Training in progress, epoch 0

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.5,
-    "eval_accuracy": 0.864516129032258,
-    "eval_loss": 0.4195975959300995,
-    "eval_runtime": 48.7546,
-    "eval_samples_per_second": 3.179,
-    "eval_steps_per_second": 1.6
+    "eval_accuracy": 0.7870967741935484,
+    "eval_loss": 0.6657504439353943,
+    "eval_runtime": 42.1495,
+    "eval_samples_per_second": 3.677,
+    "eval_steps_per_second": 1.851
 }

config.json

@@ -12,31 +12,15 @@
   "hidden_dropout_prob": 0.0,
   "hidden_size": 768,
   "id2label": {
-    "0": "ApplyEyeMakeup",
-    "1": "ApplyLipstick",
-    "2": "Archery",
-    "3": "BabyCrawling",
-    "4": "BalanceBeam",
-    "5": "BandMarching",
-    "6": "BaseballPitch",
-    "7": "Basketball",
-    "8": "BasketballDunk",
-    "9": "BenchPress"
+    "0": "Not Risky",
+    "1": "Risky"
   },
   "image_size": 224,
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "ApplyEyeMakeup": 0,
-    "ApplyLipstick": 1,
-    "Archery": 2,
-    "BabyCrawling": 3,
-    "BalanceBeam": 4,
-    "BandMarching": 5,
-    "BaseballPitch": 6,
-    "Basketball": 7,
-    "BasketballDunk": 8,
-    "BenchPress": 9
+    "Not Risky": 0,
+    "Risky": 1
   },
   "layer_norm_eps": 1e-12,
   "model_type": "videomae",

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb1a8944c5deb3d281add9f8895c3fbfb918a0b4ae2144d6902db5a18c93e325
-size 344961984
+oid sha256:aeee4826e549aec20a78f2a13875b8392b38e5eaf3d7715ea079ab583283122d
+size 344937368

runs/Jul02_18-49-41_c6d9a0630dcb/events.out.tfevents.1719946836.c6d9a0630dcb.4133.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0aa30ed1a5c4be9c241dc0494892fac26cd57ac5dcd9b7a96d068b02d43d5c5a
-size 411
+oid sha256:6fee780fc4d286169c1067496efa80e90ea0c8ae16d7a6bb146a138fca73a1a1
+size 734

runs/Jul02_19-27-25_c6d9a0630dcb/events.out.tfevents.1719948503.c6d9a0630dcb.4133.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8145403af70de9dd892277faa752f7973749eb9a0b1c8d4495fd6a53fb8d72e4
+size 5843

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.5,
-    "eval_accuracy": 0.864516129032258,
-    "eval_loss": 0.4195975959300995,
-    "eval_runtime": 48.7546,
-    "eval_samples_per_second": 3.179,
-    "eval_steps_per_second": 1.6
+    "eval_accuracy": 0.7870967741935484,
+    "eval_loss": 0.6657504439353943,
+    "eval_runtime": 42.1495,
+    "eval_samples_per_second": 3.677,
+    "eval_steps_per_second": 1.851
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8428571428571429,
+  "best_metric": 0.7857142857142857,
   "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-300",
   "epoch": 1.5,
   "eval_steps": 500,
@@ -9,236 +9,280 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.03,
+      "epoch": 0.03333333333333333,
+      "grad_norm": 12.915124893188477,
       "learning_rate": 1.6666666666666667e-05,
-      "loss": 2.3028,
+      "loss": 2.3784,
       "step": 10
     },
     {
-      "epoch": 0.07,
+      "epoch": 0.06666666666666667,
+      "grad_norm": 12.616281509399414,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 2.3514,
+      "loss": 2.3684,
       "step": 20
     },
     {
       "epoch": 0.1,
+      "grad_norm": 13.955244064331055,
       "learning_rate": 5e-05,
-      "loss": 2.3305,
+      "loss": 2.2632,
       "step": 30
     },
     {
-      "epoch": 0.13,
+      "epoch": 0.13333333333333333,
+      "grad_norm": 15.318857192993164,
       "learning_rate": 4.814814814814815e-05,
-      "loss": 2.2754,
+      "loss": 2.2229,
       "step": 40
     },
     {
-      "epoch": 0.17,
+      "epoch": 0.16666666666666666,
+      "grad_norm": 18.756649017333984,
       "learning_rate": 4.62962962962963e-05,
-      "loss": 2.0044,
+      "loss": 2.0136,
       "step": 50
     },
     {
       "epoch": 0.2,
+      "grad_norm": 16.19718360900879,
       "learning_rate": 4.4444444444444447e-05,
-      "loss": 2.1435,
+      "loss": 2.1816,
       "step": 60
     },
     {
-      "epoch": 0.23,
+      "epoch": 0.23333333333333334,
+      "grad_norm": 17.45854949951172,
       "learning_rate": 4.259259259259259e-05,
-      "loss": 2.0639,
+      "loss": 2.1317,
       "step": 70
     },
     {
-      "epoch": 0.27,
+      "epoch": 0.26666666666666666,
+      "grad_norm": 24.390531539916992,
       "learning_rate": 4.074074074074074e-05,
-      "loss": 1.6206,
+      "loss": 1.6419,
       "step": 80
     },
     {
       "epoch": 0.3,
+      "grad_norm": 18.53119468688965,
       "learning_rate": 3.888888888888889e-05,
-      "loss": 1.7889,
+      "loss": 1.8912,
       "step": 90
     },
     {
-      "epoch": 0.33,
+      "epoch": 0.3333333333333333,
+      "grad_norm": 23.2191219329834,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 1.6838,
+      "loss": 1.6417,
       "step": 100
     },
     {
-      "epoch": 0.37,
+      "epoch": 0.36666666666666664,
+      "grad_norm": 19.124753952026367,
       "learning_rate": 3.518518518518519e-05,
-      "loss": 1.3599,
+      "loss": 1.7508,
       "step": 110
     },
     {
       "epoch": 0.4,
+      "grad_norm": 7.332584381103516,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 1.4852,
+      "loss": 1.6249,
       "step": 120
     },
     {
-      "epoch": 0.43,
+      "epoch": 0.43333333333333335,
+      "grad_norm": 23.19915199279785,
       "learning_rate": 3.148148148148148e-05,
-      "loss": 1.4107,
+      "loss": 1.531,
       "step": 130
     },
     {
-      "epoch": 0.47,
+      "epoch": 0.4666666666666667,
+      "grad_norm": 16.055986404418945,
       "learning_rate": 2.962962962962963e-05,
-      "loss": 1.374,
+      "loss": 1.4367,
       "step": 140
     },
     {
       "epoch": 0.5,
+      "grad_norm": 19.868087768554688,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 1.0549,
+      "loss": 1.3617,
       "step": 150
     },
     {
       "epoch": 0.5,
-      "eval_accuracy": 0.7571428571428571,
-      "eval_loss": 0.9527812600135803,
-      "eval_runtime": 22.0926,
-      "eval_samples_per_second": 3.168,
-      "eval_steps_per_second": 1.584,
+      "eval_accuracy": 0.5,
+      "eval_loss": 1.3257172107696533,
+      "eval_runtime": 19.376,
+      "eval_samples_per_second": 3.613,
+      "eval_steps_per_second": 1.806,
       "step": 150
     },
     {
-      "epoch": 1.03,
+      "epoch": 1.0333333333333334,
+      "grad_norm": 17.58294105529785,
       "learning_rate": 2.5925925925925925e-05,
-      "loss": 0.8599,
+      "loss": 0.9594,
       "step": 160
     },
     {
-      "epoch": 1.07,
+      "epoch": 1.0666666666666667,
+      "grad_norm": 23.686206817626953,
       "learning_rate": 2.4074074074074074e-05,
-      "loss": 1.287,
+      "loss": 1.1516,
       "step": 170
     },
     {
       "epoch": 1.1,
+      "grad_norm": 26.428407669067383,
       "learning_rate": 2.2222222222222223e-05,
-      "loss": 0.9601,
+      "loss": 0.906,
       "step": 180
     },
     {
-      "epoch": 1.13,
+      "epoch": 1.1333333333333333,
+      "grad_norm": 19.449148178100586,
       "learning_rate": 2.037037037037037e-05,
-      "loss": 0.6645,
+      "loss": 0.7873,
       "step": 190
     },
     {
-      "epoch": 1.17,
+      "epoch": 1.1666666666666667,
+      "grad_norm": 40.24578857421875,
       "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.6143,
+      "loss": 1.1222,
       "step": 200
     },
     {
       "epoch": 1.2,
+      "grad_norm": 15.197308540344238,
       "learning_rate": 1.6666666666666667e-05,
-      "loss": 0.4838,
+      "loss": 0.6263,
       "step": 210
     },
     {
-      "epoch": 1.23,
+      "epoch": 1.2333333333333334,
+      "grad_norm": 15.162151336669922,
       "learning_rate": 1.4814814814814815e-05,
-      "loss": 0.4428,
+      "loss": 0.6937,
       "step": 220
     },
     {
-      "epoch": 1.27,
+      "epoch": 1.2666666666666666,
+      "grad_norm": 5.493642330169678,
       "learning_rate": 1.2962962962962962e-05,
-      "loss": 0.7375,
+      "loss": 0.806,
       "step": 230
     },
     {
       "epoch": 1.3,
+      "grad_norm": 5.520137310028076,
       "learning_rate": 1.1111111111111112e-05,
-      "loss": 0.6078,
+      "loss": 0.5812,
       "step": 240
     },
     {
-      "epoch": 1.33,
+      "epoch": 1.3333333333333333,
+      "grad_norm": 28.708820343017578,
       "learning_rate": 9.259259259259259e-06,
-      "loss": 0.4268,
+      "loss": 0.7231,
       "step": 250
     },
     {
-      "epoch": 1.37,
+      "epoch": 1.3666666666666667,
+      "grad_norm": 1.0324937105178833,
       "learning_rate": 7.4074074074074075e-06,
-      "loss": 0.2414,
+      "loss": 0.3939,
       "step": 260
     },
     {
       "epoch": 1.4,
+      "grad_norm": 31.389986038208008,
       "learning_rate": 5.555555555555556e-06,
-      "loss": 0.4216,
+      "loss": 0.4833,
       "step": 270
     },
     {
-      "epoch": 1.43,
+      "epoch": 1.4333333333333333,
+      "grad_norm": 15.664688110351562,
       "learning_rate": 3.7037037037037037e-06,
-      "loss": 0.1838,
+      "loss": 0.2994,
       "step": 280
     },
     {
-      "epoch": 1.47,
+      "epoch": 1.4666666666666668,
+      "grad_norm": 12.544212341308594,
       "learning_rate": 1.8518518518518519e-06,
-      "loss": 0.5178,
+      "loss": 0.8584,
       "step": 290
     },
     {
       "epoch": 1.5,
+      "grad_norm": 7.230623722076416,
       "learning_rate": 0.0,
-      "loss": 0.7251,
+      "loss": 0.7925,
       "step": 300
     },
     {
       "epoch": 1.5,
-      "eval_accuracy": 0.8428571428571429,
-      "eval_loss": 0.4445994794368744,
-      "eval_runtime": 21.5992,
-      "eval_samples_per_second": 3.241,
-      "eval_steps_per_second": 1.62,
+      "eval_accuracy": 0.7857142857142857,
+      "eval_loss": 0.6653470993041992,
+      "eval_runtime": 18.5919,
+      "eval_samples_per_second": 3.765,
+      "eval_steps_per_second": 1.883,
       "step": 300
     },
     {
       "epoch": 1.5,
       "step": 300,
       "total_flos": 7.47692094062592e+17,
-      "train_loss": 1.214137753645579,
-      "train_runtime": 449.6116,
-      "train_samples_per_second": 1.334,
-      "train_steps_per_second": 0.667
+      "train_loss": 1.320799682935079,
+      "train_runtime": 404.8672,
+      "train_samples_per_second": 1.482,
+      "train_steps_per_second": 0.741
     },
     {
       "epoch": 1.5,
-      "eval_accuracy": 0.864516129032258,
-      "eval_loss": 0.4195975959300995,
-      "eval_runtime": 49.2274,
-      "eval_samples_per_second": 3.149,
-      "eval_steps_per_second": 1.584,
+      "eval_accuracy": 0.7870967741935484,
+      "eval_loss": 0.6657505035400391,
+      "eval_runtime": 42.1038,
+      "eval_samples_per_second": 3.681,
+      "eval_steps_per_second": 1.853,
       "step": 300
     },
     {
       "epoch": 1.5,
-      "eval_accuracy": 0.864516129032258,
-      "eval_loss": 0.4195975959300995,
-      "eval_runtime": 48.7546,
-      "eval_samples_per_second": 3.179,
-      "eval_steps_per_second": 1.6,
+      "eval_accuracy": 0.7870967741935484,
+      "eval_loss": 0.6657504439353943,
+      "eval_runtime": 42.1495,
+      "eval_samples_per_second": 3.677,
+      "eval_steps_per_second": 1.851,
       "step": 300
     }
   ],
   "logging_steps": 10,
   "max_steps": 300,
+  "num_input_tokens_seen": 0,
   "num_train_epochs": 9223372036854775807,
   "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
   "total_flos": 7.47692094062592e+17,
+  "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e2ed80c6fe797116a572ca36cda6dde9e15ceca68db15f4c972c6726f0cb3d4
+oid sha256:5cba774569b4f69bf8b3e151f301f8a81a154c061b54bceb7919d0d01e97c129
 size 5176