End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.23,
-    "eval_accuracy": 0.8838709677419355,
-    "eval_loss": 0.40657657384872437,
-    "eval_runtime": 51.977,
-    "eval_samples_per_second": 2.982,
-    "eval_steps_per_second": 0.385
+    "eval_accuracy": 0.7012195121951219,
+    "eval_loss": 0.8817618489265442,
+    "eval_runtime": 77.5612,
+    "eval_samples_per_second": 2.114,
+    "eval_steps_per_second": 0.271
 }

runs/Feb02_06-11-44_03b2ce95466a/events.out.tfevents.1675320044.03b2ce95466a.4627.5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f53d18313f53189268c407b85b6083f0ea190f23422079a731a633b879d2902
-size 686
+oid sha256:10d44a4ed3d8385187fa7b78e2c849ecdfc92c28fc455185c587c6f5c3363c6f
+size 1009

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.23,
-    "eval_accuracy": 0.8838709677419355,
-    "eval_loss": 0.40657657384872437,
-    "eval_runtime": 51.977,
-    "eval_samples_per_second": 2.982,
-    "eval_steps_per_second": 0.385
+    "eval_accuracy": 0.7012195121951219,
+    "eval_loss": 0.8817618489265442,
+    "eval_runtime": 77.5612,
+    "eval_samples_per_second": 2.114,
+    "eval_steps_per_second": 0.271
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.8571428571428571,
+  "best_metric": 0.8421052631578947,
   "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-148",
   "epoch": 3.22972972972973,
   "global_step": 148,
@@ -10,154 +10,163 @@
     {
       "epoch": 0.07,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 2.388,
+      "loss": 2.4536,
       "step": 10
     },
     {
       "epoch": 0.14,
       "learning_rate": 4.81203007518797e-05,
-      "loss": 2.2811,
+      "loss": 2.4176,
       "step": 20
     },
     {
       "epoch": 0.2,
       "learning_rate": 4.43609022556391e-05,
-      "loss": 2.1664,
+      "loss": 2.1549,
       "step": 30
     },
     {
       "epoch": 0.26,
-      "eval_accuracy": 0.5428571428571428,
-      "eval_loss": 1.8360090255737305,
-      "eval_runtime": 21.4908,
-      "eval_samples_per_second": 3.257,
-      "eval_steps_per_second": 0.419,
+      "eval_accuracy": 0.32894736842105265,
+      "eval_loss": 2.0454306602478027,
+      "eval_runtime": 35.7982,
+      "eval_samples_per_second": 2.123,
+      "eval_steps_per_second": 0.279,
       "step": 38
     },
     {
       "epoch": 1.01,
       "learning_rate": 4.0601503759398494e-05,
-      "loss": 2.0301,
+      "loss": 1.9566,
       "step": 40
     },
     {
       "epoch": 1.08,
       "learning_rate": 3.6842105263157895e-05,
-      "loss": 1.702,
+      "loss": 1.5598,
       "step": 50
     },
     {
       "epoch": 1.15,
       "learning_rate": 3.3082706766917295e-05,
-      "loss": 1.3257,
+      "loss": 1.3703,
       "step": 60
     },
     {
       "epoch": 1.22,
       "learning_rate": 2.9323308270676693e-05,
-      "loss": 0.9836,
+      "loss": 1.153,
       "step": 70
     },
     {
       "epoch": 1.26,
-      "eval_accuracy": 0.6142857142857143,
-      "eval_loss": 1.0020188093185425,
-      "eval_runtime": 25.7828,
-      "eval_samples_per_second": 2.715,
-      "eval_steps_per_second": 0.349,
+      "eval_accuracy": 0.5526315789473685,
+      "eval_loss": 1.4139705896377563,
+      "eval_runtime": 31.628,
+      "eval_samples_per_second": 2.403,
+      "eval_steps_per_second": 0.316,
       "step": 76
     },
     {
       "epoch": 2.03,
       "learning_rate": 2.556390977443609e-05,
-      "loss": 0.9018,
+      "loss": 0.9631,
       "step": 80
     },
     {
       "epoch": 2.09,
       "learning_rate": 2.1804511278195487e-05,
-      "loss": 0.608,
+      "loss": 0.7152,
       "step": 90
     },
     {
       "epoch": 2.16,
       "learning_rate": 1.8045112781954888e-05,
-      "loss": 0.5259,
+      "loss": 0.5672,
       "step": 100
     },
     {
       "epoch": 2.23,
       "learning_rate": 1.4285714285714285e-05,
-      "loss": 0.4682,
+      "loss": 0.6531,
       "step": 110
     },
     {
       "epoch": 2.26,
-      "eval_accuracy": 0.7714285714285715,
-      "eval_loss": 0.5422177910804749,
-      "eval_runtime": 23.0731,
-      "eval_samples_per_second": 3.034,
-      "eval_steps_per_second": 0.39,
+      "eval_accuracy": 0.6052631578947368,
+      "eval_loss": 1.0039249658584595,
+      "eval_runtime": 39.6049,
+      "eval_samples_per_second": 1.919,
+      "eval_steps_per_second": 0.252,
       "step": 114
     },
     {
       "epoch": 3.04,
       "learning_rate": 1.0526315789473684e-05,
-      "loss": 0.3221,
+      "loss": 0.4358,
       "step": 120
     },
     {
       "epoch": 3.11,
       "learning_rate": 6.766917293233083e-06,
-      "loss": 0.3005,
+      "loss": 0.4124,
       "step": 130
     },
     {
       "epoch": 3.18,
       "learning_rate": 3.007518796992481e-06,
-      "loss": 0.2704,
+      "loss": 0.442,
       "step": 140
     },
     {
       "epoch": 3.23,
-      "eval_accuracy": 0.8571428571428571,
-      "eval_loss": 0.45877233147621155,
-      "eval_runtime": 25.3687,
-      "eval_samples_per_second": 2.759,
-      "eval_steps_per_second": 0.355,
+      "eval_accuracy": 0.8421052631578947,
+      "eval_loss": 0.7881976962089539,
+      "eval_runtime": 39.3029,
+      "eval_samples_per_second": 1.934,
+      "eval_steps_per_second": 0.254,
       "step": 148
     },
     {
       "epoch": 3.23,
       "step": 148,
-      "total_flos": 1.460491890402263e+18,
-      "train_loss": 1.111738093801447,
-      "train_runtime": 1447.5739,
-      "train_samples_per_second": 0.818,
-      "train_steps_per_second": 0.102
+      "total_flos": 1.4642564654333952e+18,
+      "train_loss": 1.1917882690558563,
+      "train_runtime": 1556.2056,
+      "train_samples_per_second": 0.761,
+      "train_steps_per_second": 0.095
     },
     {
       "epoch": 3.23,
-      "eval_accuracy": 0.8838709677419355,
-      "eval_loss": 0.4065764844417572,
-      "eval_runtime": 50.3372,
-      "eval_samples_per_second": 3.079,
-      "eval_steps_per_second": 0.397,
+      "eval_accuracy": 0.7012195121951219,
+      "eval_loss": 0.8817619681358337,
+      "eval_runtime": 83.6154,
+      "eval_samples_per_second": 1.961,
+      "eval_steps_per_second": 0.251,
       "step": 148
     },
     {
       "epoch": 3.23,
-      "eval_accuracy": 0.8838709677419355,
-      "eval_loss": 0.40657657384872437,
-      "eval_runtime": 51.977,
-      "eval_samples_per_second": 2.982,
-      "eval_steps_per_second": 0.385,
+      "eval_accuracy": 0.7012195121951219,
+      "eval_loss": 0.8817618489265442,
+      "eval_runtime": 81.3017,
+      "eval_samples_per_second": 2.017,
+      "eval_steps_per_second": 0.258,
+      "step": 148
+    },
+    {
+      "epoch": 3.23,
+      "eval_accuracy": 0.7012195121951219,
+      "eval_loss": 0.8817618489265442,
+      "eval_runtime": 77.5612,
+      "eval_samples_per_second": 2.114,
+      "eval_steps_per_second": 0.271,
       "step": 148
     }
   ],
   "max_steps": 148,
   "num_train_epochs": 9223372036854775807,
-  "total_flos": 1.460491890402263e+18,
+  "total_flos": 1.4642564654333952e+18,
   "trial_name": null,
   "trial_params": null
 }