End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.22972972972973,
-    "eval_accuracy": 0.8580645161290322,
-    "eval_loss": 0.435279905796051,
-    "eval_runtime": 14.7921,
-    "eval_samples_per_second": 10.479,
-    "eval_steps_per_second": 1.352
+    "eval_accuracy": 0.8193548387096774,
+    "eval_loss": 0.5397765040397644,
+    "eval_runtime": 12.6944,
+    "eval_samples_per_second": 12.21,
+    "eval_steps_per_second": 1.575
 }

runs/May18_20-34-00_DESKTOP-SGO0ICV/events.out.tfevents.1716035945.DESKTOP-SGO0ICV.24484.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71a61a0f8e2f70009003f9e31c74782006bc0cfb1437e8d361ef555d84153834
-size 411
+oid sha256:a57fcbb21c777ebca6ec0908fa8991587f1c75cae2f309f174579ce373eb2bed
+size 734

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.22972972972973,
-    "eval_accuracy": 0.8580645161290322,
-    "eval_loss": 0.435279905796051,
-    "eval_runtime": 14.7921,
-    "eval_samples_per_second": 10.479,
-    "eval_steps_per_second": 1.352
+    "eval_accuracy": 0.8193548387096774,
+    "eval_loss": 0.5397765040397644,
+    "eval_runtime": 12.6944,
+    "eval_samples_per_second": 12.21,
+    "eval_steps_per_second": 1.575
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8857142857142857,
+  "best_metric": 0.8285714285714286,
   "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": 6.154857635498047,
+      "grad_norm": 6.302127838134766,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 2.326,
+      "loss": 2.2754,
       "step": 10
     },
     {
       "epoch": 0.13513513513513514,
-      "grad_norm": 7.193238735198975,
+      "grad_norm": 7.385843276977539,
       "learning_rate": 4.81203007518797e-05,
-      "loss": 2.2205,
+      "loss": 2.237,
       "step": 20
     },
     {
       "epoch": 0.20270270270270271,
-      "grad_norm": 7.481975555419922,
+      "grad_norm": 8.139371871948242,
       "learning_rate": 4.43609022556391e-05,
-      "loss": 2.16,
+      "loss": 2.1597,
       "step": 30
     },
     {
       "epoch": 0.25675675675675674,
-      "eval_accuracy": 0.5,
-      "eval_loss": 1.7993043661117554,
-      "eval_runtime": 6.6774,
-      "eval_samples_per_second": 10.483,
-      "eval_steps_per_second": 1.348,
+      "eval_accuracy": 0.4857142857142857,
+      "eval_loss": 1.856022596359253,
+      "eval_runtime": 5.9364,
+      "eval_samples_per_second": 11.792,
+      "eval_steps_per_second": 1.516,
       "step": 38
     },
     {
       "epoch": 1.0135135135135136,
-      "grad_norm": 10.227862358093262,
+      "grad_norm": 10.158555030822754,
       "learning_rate": 4.0601503759398494e-05,
-      "loss": 1.9946,
+      "loss": 1.9968,
       "step": 40
     },
     {
       "epoch": 1.0810810810810811,
-      "grad_norm": 10.306306838989258,
+      "grad_norm": 11.944147109985352,
       "learning_rate": 3.6842105263157895e-05,
-      "loss": 1.5987,
+      "loss": 1.6488,
       "step": 50
     },
     {
       "epoch": 1.1486486486486487,
-      "grad_norm": 10.154570579528809,
+      "grad_norm": 14.284652709960938,
       "learning_rate": 3.3082706766917295e-05,
-      "loss": 1.197,
+      "loss": 1.3385,
       "step": 60
     },
     {
       "epoch": 1.2162162162162162,
-      "grad_norm": 8.003480911254883,
+      "grad_norm": 8.205511093139648,
       "learning_rate": 2.9323308270676693e-05,
-      "loss": 0.8394,
+      "loss": 0.9646,
       "step": 70
     },
     {
       "epoch": 1.2567567567567568,
-      "eval_accuracy": 0.7,
-      "eval_loss": 0.8820496201515198,
-      "eval_runtime": 6.7555,
-      "eval_samples_per_second": 10.362,
-      "eval_steps_per_second": 1.332,
+      "eval_accuracy": 0.6285714285714286,
+      "eval_loss": 1.090829849243164,
+      "eval_runtime": 5.8872,
+      "eval_samples_per_second": 11.89,
+      "eval_steps_per_second": 1.529,
       "step": 76
     },
     {
       "epoch": 2.027027027027027,
-      "grad_norm": 6.113378047943115,
+      "grad_norm": 4.834934711456299,
       "learning_rate": 2.556390977443609e-05,
-      "loss": 0.8036,
+      "loss": 0.9288,
       "step": 80
     },
     {
       "epoch": 2.0945945945945947,
-      "grad_norm": 13.039664268493652,
+      "grad_norm": 9.32314395904541,
       "learning_rate": 2.1804511278195487e-05,
-      "loss": 0.5384,
+      "loss": 0.6638,
       "step": 90
     },
     {
       "epoch": 2.1621621621621623,
-      "grad_norm": 14.060022354125977,
+      "grad_norm": 16.83292007446289,
       "learning_rate": 1.8045112781954888e-05,
-      "loss": 0.6038,
+      "loss": 0.6089,
       "step": 100
     },
     {
       "epoch": 2.22972972972973,
-      "grad_norm": 6.143197536468506,
+      "grad_norm": 6.560121536254883,
       "learning_rate": 1.4285714285714285e-05,
-      "loss": 0.4228,
+      "loss": 0.4806,
       "step": 110
     },
     {
       "epoch": 2.2567567567567566,
-      "eval_accuracy": 0.8428571428571429,
-      "eval_loss": 0.5060061812400818,
-      "eval_runtime": 7.8853,
-      "eval_samples_per_second": 8.877,
-      "eval_steps_per_second": 1.141,
+      "eval_accuracy": 0.7857142857142857,
+      "eval_loss": 0.5811253786087036,
+      "eval_runtime": 6.0891,
+      "eval_samples_per_second": 11.496,
+      "eval_steps_per_second": 1.478,
       "step": 114
     },
     {
       "epoch": 3.0405405405405403,
-      "grad_norm": 5.005883693695068,
+      "grad_norm": 3.987687587738037,
       "learning_rate": 1.0526315789473684e-05,
-      "loss": 0.3779,
+      "loss": 0.3834,
       "step": 120
     },
     {
       "epoch": 3.108108108108108,
-      "grad_norm": 4.216254711151123,
+      "grad_norm": 4.402498722076416,
       "learning_rate": 6.766917293233083e-06,
-      "loss": 0.2576,
+      "loss": 0.3325,
       "step": 130
     },
     {
       "epoch": 3.175675675675676,
-      "grad_norm": 8.001497268676758,
+      "grad_norm": 9.72280216217041,
       "learning_rate": 3.007518796992481e-06,
-      "loss": 0.3029,
+      "loss": 0.3196,
       "step": 140
     },
     {
       "epoch": 3.22972972972973,
-      "eval_accuracy": 0.8857142857142857,
-      "eval_loss": 0.38448596000671387,
-      "eval_runtime": 7.6412,
-      "eval_samples_per_second": 9.161,
-      "eval_steps_per_second": 1.178,
+      "eval_accuracy": 0.8285714285714286,
+      "eval_loss": 0.4874313771724701,
+      "eval_runtime": 6.4339,
+      "eval_samples_per_second": 10.88,
+      "eval_steps_per_second": 1.399,
       "step": 148
     },
     {
       "epoch": 3.22972972972973,
       "step": 148,
       "total_flos": 1.460491890402263e+18,
-      "train_loss": 1.0705470426662549,
-      "train_runtime": 391.8273,
-      "train_samples_per_second": 3.022,
-      "train_steps_per_second": 0.378
+      "train_loss": 1.1212589031941183,
+      "train_runtime": 252.7655,
+      "train_samples_per_second": 4.684,
+      "train_steps_per_second": 0.586
     },
     {
       "epoch": 3.22972972972973,
-      "eval_accuracy": 0.8580645161290322,
-      "eval_loss": 0.4352799952030182,
-      "eval_runtime": 17.7551,
-      "eval_samples_per_second": 8.73,
-      "eval_steps_per_second": 1.126,
+      "eval_accuracy": 0.8193548387096774,
+      "eval_loss": 0.5397763848304749,
+      "eval_runtime": 13.5527,
+      "eval_samples_per_second": 11.437,
+      "eval_steps_per_second": 1.476,
       "step": 148
     },
     {
       "epoch": 3.22972972972973,
-      "eval_accuracy": 0.8580645161290322,
-      "eval_loss": 0.435279905796051,
-      "eval_runtime": 14.7921,
-      "eval_samples_per_second": 10.479,
-      "eval_steps_per_second": 1.352,
+      "eval_accuracy": 0.8193548387096774,
+      "eval_loss": 0.5397765040397644,
+      "eval_runtime": 12.6944,
+      "eval_samples_per_second": 12.21,
+      "eval_steps_per_second": 1.575,
       "step": 148
     }
   ],