End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.3225806451612903,
-    "eval_loss": 1.8583415746688843,
-    "eval_runtime": 28.2276,
-    "eval_samples_per_second": 5.491,
-    "eval_steps_per_second": 1.382
+    "eval_accuracy": 0.432258064516129,
+    "eval_loss": 1.879202961921692,
+    "eval_runtime": 28.2061,
+    "eval_samples_per_second": 5.495,
+    "eval_steps_per_second": 1.383
 }

runs/May24_22-11-50_2241dd2ff75a/events.out.tfevents.1716589127.2241dd2ff75a.34.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2e7c1e2c11340c2c46f902eed5c9acda18552122ab8125e2a53f85f87f5d2c5
-size 405
+oid sha256:769f0a25f3d8a6a53fe2878ced50f28c70eb6e3fa0f84e82d116b25947a9fa5a
+size 722

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.3225806451612903,
-    "eval_loss": 1.8583415746688843,
-    "eval_runtime": 28.2276,
-    "eval_samples_per_second": 5.491,
-    "eval_steps_per_second": 1.382
+    "eval_accuracy": 0.432258064516129,
+    "eval_loss": 1.879202961921692,
+    "eval_runtime": 28.2061,
+    "eval_samples_per_second": 5.495,
+    "eval_steps_per_second": 1.383
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.5714285714285714,
+  "best_metric": 0.5285714285714286,
   "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": 9.74156379699707,
+      "grad_norm": 9.097260475158691,
       "learning_rate": 4.850746268656717e-05,
-      "loss": 2.3609,
+      "loss": 2.3267,
       "step": 10
     },
     {
       "epoch": 0.27,
-      "grad_norm": 9.950693130493164,
+      "grad_norm": 9.448296546936035,
       "learning_rate": 4.104477611940299e-05,
-      "loss": 2.2224,
+      "loss": 2.1526,
       "step": 20
     },
     {
       "epoch": 0.4,
-      "grad_norm": 9.031251907348633,
+      "grad_norm": 8.596159934997559,
       "learning_rate": 3.358208955223881e-05,
-      "loss": 2.1254,
+      "loss": 2.1627,
       "step": 30
     },
     {
       "epoch": 0.53,
-      "grad_norm": 9.428235054016113,
+      "grad_norm": 9.472164154052734,
       "learning_rate": 2.6119402985074626e-05,
-      "loss": 1.8965,
+      "loss": 2.0586,
       "step": 40
     },
     {
       "epoch": 0.67,
-      "grad_norm": 16.60507583618164,
+      "grad_norm": 13.980053901672363,
       "learning_rate": 1.865671641791045e-05,
-      "loss": 1.9915,
+      "loss": 2.1303,
       "step": 50
     },
     {
       "epoch": 0.8,
-      "grad_norm": 10.332749366760254,
+      "grad_norm": 8.99145793914795,
       "learning_rate": 1.119402985074627e-05,
-      "loss": 1.7514,
+      "loss": 1.8887,
       "step": 60
     },
     {
       "epoch": 0.93,
-      "grad_norm": 13.279064178466797,
+      "grad_norm": 13.135601043701172,
       "learning_rate": 3.7313432835820893e-06,
-      "loss": 1.8974,
+      "loss": 1.841,
       "step": 70
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5714285714285714,
-      "eval_loss": 1.5608302354812622,
-      "eval_runtime": 12.4656,
-      "eval_samples_per_second": 5.615,
-      "eval_steps_per_second": 1.444,
+      "eval_accuracy": 0.5285714285714286,
+      "eval_loss": 1.7372936010360718,
+      "eval_runtime": 13.1282,
+      "eval_samples_per_second": 5.332,
+      "eval_steps_per_second": 1.371,
       "step": 75
     },
     {
       "epoch": 1.0,
       "step": 75,
       "total_flos": 3.73846047031296e+17,
-      "train_loss": 2.0024278195699057,
-      "train_runtime": 118.6782,
-      "train_samples_per_second": 2.528,
-      "train_steps_per_second": 0.632
+      "train_loss": 2.053947308858236,
+      "train_runtime": 151.5596,
+      "train_samples_per_second": 1.979,
+      "train_steps_per_second": 0.495
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.3225806451612903,
-      "eval_loss": 1.8583416938781738,
-      "eval_runtime": 27.5747,
-      "eval_samples_per_second": 5.621,
-      "eval_steps_per_second": 1.414,
+      "eval_accuracy": 0.432258064516129,
+      "eval_loss": 1.8792027235031128,
+      "eval_runtime": 29.2338,
+      "eval_samples_per_second": 5.302,
+      "eval_steps_per_second": 1.334,
       "step": 75
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.3225806451612903,
-      "eval_loss": 1.8583415746688843,
-      "eval_runtime": 28.2276,
-      "eval_samples_per_second": 5.491,
-      "eval_steps_per_second": 1.382,
+      "eval_accuracy": 0.432258064516129,
+      "eval_loss": 1.879202961921692,
+      "eval_runtime": 28.2061,
+      "eval_samples_per_second": 5.495,
+      "eval_steps_per_second": 1.383,
       "step": 75
     }
   ],