End of training

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.23,
-    "eval_accuracy": 0.8,
-    "eval_loss": 0.5069947242736816,
-    "eval_runtime": 282.168,
-    "eval_samples_per_second": 0.549,
-    "eval_steps_per_second": 0.071
+    "epoch": 3.16,
+    "eval_accuracy": 0.26691838512924776,
+    "eval_loss": 1.1199040412902832,
+    "eval_runtime": 8842.1517,
+    "eval_samples_per_second": 0.389,
+    "eval_steps_per_second": 0.049
 }

runs/Dec13_13-08-09_phillips-personal-workstation/events.out.tfevents.1702507967.phillips-personal-workstation.62087.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd8c5c0a61d3cb8957bdb189b690af8ceeb7cc13ed891903c3542aef0a9a5c31
-size 357
+oid sha256:b6383836360558a638299074c96aa5aaf8dc0c13a9906d4865769091f3968ca3
+size 674

test_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.23,
-    "eval_accuracy": 0.8,
-    "eval_loss": 0.5069947242736816,
-    "eval_runtime": 282.168,
-    "eval_samples_per_second": 0.549,
-    "eval_steps_per_second": 0.071
+    "epoch": 3.16,
+    "eval_accuracy": 0.26691838512924776,
+    "eval_loss": 1.1199040412902832,
+    "eval_runtime": 8842.1517,
+    "eval_samples_per_second": 0.389,
+    "eval_steps_per_second": 0.049
 }

trainer_state.json

@@ -1,163 +1,97 @@
 {
-  "best_metric": 0.8142857142857143,
-  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-114",
-  "epoch": 3.22972972972973,
-  "global_step": 148,
+  "best_metric": 0.4974337040205304,
+  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-32",
+  "epoch": 3.15625,
+  "global_step": 32,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.07,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 2.3201,
-      "step": 10
-    },
-    {
-      "epoch": 0.14,
-      "learning_rate": 4.81203007518797e-05,
-      "loss": 2.2057,
-      "step": 20
-    },
-    {
-      "epoch": 0.2,
-      "learning_rate": 4.43609022556391e-05,
-      "loss": 2.1013,
-      "step": 30
-    },
-    {
-      "epoch": 0.26,
-      "eval_accuracy": 0.5428571428571428,
-      "eval_loss": 1.6601073741912842,
-      "eval_runtime": 176.7487,
-      "eval_samples_per_second": 0.396,
-      "eval_steps_per_second": 0.051,
-      "step": 38
-    },
-    {
-      "epoch": 1.01,
-      "learning_rate": 4.0601503759398494e-05,
-      "loss": 1.835,
-      "step": 40
-    },
-    {
-      "epoch": 1.08,
-      "learning_rate": 3.6842105263157895e-05,
-      "loss": 1.4632,
-      "step": 50
-    },
-    {
-      "epoch": 1.15,
-      "learning_rate": 3.3082706766917295e-05,
-      "loss": 1.2188,
-      "step": 60
-    },
-    {
-      "epoch": 1.22,
-      "learning_rate": 2.9323308270676693e-05,
-      "loss": 0.8317,
-      "step": 70
+      "epoch": 0.28,
+      "eval_accuracy": 0.3678357570573139,
+      "eval_loss": 1.1095340251922607,
+      "eval_runtime": 6781.1418,
+      "eval_samples_per_second": 0.345,
+      "eval_steps_per_second": 0.043,
+      "step": 9
     },
     {
-      "epoch": 1.26,
-      "eval_accuracy": 0.5428571428571428,
-      "eval_loss": 1.0476030111312866,
-      "eval_runtime": 178.2896,
-      "eval_samples_per_second": 0.393,
-      "eval_steps_per_second": 0.05,
-      "step": 76
-    },
-    {
-      "epoch": 2.03,
-      "learning_rate": 2.556390977443609e-05,
-      "loss": 0.8961,
-      "step": 80
-    },
-    {
-      "epoch": 2.09,
-      "learning_rate": 2.1804511278195487e-05,
-      "loss": 0.6745,
-      "step": 90
-    },
-    {
-      "epoch": 2.16,
-      "learning_rate": 1.8045112781954888e-05,
-      "loss": 0.6848,
-      "step": 100
-    },
-    {
-      "epoch": 2.23,
-      "learning_rate": 1.4285714285714285e-05,
-      "loss": 0.4313,
-      "step": 110
-    },
-    {
-      "epoch": 2.26,
-      "eval_accuracy": 0.8142857142857143,
-      "eval_loss": 0.5404379963874817,
-      "eval_runtime": 138.1129,
-      "eval_samples_per_second": 0.507,
-      "eval_steps_per_second": 0.065,
-      "step": 114
-    },
-    {
-      "epoch": 3.04,
-      "learning_rate": 1.0526315789473684e-05,
-      "loss": 0.3644,
-      "step": 120
-    },
-    {
-      "epoch": 3.11,
-      "learning_rate": 6.766917293233083e-06,
-      "loss": 0.2791,
-      "step": 130
+      "epoch": 1.03,
+      "learning_rate": 3.928571428571429e-05,
+      "loss": 1.1547,
+      "step": 10
     },
     {
-      "epoch": 3.18,
-      "learning_rate": 3.007518796992481e-06,
-      "loss": 0.2871,
-      "step": 140
+      "epoch": 1.28,
+      "eval_accuracy": 0.25149700598802394,
+      "eval_loss": 1.0840415954589844,
+      "eval_runtime": 6565.429,
+      "eval_samples_per_second": 0.356,
+      "eval_steps_per_second": 0.045,
+      "step": 18
     },
     {
-      "epoch": 3.23,
-      "eval_accuracy": 0.8142857142857143,
-      "eval_loss": 0.455091267824173,
-      "eval_runtime": 134.2932,
-      "eval_samples_per_second": 0.521,
-      "eval_steps_per_second": 0.067,
-      "step": 148
+      "epoch": 2.06,
+      "learning_rate": 2.1428571428571428e-05,
+      "loss": 1.0917,
+      "step": 20
     },
     {
-      "epoch": 3.23,
-      "step": 148,
-      "total_flos": 1.460491890402263e+18,
-      "train_loss": 1.0675029545216947,
-      "train_runtime": 11717.5942,
-      "train_samples_per_second": 0.101,
-      "train_steps_per_second": 0.013
+      "epoch": 2.28,
+      "eval_accuracy": 0.48545765611633873,
+      "eval_loss": 1.037040114402771,
+      "eval_runtime": 6528.1498,
+      "eval_samples_per_second": 0.358,
+      "eval_steps_per_second": 0.045,
+      "step": 27
     },
     {
-      "epoch": 3.23,
-      "eval_accuracy": 0.8,
-      "eval_loss": 0.5069947838783264,
-      "eval_runtime": 284.4819,
-      "eval_samples_per_second": 0.545,
-      "eval_steps_per_second": 0.07,
-      "step": 148
+      "epoch": 3.09,
+      "learning_rate": 3.5714285714285714e-06,
+      "loss": 1.0976,
+      "step": 30
     },
     {
-      "epoch": 3.23,
-      "eval_accuracy": 0.8,
-      "eval_loss": 0.5069947242736816,
-      "eval_runtime": 282.168,
-      "eval_samples_per_second": 0.549,
-      "eval_steps_per_second": 0.071,
-      "step": 148
+      "epoch": 3.16,
+      "eval_accuracy": 0.4974337040205304,
+      "eval_loss": 1.0295767784118652,
+      "eval_runtime": 6784.8017,
+      "eval_samples_per_second": 0.345,
+      "eval_steps_per_second": 0.043,
+      "step": 32
+    },
+    {
+      "epoch": 3.16,
+      "step": 32,
+      "total_flos": 2.9282778998194176e+17,
+      "train_loss": 1.1151684671640396,
+      "train_runtime": 29569.9652,
+      "train_samples_per_second": 0.009,
+      "train_steps_per_second": 0.001
+    },
+    {
+      "epoch": 3.16,
+      "eval_accuracy": 0.26691838512924776,
+      "eval_loss": 1.1199040412902832,
+      "eval_runtime": 9068.5155,
+      "eval_samples_per_second": 0.38,
+      "eval_steps_per_second": 0.048,
+      "step": 32
+    },
+    {
+      "epoch": 3.16,
+      "eval_accuracy": 0.26691838512924776,
+      "eval_loss": 1.1199040412902832,
+      "eval_runtime": 8842.1517,
+      "eval_samples_per_second": 0.389,
+      "eval_steps_per_second": 0.049,
+      "step": 32
     }
   ],
-  "max_steps": 148,
+  "max_steps": 32,
   "num_train_epochs": 9223372036854775807,
-  "total_flos": 1.460491890402263e+18,
+  "total_flos": 2.9282778998194176e+17,
   "trial_name": null,
   "trial_params": null
 }