End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.432258064516129,
-    "eval_loss": 1.879202961921692,
-    "eval_runtime": 28.2061,
-    "eval_samples_per_second": 5.495,
-    "eval_steps_per_second": 1.383
+    "eval_accuracy": 0.23225806451612904,
+    "eval_loss": 1.9978936910629272,
+    "eval_runtime": 28.5135,
+    "eval_samples_per_second": 5.436,
+    "eval_steps_per_second": 1.368
 }

runs/May24_22-28-39_2241dd2ff75a/events.out.tfevents.1716589901.2241dd2ff75a.34.3

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff26ffc6970a6dae145346a2227b15aa8630afbfb7a0b728a2c3176b90d04414
-size 405
+oid sha256:d3431aeb06668d26674f0463e5d55f0752f881d06011c509e90a6eb74ee39ba2
+size 722

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.432258064516129,
-    "eval_loss": 1.879202961921692,
-    "eval_runtime": 28.2061,
-    "eval_samples_per_second": 5.495,
-    "eval_steps_per_second": 1.383
+    "eval_accuracy": 0.23225806451612904,
+    "eval_loss": 1.9978936910629272,
+    "eval_runtime": 28.5135,
+    "eval_samples_per_second": 5.436,
+    "eval_steps_per_second": 1.368
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.5285714285714286,
+  "best_metric": 0.4857142857142857,
   "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": 9.097260475158691,
+      "grad_norm": 10.20085620880127,
       "learning_rate": 4.850746268656717e-05,
-      "loss": 2.3267,
+      "loss": 2.3596,
       "step": 10
     },
     {
       "epoch": 0.27,
-      "grad_norm": 9.448296546936035,
+      "grad_norm": 10.088861465454102,
       "learning_rate": 4.104477611940299e-05,
-      "loss": 2.1526,
+      "loss": 2.2165,
       "step": 20
     },
     {
       "epoch": 0.4,
-      "grad_norm": 8.596159934997559,
+      "grad_norm": 9.590352058410645,
       "learning_rate": 3.358208955223881e-05,
-      "loss": 2.1627,
+      "loss": 2.1373,
       "step": 30
     },
     {
       "epoch": 0.53,
-      "grad_norm": 9.472164154052734,
+      "grad_norm": 9.82310676574707,
       "learning_rate": 2.6119402985074626e-05,
-      "loss": 2.0586,
+      "loss": 2.0439,
       "step": 40
     },
     {
       "epoch": 0.67,
-      "grad_norm": 13.980053901672363,
+      "grad_norm": 14.587775230407715,
       "learning_rate": 1.865671641791045e-05,
-      "loss": 2.1303,
+      "loss": 1.9888,
       "step": 50
     },
     {
       "epoch": 0.8,
-      "grad_norm": 8.99145793914795,
+      "grad_norm": 9.921350479125977,
       "learning_rate": 1.119402985074627e-05,
-      "loss": 1.8887,
+      "loss": 1.9229,
       "step": 60
     },
     {
       "epoch": 0.93,
-      "grad_norm": 13.135601043701172,
+      "grad_norm": 13.914346694946289,
       "learning_rate": 3.7313432835820893e-06,
-      "loss": 1.841,
+      "loss": 1.9513,
       "step": 70
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5285714285714286,
-      "eval_loss": 1.7372936010360718,
-      "eval_runtime": 13.1282,
-      "eval_samples_per_second": 5.332,
-      "eval_steps_per_second": 1.371,
+      "eval_accuracy": 0.4857142857142857,
+      "eval_loss": 1.798818588256836,
+      "eval_runtime": 12.7163,
+      "eval_samples_per_second": 5.505,
+      "eval_steps_per_second": 1.416,
       "step": 75
     },
     {
       "epoch": 1.0,
       "step": 75,
       "total_flos": 3.73846047031296e+17,
-      "train_loss": 2.053947308858236,
-      "train_runtime": 151.5596,
-      "train_samples_per_second": 1.979,
-      "train_steps_per_second": 0.495
+      "train_loss": 2.0687105814615885,
+      "train_runtime": 114.0769,
+      "train_samples_per_second": 2.63,
+      "train_steps_per_second": 0.657
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.432258064516129,
-      "eval_loss": 1.8792027235031128,
-      "eval_runtime": 29.2338,
-      "eval_samples_per_second": 5.302,
-      "eval_steps_per_second": 1.334,
+      "eval_accuracy": 0.23225806451612904,
+      "eval_loss": 1.9978936910629272,
+      "eval_runtime": 28.0596,
+      "eval_samples_per_second": 5.524,
+      "eval_steps_per_second": 1.39,
       "step": 75
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.432258064516129,
-      "eval_loss": 1.879202961921692,
-      "eval_runtime": 28.2061,
-      "eval_samples_per_second": 5.495,
-      "eval_steps_per_second": 1.383,
+      "eval_accuracy": 0.23225806451612904,
+      "eval_loss": 1.9978936910629272,
+      "eval_runtime": 28.5135,
+      "eval_samples_per_second": 5.436,
+      "eval_steps_per_second": 1.368,
       "step": 75
     }
   ],