End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.23,
-    "eval_accuracy": 0.7012195121951219,
-    "eval_loss": 0.8817618489265442,
-    "eval_runtime": 77.5612,
-    "eval_samples_per_second": 2.114,
-    "eval_steps_per_second": 0.271
+    "eval_accuracy": 0.7935483870967742,
+    "eval_loss": 0.5277948379516602,
+    "eval_runtime": 72.4283,
+    "eval_samples_per_second": 2.14,
+    "eval_steps_per_second": 0.276
 }

runs/Feb02_07-27-25_03b2ce95466a/events.out.tfevents.1675324412.03b2ce95466a.384474.2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:77274f206d2b150c2fee7229d1e9f7b60a1c52e62b93a9d048c9f33c4e90dbc3
-size 363
+oid sha256:0cf81dea5620fbf598327df0ac53bf1712a5ca90b7625378787873566aec3637
+size 686

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.23,
-    "eval_accuracy": 0.7012195121951219,
-    "eval_loss": 0.8817618489265442,
-    "eval_runtime": 77.5612,
-    "eval_samples_per_second": 2.114,
-    "eval_steps_per_second": 0.271
+    "eval_accuracy": 0.7935483870967742,
+    "eval_loss": 0.5277948379516602,
+    "eval_runtime": 72.4283,
+    "eval_samples_per_second": 2.14,
+    "eval_steps_per_second": 0.276
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8421052631578947,
+  "best_metric": 0.8857142857142857,
   "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-148",
   "epoch": 3.22972972972973,
   "global_step": 148,
@@ -10,163 +10,154 @@
     {
       "epoch": 0.07,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 2.4536,
+      "loss": 2.3448,
       "step": 10
     },
     {
       "epoch": 0.14,
       "learning_rate": 4.81203007518797e-05,
-      "loss": 2.4176,
+      "loss": 2.2048,
       "step": 20
     },
     {
       "epoch": 0.2,
       "learning_rate": 4.43609022556391e-05,
-      "loss": 2.1549,
+      "loss": 2.1711,
       "step": 30
     },
     {
       "epoch": 0.26,
-      "eval_accuracy": 0.32894736842105265,
-      "eval_loss": 2.0454306602478027,
-      "eval_runtime": 35.7982,
-      "eval_samples_per_second": 2.123,
-      "eval_steps_per_second": 0.279,
+      "eval_accuracy": 0.5142857142857142,
+      "eval_loss": 1.7582899332046509,
+      "eval_runtime": 23.3861,
+      "eval_samples_per_second": 2.993,
+      "eval_steps_per_second": 0.385,
       "step": 38
     },
     {
       "epoch": 1.01,
       "learning_rate": 4.0601503759398494e-05,
-      "loss": 1.9566,
+      "loss": 2.0056,
       "step": 40
     },
     {
       "epoch": 1.08,
       "learning_rate": 3.6842105263157895e-05,
-      "loss": 1.5598,
+      "loss": 1.6312,
       "step": 50
     },
     {
       "epoch": 1.15,
       "learning_rate": 3.3082706766917295e-05,
-      "loss": 1.3703,
+      "loss": 1.2896,
       "step": 60
     },
     {
       "epoch": 1.22,
       "learning_rate": 2.9323308270676693e-05,
-      "loss": 1.153,
+      "loss": 0.8671,
       "step": 70
     },
     {
       "epoch": 1.26,
-      "eval_accuracy": 0.5526315789473685,
-      "eval_loss": 1.4139705896377563,
-      "eval_runtime": 31.628,
-      "eval_samples_per_second": 2.403,
-      "eval_steps_per_second": 0.316,
+      "eval_accuracy": 0.7714285714285715,
+      "eval_loss": 0.8745585680007935,
+      "eval_runtime": 26.4231,
+      "eval_samples_per_second": 2.649,
+      "eval_steps_per_second": 0.341,
       "step": 76
     },
     {
       "epoch": 2.03,
       "learning_rate": 2.556390977443609e-05,
-      "loss": 0.9631,
+      "loss": 0.8836,
       "step": 80
     },
     {
       "epoch": 2.09,
       "learning_rate": 2.1804511278195487e-05,
-      "loss": 0.7152,
+      "loss": 0.7118,
       "step": 90
     },
     {
       "epoch": 2.16,
       "learning_rate": 1.8045112781954888e-05,
-      "loss": 0.5672,
+      "loss": 0.661,
       "step": 100
     },
     {
       "epoch": 2.23,
       "learning_rate": 1.4285714285714285e-05,
-      "loss": 0.6531,
+      "loss": 0.5099,
       "step": 110
     },
     {
       "epoch": 2.26,
-      "eval_accuracy": 0.6052631578947368,
-      "eval_loss": 1.0039249658584595,
-      "eval_runtime": 39.6049,
-      "eval_samples_per_second": 1.919,
-      "eval_steps_per_second": 0.252,
+      "eval_accuracy": 0.8142857142857143,
+      "eval_loss": 0.5396287441253662,
+      "eval_runtime": 23.1042,
+      "eval_samples_per_second": 3.03,
+      "eval_steps_per_second": 0.39,
       "step": 114
     },
     {
       "epoch": 3.04,
       "learning_rate": 1.0526315789473684e-05,
-      "loss": 0.4358,
+      "loss": 0.3588,
       "step": 120
     },
     {
       "epoch": 3.11,
       "learning_rate": 6.766917293233083e-06,
-      "loss": 0.4124,
+      "loss": 0.3546,
       "step": 130
     },
     {
       "epoch": 3.18,
       "learning_rate": 3.007518796992481e-06,
-      "loss": 0.442,
+      "loss": 0.3309,
       "step": 140
     },
     {
       "epoch": 3.23,
-      "eval_accuracy": 0.8421052631578947,
-      "eval_loss": 0.7881976962089539,
-      "eval_runtime": 39.3029,
-      "eval_samples_per_second": 1.934,
-      "eval_steps_per_second": 0.254,
+      "eval_accuracy": 0.8857142857142857,
+      "eval_loss": 0.42734411358833313,
+      "eval_runtime": 33.9304,
+      "eval_samples_per_second": 2.063,
+      "eval_steps_per_second": 0.265,
       "step": 148
     },
     {
       "epoch": 3.23,
       "step": 148,
-      "total_flos": 1.4642564654333952e+18,
-      "train_loss": 1.1917882690558563,
-      "train_runtime": 1556.2056,
-      "train_samples_per_second": 0.761,
-      "train_steps_per_second": 0.095
+      "total_flos": 1.460491890402263e+18,
+      "train_loss": 1.1209931856877096,
+      "train_runtime": 1473.4643,
+      "train_samples_per_second": 0.804,
+      "train_steps_per_second": 0.1
     },
     {
       "epoch": 3.23,
-      "eval_accuracy": 0.7012195121951219,
-      "eval_loss": 0.8817619681358337,
-      "eval_runtime": 83.6154,
-      "eval_samples_per_second": 1.961,
-      "eval_steps_per_second": 0.251,
+      "eval_accuracy": 0.7935483870967742,
+      "eval_loss": 0.5277948975563049,
+      "eval_runtime": 75.0622,
+      "eval_samples_per_second": 2.065,
+      "eval_steps_per_second": 0.266,
       "step": 148
     },
     {
       "epoch": 3.23,
-      "eval_accuracy": 0.7012195121951219,
-      "eval_loss": 0.8817618489265442,
-      "eval_runtime": 81.3017,
-      "eval_samples_per_second": 2.017,
-      "eval_steps_per_second": 0.258,
-      "step": 148
-    },
-    {
-      "epoch": 3.23,
-      "eval_accuracy": 0.7012195121951219,
-      "eval_loss": 0.8817618489265442,
-      "eval_runtime": 77.5612,
-      "eval_samples_per_second": 2.114,
-      "eval_steps_per_second": 0.271,
+      "eval_accuracy": 0.7935483870967742,
+      "eval_loss": 0.5277948379516602,
+      "eval_runtime": 72.4283,
+      "eval_samples_per_second": 2.14,
+      "eval_steps_per_second": 0.276,
       "step": 148
     }
   ],
   "max_steps": 148,
   "num_train_epochs": 9223372036854775807,
-  "total_flos": 1.4642564654333952e+18,
+  "total_flos": 1.460491890402263e+18,
   "trial_name": null,
   "trial_params": null
 }