End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.2838709677419355,
-    "eval_loss": 1.9679656028747559,
-    "eval_runtime": 26.6207,
-    "eval_samples_per_second": 5.823,
-    "eval_steps_per_second": 1.465
+    "eval_accuracy": 0.27741935483870966,
+    "eval_loss": 1.9401745796203613,
+    "eval_runtime": 28.1588,
+    "eval_samples_per_second": 5.505,
+    "eval_steps_per_second": 1.385
 }

runs/May25_00-05-58_f1bac041167e/events.out.tfevents.1716595760.f1bac041167e.34.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f10ee2b297deb7126f4447f18d9df249655a8b2b0251971a9382a3727f2be9d4
-size 405
+oid sha256:e90073794d9b9b7a0aedf61f71a7bebcd74c896b94a508153db6b0c2a4d95157
+size 722

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.2838709677419355,
-    "eval_loss": 1.9679656028747559,
-    "eval_runtime": 26.6207,
-    "eval_samples_per_second": 5.823,
-    "eval_steps_per_second": 1.465
+    "eval_accuracy": 0.27741935483870966,
+    "eval_loss": 1.9401745796203613,
+    "eval_runtime": 28.1588,
+    "eval_samples_per_second": 5.505,
+    "eval_steps_per_second": 1.385
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.5428571428571428,
+  "best_metric": 0.35714285714285715,
   "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-75",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -10,87 +10,87 @@
   "log_history": [
     {
       "epoch": 0.13,
-      "grad_norm": 11.617264747619629,
+      "grad_norm": 9.760223388671875,
       "learning_rate": 4.850746268656717e-05,
-      "loss": 2.377,
+      "loss": 2.2528,
       "step": 10
     },
     {
       "epoch": 0.27,
-      "grad_norm": 10.894355773925781,
+      "grad_norm": 10.098638534545898,
       "learning_rate": 4.104477611940299e-05,
-      "loss": 2.2211,
+      "loss": 2.2017,
       "step": 20
     },
     {
       "epoch": 0.4,
-      "grad_norm": 8.908524513244629,
+      "grad_norm": 9.566693305969238,
       "learning_rate": 3.358208955223881e-05,
-      "loss": 2.1478,
+      "loss": 2.106,
       "step": 30
     },
     {
       "epoch": 0.53,
-      "grad_norm": 9.142258644104004,
+      "grad_norm": 9.71388053894043,
       "learning_rate": 2.6119402985074626e-05,
-      "loss": 2.109,
+      "loss": 2.0158,
       "step": 40
     },
     {
       "epoch": 0.67,
-      "grad_norm": 11.492398262023926,
+      "grad_norm": 13.246223449707031,
       "learning_rate": 1.865671641791045e-05,
-      "loss": 2.077,
+      "loss": 2.0417,
       "step": 50
     },
     {
       "epoch": 0.8,
-      "grad_norm": 9.018352508544922,
+      "grad_norm": 10.271443367004395,
       "learning_rate": 1.119402985074627e-05,
-      "loss": 1.8977,
+      "loss": 1.8307,
       "step": 60
     },
     {
       "epoch": 0.93,
-      "grad_norm": 14.4193115234375,
+      "grad_norm": 16.004535675048828,
       "learning_rate": 3.7313432835820893e-06,
-      "loss": 1.9343,
+      "loss": 1.9649,
       "step": 70
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5428571428571428,
-      "eval_loss": 1.724652886390686,
-      "eval_runtime": 13.0426,
-      "eval_samples_per_second": 5.367,
-      "eval_steps_per_second": 1.38,
+      "eval_accuracy": 0.35714285714285715,
+      "eval_loss": 1.7548341751098633,
+      "eval_runtime": 12.6786,
+      "eval_samples_per_second": 5.521,
+      "eval_steps_per_second": 1.42,
       "step": 75
     },
     {
       "epoch": 1.0,
       "step": 75,
       "total_flos": 3.73846047031296e+17,
-      "train_loss": 2.0760047658284506,
-      "train_runtime": 147.1173,
-      "train_samples_per_second": 2.039,
-      "train_steps_per_second": 0.51
+      "train_loss": 2.0365064112345377,
+      "train_runtime": 139.8497,
+      "train_samples_per_second": 2.145,
+      "train_steps_per_second": 0.536
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.2838709677419355,
-      "eval_loss": 1.9679653644561768,
-      "eval_runtime": 28.6695,
-      "eval_samples_per_second": 5.406,
-      "eval_steps_per_second": 1.36,
+      "eval_accuracy": 0.27741935483870966,
+      "eval_loss": 1.9401741027832031,
+      "eval_runtime": 29.4337,
+      "eval_samples_per_second": 5.266,
+      "eval_steps_per_second": 1.325,
       "step": 75
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.2838709677419355,
-      "eval_loss": 1.9679656028747559,
-      "eval_runtime": 26.6207,
-      "eval_samples_per_second": 5.823,
-      "eval_steps_per_second": 1.465,
+      "eval_accuracy": 0.27741935483870966,
+      "eval_loss": 1.9401745796203613,
+      "eval_runtime": 28.1588,
+      "eval_samples_per_second": 5.505,
+      "eval_steps_per_second": 1.385,
       "step": 75
     }
   ],