End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.23225806451612904,
-    "eval_loss": 1.9978936910629272,
-    "eval_runtime": 28.5135,
-    "eval_samples_per_second": 5.436,
-    "eval_steps_per_second": 1.368
+    "eval_accuracy": 0.2838709677419355,
+    "eval_loss": 1.9679656028747559,
+    "eval_runtime": 26.6207,
+    "eval_samples_per_second": 5.823,
+    "eval_steps_per_second": 1.465
 }

runs/May24_22-41-06_5ebd624037e4/events.out.tfevents.1716590680.5ebd624037e4.34.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e30da285e4c07174147330d0b90cebca00ce5400ff51cbf89ffd94b04fb9535
-size 405
+oid sha256:a6c02ea46edd6f7f9c16d14f124bac107177889c26a5ac8311360d7af4337f4c
+size 722

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.23225806451612904,
-    "eval_loss": 1.9978936910629272,
-    "eval_runtime": 28.5135,
-    "eval_samples_per_second": 5.436,
-    "eval_steps_per_second": 1.368
+    "eval_accuracy": 0.2838709677419355,
+    "eval_loss": 1.9679656028747559,
+    "eval_runtime": 26.6207,
+    "eval_samples_per_second": 5.823,
+    "eval_steps_per_second": 1.465
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.4857142857142857,
+  "best_metric": 0.5428571428571428,
   "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": 10.20085620880127,
+      "grad_norm": 11.617264747619629,
       "learning_rate": 4.850746268656717e-05,
-      "loss": 2.3596,
+      "loss": 2.377,
       "step": 10
     },
     {
       "epoch": 0.27,
-      "grad_norm": 10.088861465454102,
+      "grad_norm": 10.894355773925781,
       "learning_rate": 4.104477611940299e-05,
-      "loss": 2.2165,
+      "loss": 2.2211,
       "step": 20
     },
     {
       "epoch": 0.4,
-      "grad_norm": 9.590352058410645,
+      "grad_norm": 8.908524513244629,
       "learning_rate": 3.358208955223881e-05,
-      "loss": 2.1373,
+      "loss": 2.1478,
       "step": 30
     },
     {
       "epoch": 0.53,
-      "grad_norm": 9.82310676574707,
+      "grad_norm": 9.142258644104004,
       "learning_rate": 2.6119402985074626e-05,
-      "loss": 2.0439,
+      "loss": 2.109,
       "step": 40
     },
     {
       "epoch": 0.67,
-      "grad_norm": 14.587775230407715,
+      "grad_norm": 11.492398262023926,
       "learning_rate": 1.865671641791045e-05,
-      "loss": 1.9888,
+      "loss": 2.077,
       "step": 50
     },
     {
       "epoch": 0.8,
-      "grad_norm": 9.921350479125977,
+      "grad_norm": 9.018352508544922,
       "learning_rate": 1.119402985074627e-05,
-      "loss": 1.9229,
+      "loss": 1.8977,
       "step": 60
     },
     {
       "epoch": 0.93,
-      "grad_norm": 13.914346694946289,
+      "grad_norm": 14.4193115234375,
       "learning_rate": 3.7313432835820893e-06,
-      "loss": 1.9513,
+      "loss": 1.9343,
       "step": 70
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.4857142857142857,
-      "eval_loss": 1.798818588256836,
-      "eval_runtime": 12.7163,
-      "eval_samples_per_second": 5.505,
-      "eval_steps_per_second": 1.416,
+      "eval_accuracy": 0.5428571428571428,
+      "eval_loss": 1.724652886390686,
+      "eval_runtime": 13.0426,
+      "eval_samples_per_second": 5.367,
+      "eval_steps_per_second": 1.38,
       "step": 75
     },
     {
       "epoch": 1.0,
       "step": 75,
       "total_flos": 3.73846047031296e+17,
-      "train_loss": 2.0687105814615885,
-      "train_runtime": 114.0769,
-      "train_samples_per_second": 2.63,
-      "train_steps_per_second": 0.657
+      "train_loss": 2.0760047658284506,
+      "train_runtime": 147.1173,
+      "train_samples_per_second": 2.039,
+      "train_steps_per_second": 0.51
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.23225806451612904,
-      "eval_loss": 1.9978936910629272,
-      "eval_runtime": 28.0596,
-      "eval_samples_per_second": 5.524,
-      "eval_steps_per_second": 1.39,
+      "eval_accuracy": 0.2838709677419355,
+      "eval_loss": 1.9679653644561768,
+      "eval_runtime": 28.6695,
+      "eval_samples_per_second": 5.406,
+      "eval_steps_per_second": 1.36,
       "step": 75
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.23225806451612904,
-      "eval_loss": 1.9978936910629272,
-      "eval_runtime": 28.5135,
-      "eval_samples_per_second": 5.436,
-      "eval_steps_per_second": 1.368,
+      "eval_accuracy": 0.2838709677419355,
+      "eval_loss": 1.9679656028747559,
+      "eval_runtime": 26.6207,
+      "eval_samples_per_second": 5.823,
+      "eval_steps_per_second": 1.465,
       "step": 75
     }
   ],