End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.22972972972973,
-    "eval_accuracy": 0.8516129032258064,
-    "eval_loss": 0.40030112862586975,
-    "eval_runtime": 12.2214,
-    "eval_samples_per_second": 12.683,
-    "eval_steps_per_second": 1.636
+    "eval_accuracy": 0.9032258064516129,
+    "eval_loss": 0.366041898727417,
+    "eval_runtime": 14.264,
+    "eval_samples_per_second": 10.866,
+    "eval_steps_per_second": 1.402
 }

runs/Oct14_15-11-34_71fb6b0700bc/events.out.tfevents.1728918935.71fb6b0700bc.15107.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce284a92b40765bea808299d6056894738f48e8a2387e585aea40879203250ae
-size 411
+oid sha256:0fcfcf2800a8eaeb8337b6189e39cb2cb510f8d10ce4e5c5330164c74e04e7cc
+size 734

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.22972972972973,
-    "eval_accuracy": 0.8516129032258064,
-    "eval_loss": 0.40030112862586975,
-    "eval_runtime": 12.2214,
-    "eval_samples_per_second": 12.683,
-    "eval_steps_per_second": 1.636
+    "eval_accuracy": 0.9032258064516129,
+    "eval_loss": 0.366041898727417,
+    "eval_runtime": 14.264,
+    "eval_samples_per_second": 10.866,
+    "eval_steps_per_second": 1.402
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8714285714285714,
+  "best_metric": 0.9,
   "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-148",
   "epoch": 3.22972972972973,
   "eval_steps": 500,
@@ -10,135 +10,135 @@
   "log_history": [
     {
       "epoch": 0.06756756756756757,
-      "grad_norm": 6.236786842346191,
+      "grad_norm": 6.507430553436279,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 2.3922,
+      "loss": 2.3814,
       "step": 10
     },
     {
       "epoch": 0.13513513513513514,
-      "grad_norm": 6.904232978820801,
+      "grad_norm": 6.589715003967285,
       "learning_rate": 4.81203007518797e-05,
-      "loss": 2.2012,
+      "loss": 2.1846,
       "step": 20
     },
     {
       "epoch": 0.20270270270270271,
-      "grad_norm": 7.755519390106201,
+      "grad_norm": 7.477535247802734,
       "learning_rate": 4.43609022556391e-05,
-      "loss": 2.1814,
+      "loss": 2.0976,
       "step": 30
     },
     {
       "epoch": 0.25675675675675674,
-      "eval_accuracy": 0.4714285714285714,
-      "eval_loss": 1.857365369796753,
-      "eval_runtime": 6.4158,
-      "eval_samples_per_second": 10.911,
-      "eval_steps_per_second": 1.403,
+      "eval_accuracy": 0.6428571428571429,
+      "eval_loss": 1.6405479907989502,
+      "eval_runtime": 6.199,
+      "eval_samples_per_second": 11.292,
+      "eval_steps_per_second": 1.452,
       "step": 38
     },
     {
       "epoch": 1.0135135135135136,
-      "grad_norm": 8.18625259399414,
+      "grad_norm": 12.163784980773926,
       "learning_rate": 4.0601503759398494e-05,
-      "loss": 2.0015,
+      "loss": 1.8387,
       "step": 40
     },
     {
       "epoch": 1.0810810810810811,
-      "grad_norm": 11.397420883178711,
+      "grad_norm": 22.693099975585938,
       "learning_rate": 3.6842105263157895e-05,
-      "loss": 1.6643,
+      "loss": 1.3768,
       "step": 50
     },
     {
       "epoch": 1.1486486486486487,
-      "grad_norm": 11.84655475616455,
+      "grad_norm": 9.277478218078613,
       "learning_rate": 3.3082706766917295e-05,
-      "loss": 1.2146,
+      "loss": 1.0588,
       "step": 60
     },
     {
       "epoch": 1.2162162162162162,
-      "grad_norm": 6.57381534576416,
+      "grad_norm": 5.01579475402832,
       "learning_rate": 2.9323308270676693e-05,
-      "loss": 0.8234,
+      "loss": 0.682,
       "step": 70
     },
     {
       "epoch": 1.2567567567567568,
-      "eval_accuracy": 0.7,
-      "eval_loss": 0.9412073493003845,
-      "eval_runtime": 6.491,
-      "eval_samples_per_second": 10.784,
-      "eval_steps_per_second": 1.387,
+      "eval_accuracy": 0.8,
+      "eval_loss": 0.7726017832756042,
+      "eval_runtime": 6.3464,
+      "eval_samples_per_second": 11.03,
+      "eval_steps_per_second": 1.418,
       "step": 76
     },
     {
       "epoch": 2.027027027027027,
-      "grad_norm": 7.174142837524414,
+      "grad_norm": 3.855884313583374,
       "learning_rate": 2.556390977443609e-05,
-      "loss": 0.8626,
+      "loss": 0.7679,
       "step": 80
     },
     {
       "epoch": 2.0945945945945947,
-      "grad_norm": 9.910486221313477,
+      "grad_norm": 7.461369514465332,
       "learning_rate": 2.1804511278195487e-05,
-      "loss": 0.6279,
+      "loss": 0.554,
       "step": 90
     },
     {
       "epoch": 2.1621621621621623,
-      "grad_norm": 18.203344345092773,
+      "grad_norm": 10.099495887756348,
       "learning_rate": 1.8045112781954888e-05,
-      "loss": 0.5971,
+      "loss": 0.4587,
       "step": 100
     },
     {
       "epoch": 2.22972972972973,
-      "grad_norm": 5.582463264465332,
+      "grad_norm": 11.261001586914062,
       "learning_rate": 1.4285714285714285e-05,
-      "loss": 0.467,
+      "loss": 0.4104,
       "step": 110
     },
     {
       "epoch": 2.2567567567567566,
-      "eval_accuracy": 0.7571428571428571,
-      "eval_loss": 0.5703738331794739,
-      "eval_runtime": 6.3262,
-      "eval_samples_per_second": 11.065,
-      "eval_steps_per_second": 1.423,
+      "eval_accuracy": 0.8571428571428571,
+      "eval_loss": 0.42056986689567566,
+      "eval_runtime": 6.2204,
+      "eval_samples_per_second": 11.253,
+      "eval_steps_per_second": 1.447,
       "step": 114
     },
     {
       "epoch": 3.0405405405405403,
-      "grad_norm": 3.68727707862854,
+      "grad_norm": 3.727363109588623,
       "learning_rate": 1.0526315789473684e-05,
-      "loss": 0.4091,
+      "loss": 0.2675,
       "step": 120
     },
     {
       "epoch": 3.108108108108108,
-      "grad_norm": 3.5390589237213135,
+      "grad_norm": 3.014127731323242,
       "learning_rate": 6.766917293233083e-06,
-      "loss": 0.2842,
+      "loss": 0.236,
       "step": 130
     },
     {
       "epoch": 3.175675675675676,
-      "grad_norm": 8.368684768676758,
+      "grad_norm": 8.83653450012207,
       "learning_rate": 3.007518796992481e-06,
-      "loss": 0.3374,
+      "loss": 0.2605,
       "step": 140
     },
     {
       "epoch": 3.22972972972973,
-      "eval_accuracy": 0.8714285714285714,
-      "eval_loss": 0.42163416743278503,
-      "eval_runtime": 6.6066,
-      "eval_samples_per_second": 10.595,
+      "eval_accuracy": 0.9,
+      "eval_loss": 0.3387860059738159,
+      "eval_runtime": 6.6102,
+      "eval_samples_per_second": 10.59,
       "eval_steps_per_second": 1.362,
       "step": 148
     },
@@ -146,27 +146,27 @@
       "epoch": 3.22972972972973,
       "step": 148,
       "total_flos": 1.460491890402263e+18,
-      "train_loss": 1.0991543885823842,
-      "train_runtime": 217.7469,
-      "train_samples_per_second": 5.438,
-      "train_steps_per_second": 0.68
+      "train_loss": 0.9973534748360917,
+      "train_runtime": 219.7293,
+      "train_samples_per_second": 5.388,
+      "train_steps_per_second": 0.674
     },
     {
       "epoch": 3.22972972972973,
-      "eval_accuracy": 0.8516129032258064,
-      "eval_loss": 0.40030112862586975,
-      "eval_runtime": 13.2529,
-      "eval_samples_per_second": 11.696,
-      "eval_steps_per_second": 1.509,
+      "eval_accuracy": 0.9032258064516129,
+      "eval_loss": 0.366041898727417,
+      "eval_runtime": 15.0666,
+      "eval_samples_per_second": 10.288,
+      "eval_steps_per_second": 1.327,
       "step": 148
     },
     {
       "epoch": 3.22972972972973,
-      "eval_accuracy": 0.8516129032258064,
-      "eval_loss": 0.40030112862586975,
-      "eval_runtime": 12.2214,
-      "eval_samples_per_second": 12.683,
-      "eval_steps_per_second": 1.636,
+      "eval_accuracy": 0.9032258064516129,
+      "eval_loss": 0.366041898727417,
+      "eval_runtime": 14.264,
+      "eval_samples_per_second": 10.866,
+      "eval_steps_per_second": 1.402,
       "step": 148
     }
   ],