End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.36774193548387096,
-    "eval_loss": 1.830082893371582,
-    "eval_runtime": 28.5402,
-    "eval_samples_per_second": 5.431,
-    "eval_steps_per_second": 1.366
+    "eval_accuracy": 0.3225806451612903,
+    "eval_loss": 1.8583415746688843,
+    "eval_runtime": 28.2276,
+    "eval_samples_per_second": 5.491,
+    "eval_steps_per_second": 1.382
 }

runs/May24_14-33-48_4b27c78cc6c7/events.out.tfevents.1716561418.4b27c78cc6c7.35.3

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17489f31e866cf9957a2850abae7266377c6084c31b5131eab77e48295c781a1
-size 405
+oid sha256:d9c8df45eb087425df3e3bbf70b0e3473fffc16f1733808746811069574745c2
+size 722

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.36774193548387096,
-    "eval_loss": 1.830082893371582,
-    "eval_runtime": 28.5402,
-    "eval_samples_per_second": 5.431,
-    "eval_steps_per_second": 1.366
+    "eval_accuracy": 0.3225806451612903,
+    "eval_loss": 1.8583415746688843,
+    "eval_runtime": 28.2276,
+    "eval_samples_per_second": 5.491,
+    "eval_steps_per_second": 1.382
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.34285714285714286,
+  "best_metric": 0.5714285714285714,
   "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-75",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -10,105 +10,87 @@
   "log_history": [
     {
       "epoch": 0.13,
-      "grad_norm": 10.281960487365723,
+      "grad_norm": 9.74156379699707,
       "learning_rate": 4.850746268656717e-05,
-      "loss": 2.3601,
+      "loss": 2.3609,
       "step": 10
     },
     {
       "epoch": 0.27,
-      "grad_norm": 11.063933372497559,
+      "grad_norm": 9.950693130493164,
       "learning_rate": 4.104477611940299e-05,
-      "loss": 2.211,
+      "loss": 2.2224,
       "step": 20
     },
     {
       "epoch": 0.4,
-      "grad_norm": 10.219429016113281,
+      "grad_norm": 9.031251907348633,
       "learning_rate": 3.358208955223881e-05,
-      "loss": 2.0991,
+      "loss": 2.1254,
       "step": 30
     },
     {
       "epoch": 0.53,
-      "grad_norm": 10.343681335449219,
+      "grad_norm": 9.428235054016113,
       "learning_rate": 2.6119402985074626e-05,
-      "loss": 2.0743,
+      "loss": 1.8965,
       "step": 40
     },
     {
       "epoch": 0.67,
-      "grad_norm": 16.12186622619629,
+      "grad_norm": 16.60507583618164,
       "learning_rate": 1.865671641791045e-05,
-      "loss": 2.0251,
+      "loss": 1.9915,
       "step": 50
     },
     {
       "epoch": 0.8,
-      "grad_norm": 11.030004501342773,
+      "grad_norm": 10.332749366760254,
       "learning_rate": 1.119402985074627e-05,
-      "loss": 1.7871,
+      "loss": 1.7514,
       "step": 60
     },
     {
       "epoch": 0.93,
-      "grad_norm": 16.10842514038086,
+      "grad_norm": 13.279064178466797,
       "learning_rate": 3.7313432835820893e-06,
-      "loss": 1.886,
+      "loss": 1.8974,
       "step": 70
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.34285714285714286,
-      "eval_loss": 1.6963948011398315,
-      "eval_runtime": 13.6425,
-      "eval_samples_per_second": 5.131,
-      "eval_steps_per_second": 1.319,
+      "eval_accuracy": 0.5714285714285714,
+      "eval_loss": 1.5608302354812622,
+      "eval_runtime": 12.4656,
+      "eval_samples_per_second": 5.615,
+      "eval_steps_per_second": 1.444,
       "step": 75
     },
     {
       "epoch": 1.0,
       "step": 75,
       "total_flos": 3.73846047031296e+17,
-      "train_loss": 2.0343732452392578,
-      "train_runtime": 184.675,
-      "train_samples_per_second": 1.624,
-      "train_steps_per_second": 0.406
+      "train_loss": 2.0024278195699057,
+      "train_runtime": 118.6782,
+      "train_samples_per_second": 2.528,
+      "train_steps_per_second": 0.632
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.36774193548387096,
-      "eval_loss": 1.8300830125808716,
-      "eval_runtime": 29.3467,
-      "eval_samples_per_second": 5.282,
-      "eval_steps_per_second": 1.329,
+      "eval_accuracy": 0.3225806451612903,
+      "eval_loss": 1.8583416938781738,
+      "eval_runtime": 27.5747,
+      "eval_samples_per_second": 5.621,
+      "eval_steps_per_second": 1.414,
       "step": 75
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.36774193548387096,
-      "eval_loss": 1.830082893371582,
-      "eval_runtime": 27.9297,
-      "eval_samples_per_second": 5.55,
-      "eval_steps_per_second": 1.396,
-      "step": 75
-    },
-    {
-      "epoch": 1.0,
-      "eval_accuracy": 0.36774193548387096,
-      "eval_loss": 1.8300830125808716,
-      "eval_runtime": 28.0549,
-      "eval_samples_per_second": 5.525,
-      "eval_steps_per_second": 1.39,
-      "step": 75
-    },
-    {
-      "epoch": 1.0,
-      "eval_accuracy": 0.36774193548387096,
-      "eval_loss": 1.830082893371582,
-      "eval_runtime": 28.5402,
-      "eval_samples_per_second": 5.431,
-      "eval_steps_per_second": 1.366,
+      "eval_accuracy": 0.3225806451612903,
+      "eval_loss": 1.8583415746688843,
+      "eval_runtime": 28.2276,
+      "eval_samples_per_second": 5.491,
+      "eval_steps_per_second": 1.382,
       "step": 75
     }
   ],