End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.22972972972973,
-    "eval_accuracy": 0.9032258064516129,
-    "eval_loss": 0.366041898727417,
-    "eval_runtime": 14.264,
-    "eval_samples_per_second": 10.866,
-    "eval_steps_per_second": 1.402
+    "eval_accuracy": 0.8709677419354839,
+    "eval_loss": 0.5119308829307556,
+    "eval_runtime": 12.5062,
+    "eval_samples_per_second": 12.394,
+    "eval_steps_per_second": 1.599
 }

runs/Oct14_15-28-32_119f7f22da2a/events.out.tfevents.1728919960.119f7f22da2a.2416.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:320ead990dbfa000bce970c8d7c589040fc815688f438473c6932a7ea933835e
-size 411
+oid sha256:83867a11bb91a5a6f15b639b868d48bcdd4bbf327e9d7bdd733e981024ee68dc
+size 734

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.22972972972973,
-    "eval_accuracy": 0.9032258064516129,
-    "eval_loss": 0.366041898727417,
-    "eval_runtime": 14.264,
-    "eval_samples_per_second": 10.866,
-    "eval_steps_per_second": 1.402
+    "eval_accuracy": 0.8709677419354839,
+    "eval_loss": 0.5119308829307556,
+    "eval_runtime": 12.5062,
+    "eval_samples_per_second": 12.394,
+    "eval_steps_per_second": 1.599
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9,
+  "best_metric": 0.9142857142857143,
   "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-148",
   "epoch": 3.22972972972973,
   "eval_steps": 500,
@@ -10,163 +10,163 @@
   "log_history": [
     {
       "epoch": 0.06756756756756757,
-      "grad_norm": 6.507430553436279,
+      "grad_norm": 6.351539134979248,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 2.3814,
+      "loss": 2.2741,
       "step": 10
     },
     {
       "epoch": 0.13513513513513514,
-      "grad_norm": 6.589715003967285,
+      "grad_norm": 7.739011287689209,
       "learning_rate": 4.81203007518797e-05,
-      "loss": 2.1846,
+      "loss": 2.2121,
       "step": 20
     },
     {
       "epoch": 0.20270270270270271,
-      "grad_norm": 7.477535247802734,
+      "grad_norm": 8.185063362121582,
       "learning_rate": 4.43609022556391e-05,
-      "loss": 2.0976,
+      "loss": 2.1008,
       "step": 30
     },
     {
       "epoch": 0.25675675675675674,
-      "eval_accuracy": 0.6428571428571429,
-      "eval_loss": 1.6405479907989502,
-      "eval_runtime": 6.199,
-      "eval_samples_per_second": 11.292,
-      "eval_steps_per_second": 1.452,
+      "eval_accuracy": 0.4142857142857143,
+      "eval_loss": 1.834967017173767,
+      "eval_runtime": 6.5432,
+      "eval_samples_per_second": 10.698,
+      "eval_steps_per_second": 1.375,
       "step": 38
     },
     {
       "epoch": 1.0135135135135136,
-      "grad_norm": 12.163784980773926,
+      "grad_norm": 8.868829727172852,
       "learning_rate": 4.0601503759398494e-05,
-      "loss": 1.8387,
+      "loss": 1.952,
       "step": 40
     },
     {
       "epoch": 1.0810810810810811,
-      "grad_norm": 22.693099975585938,
+      "grad_norm": 10.57504940032959,
       "learning_rate": 3.6842105263157895e-05,
-      "loss": 1.3768,
+      "loss": 1.5486,
       "step": 50
     },
     {
       "epoch": 1.1486486486486487,
-      "grad_norm": 9.277478218078613,
+      "grad_norm": 9.441951751708984,
       "learning_rate": 3.3082706766917295e-05,
-      "loss": 1.0588,
+      "loss": 1.2984,
       "step": 60
     },
     {
       "epoch": 1.2162162162162162,
-      "grad_norm": 5.01579475402832,
+      "grad_norm": 7.836465358734131,
       "learning_rate": 2.9323308270676693e-05,
-      "loss": 0.682,
+      "loss": 0.9279,
       "step": 70
     },
     {
       "epoch": 1.2567567567567568,
-      "eval_accuracy": 0.8,
-      "eval_loss": 0.7726017832756042,
-      "eval_runtime": 6.3464,
-      "eval_samples_per_second": 11.03,
-      "eval_steps_per_second": 1.418,
+      "eval_accuracy": 0.7142857142857143,
+      "eval_loss": 1.0198957920074463,
+      "eval_runtime": 6.5052,
+      "eval_samples_per_second": 10.761,
+      "eval_steps_per_second": 1.384,
       "step": 76
     },
     {
       "epoch": 2.027027027027027,
-      "grad_norm": 3.855884313583374,
+      "grad_norm": 10.87349796295166,
       "learning_rate": 2.556390977443609e-05,
-      "loss": 0.7679,
+      "loss": 1.0519,
       "step": 80
     },
     {
       "epoch": 2.0945945945945947,
-      "grad_norm": 7.461369514465332,
+      "grad_norm": 9.65088176727295,
       "learning_rate": 2.1804511278195487e-05,
-      "loss": 0.554,
+      "loss": 0.7748,
       "step": 90
     },
     {
       "epoch": 2.1621621621621623,
-      "grad_norm": 10.099495887756348,
+      "grad_norm": 12.413875579833984,
       "learning_rate": 1.8045112781954888e-05,
-      "loss": 0.4587,
+      "loss": 0.6615,
       "step": 100
     },
     {
       "epoch": 2.22972972972973,
-      "grad_norm": 11.261001586914062,
+      "grad_norm": 7.036044597625732,
       "learning_rate": 1.4285714285714285e-05,
-      "loss": 0.4104,
+      "loss": 0.5482,
       "step": 110
     },
     {
       "epoch": 2.2567567567567566,
-      "eval_accuracy": 0.8571428571428571,
-      "eval_loss": 0.42056986689567566,
-      "eval_runtime": 6.2204,
-      "eval_samples_per_second": 11.253,
-      "eval_steps_per_second": 1.447,
+      "eval_accuracy": 0.8,
+      "eval_loss": 0.6185019612312317,
+      "eval_runtime": 6.5402,
+      "eval_samples_per_second": 10.703,
+      "eval_steps_per_second": 1.376,
       "step": 114
     },
     {
       "epoch": 3.0405405405405403,
-      "grad_norm": 3.727363109588623,
+      "grad_norm": 3.553114175796509,
       "learning_rate": 1.0526315789473684e-05,
-      "loss": 0.2675,
+      "loss": 0.4077,
       "step": 120
     },
     {
       "epoch": 3.108108108108108,
-      "grad_norm": 3.014127731323242,
+      "grad_norm": 3.6690454483032227,
       "learning_rate": 6.766917293233083e-06,
-      "loss": 0.236,
+      "loss": 0.3242,
       "step": 130
     },
     {
       "epoch": 3.175675675675676,
-      "grad_norm": 8.83653450012207,
+      "grad_norm": 13.059000015258789,
       "learning_rate": 3.007518796992481e-06,
-      "loss": 0.2605,
+      "loss": 0.3476,
       "step": 140
     },
     {
       "epoch": 3.22972972972973,
-      "eval_accuracy": 0.9,
-      "eval_loss": 0.3387860059738159,
-      "eval_runtime": 6.6102,
-      "eval_samples_per_second": 10.59,
-      "eval_steps_per_second": 1.362,
+      "eval_accuracy": 0.9142857142857143,
+      "eval_loss": 0.40149447321891785,
+      "eval_runtime": 6.6746,
+      "eval_samples_per_second": 10.487,
+      "eval_steps_per_second": 1.348,
       "step": 148
     },
     {
       "epoch": 3.22972972972973,
       "step": 148,
       "total_flos": 1.460491890402263e+18,
-      "train_loss": 0.9973534748360917,
-      "train_runtime": 219.7293,
-      "train_samples_per_second": 5.388,
-      "train_steps_per_second": 0.674
+      "train_loss": 1.1264432153186283,
+      "train_runtime": 223.8174,
+      "train_samples_per_second": 5.29,
+      "train_steps_per_second": 0.661
     },
     {
       "epoch": 3.22972972972973,
-      "eval_accuracy": 0.9032258064516129,
-      "eval_loss": 0.366041898727417,
-      "eval_runtime": 15.0666,
-      "eval_samples_per_second": 10.288,
-      "eval_steps_per_second": 1.327,
+      "eval_accuracy": 0.8709677419354839,
+      "eval_loss": 0.5119309425354004,
+      "eval_runtime": 13.3003,
+      "eval_samples_per_second": 11.654,
+      "eval_steps_per_second": 1.504,
       "step": 148
     },
     {
       "epoch": 3.22972972972973,
-      "eval_accuracy": 0.9032258064516129,
-      "eval_loss": 0.366041898727417,
-      "eval_runtime": 14.264,
-      "eval_samples_per_second": 10.866,
-      "eval_steps_per_second": 1.402,
+      "eval_accuracy": 0.8709677419354839,
+      "eval_loss": 0.5119308829307556,
+      "eval_runtime": 12.5062,
+      "eval_samples_per_second": 12.394,
+      "eval_steps_per_second": 1.599,
       "step": 148
     }
   ],