1": 1.0,
-      "train_auc_class_2": 1.0,
-      "train_auc_overall": 1.0,
-      "train_loss": 0.27592089772224426,
-      "train_runtime": 2.0142,
-      "train_samples_per_second": 7.447,
-      "train_steps_per_second": 0.496
-    },
-    {
-      "epoch": 7.0,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_auc_class_0": 1.0,
-      "eval_auc_class_1": 1.0,
-      "eval_auc_class_2": 1.0,
-      "eval_auc_overall": 1.0,
-      "eval_loss": 0.6634145379066467,
-      "eval_runtime": 0.7768,
-      "eval_samples_per_second": 7.724,
-      "eval_steps_per_second": 1.287,
-      "step": 7
-    },
-    {
-      "epoch": 7.0,
-      "step": 7,
-      "total_flos": 8226380544399360.0,
-      "train_loss": 0.14502547468457902,
-      "train_runtime": 156.7549,
-      "train_samples_per_second": 0.67,
-      "train_steps_per_second": 0.045
+      "eval_accuracy": 0.8720870015535992,
+      "eval_auc_class_0": 0.9470949198392808,
+      "eval_auc_class_1": 0.9577675819055128,
+      "eval_auc_class_2": 0.961633814620179,
+      "eval_auc_overall": 0.9554987721216577,
+      "eval_loss": 0.32258161902427673,
+      "eval_runtime": 204.0189,
+      "eval_samples_per_second": 9.465,
+      "eval_steps_per_second": 0.078,
+      "step": 61
+    },
+    {
+      "epoch": 4.590163934426229,
+      "grad_norm": 2.246328592300415,
+      "learning_rate": 1.8617021276595745e-05,
+      "loss": 0.3257,
+      "step": 70
+    },
+    {
+      "epoch": 4.983606557377049,
+      "step": 76,
+      "train_accuracy": 0.879798371461807,
+      "train_auc_class_0": 0.9544922319365635,
+      "train_auc_class_1": 0.9627755801666535,
+      "train_auc_class_2": 0.9628861811778814,
+      "train_auc_overall": 0.9600513310936994,
+      "train_loss": 0.29983675479888916,
+      "train_runtime": 688.2036,
+      "train_samples_per_second": 11.242,
+      "train_steps_per_second": 0.089
+    },
+    {
+      "epoch": 4.983606557377049,
+      "eval_accuracy": 0.8741584671154842,
+      "eval_auc_class_0": 0.9478719119226638,
+      "eval_auc_class_1": 0.9577360384256937,
+      "eval_auc_class_2": 0.9608001466557257,
+      "eval_auc_overall": 0.9554693656680278,
+      "eval_loss": 0.32330793142318726,
+      "eval_runtime": 205.4973,
+      "eval_samples_per_second": 9.397,
+      "eval_steps_per_second": 0.078,
+      "step": 76
+    },
+    {
+      "epoch": 5.245901639344262,
+      "grad_norm": 1.7804102897644043,
+      "learning_rate": 1.3297872340425532e-05,
+      "loss": 0.2972,
+      "step": 80
+    },
+    {
+      "epoch": 5.901639344262295,
+      "grad_norm": 1.6500009298324585,
+      "learning_rate": 7.97872340425532e-06,
+      "loss": 0.2923,
+      "step": 90
+    },
+    {
+      "epoch": 5.967213114754099,
+      "step": 91,
+      "train_accuracy": 0.8857438283572444,
+      "train_auc_class_0": 0.9581288331197666,
+      "train_auc_class_1": 0.966850418082771,
+      "train_auc_class_2": 0.9694481750082071,
+      "train_auc_overall": 0.9648091420702481,
+      "train_loss": 0.28071513772010803,
+      "train_runtime": 686.7084,
+      "train_samples_per_second": 11.267,
+      "train_steps_per_second": 0.089
+    },
+    {
+      "epoch": 5.967213114754099,
+      "eval_accuracy": 0.8736406007250129,
+      "eval_auc_class_0": 0.9477737001233243,
+      "eval_auc_class_1": 0.9590063803856907,
+      "eval_auc_class_2": 0.9627250030553275,
+      "eval_auc_overall": 0.9565016945214474,
+      "eval_loss": 0.3128957450389862,
+      "eval_runtime": 206.0937,
+      "eval_samples_per_second": 9.37,
+      "eval_steps_per_second": 0.078,
+      "step": 91
+    },
+    {
+      "epoch": 6.557377049180328,
+      "grad_norm": 2.4247124195098877,
+      "learning_rate": 2.6595744680851065e-06,
+      "loss": 0.2794,
+      "step": 100
+    },
+    {
+      "epoch": 6.885245901639344,
+      "step": 105,
+      "train_accuracy": 0.8973762440222308,
+      "train_auc_class_0": 0.9638374809913217,
+      "train_auc_class_1": 0.9706794670135619,
+      "train_auc_class_2": 0.9722462334871682,
+      "train_auc_overall": 0.9689210604973506,
+      "train_loss": 0.26410263776779175,
+      "train_runtime": 684.8081,
+      "train_samples_per_second": 11.298,
+      "train_steps_per_second": 0.089
+    },
+    {
+      "epoch": 6.885245901639344,
+      "eval_accuracy": 0.8772656654583117,
+      "eval_auc_class_0": 0.9495477284481044,
+      "eval_auc_class_1": 0.9596716610509716,
+      "eval_auc_class_2": 0.9626115194580721,
+      "eval_auc_overall": 0.9572769696523826,
+      "eval_loss": 0.30786964297294617,
+      "eval_runtime": 205.1702,
+      "eval_samples_per_second": 9.412,
+      "eval_steps_per_second": 0.078,
+      "step": 105
+    },
+    {
+      "epoch": 6.885245901639344,
+      "step": 105,
+      "total_flos": 4.1785312376666235e+18,
+      "train_loss": 0.38806929134187246,
+      "train_runtime": 12116.6555,
+      "train_samples_per_second": 4.47,
+      "train_steps_per_second": 0.009
     }
   ],
   "logging_steps": 10,
-  "max_steps": 7,
+  "max_steps": 105,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 7,
   "save_steps": 500,
@@ -217,7 +287,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8226380544399360.0,
+  "total_flos": 4.1785312376666235e+18,
   "train_batch_size": 128,
   "trial_name": null,
   "trial_params": null