End of training

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.32,
-    "eval_accuracy": 0.8194444444444444,
-    "eval_loss": 0.5837830901145935,
-    "eval_runtime": 17.7379,
-    "eval_samples_per_second": 12.177,
-    "eval_steps_per_second": 1.522
+    "epoch": 3.23,
+    "eval_accuracy": 0.8698224852071006,
+    "eval_loss": 0.4558912515640259,
+    "eval_runtime": 14.3413,
+    "eval_samples_per_second": 11.784,
+    "eval_steps_per_second": 1.534
 }

runs/Nov10_07-39-57_8182e899e964/events.out.tfevents.1668066300.8182e899e964.742.2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d9caa80b716bb30108de9924a7b389441dbafcca07dab84d216baab312d7c028
-size 363
+oid sha256:2c056c17d5713f6b5fe12c256efc63c372ef10b41440213cacd822a3b2b9bb75
+size 686

test_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.32,
-    "eval_accuracy": 0.8194444444444444,
-    "eval_loss": 0.5837830901145935,
-    "eval_runtime": 17.7379,
-    "eval_samples_per_second": 12.177,
-    "eval_steps_per_second": 1.522
+    "epoch": 3.23,
+    "eval_accuracy": 0.8698224852071006,
+    "eval_loss": 0.4558912515640259,
+    "eval_runtime": 14.3413,
+    "eval_samples_per_second": 11.784,
+    "eval_steps_per_second": 1.534
 }

trainer_state.json

@@ -1,136 +1,163 @@
 {
-  "best_metric": 0.8247863247863247,
-  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-111",
-  "epoch": 2.315315315315315,
-  "global_step": 111,
+  "best_metric": 0.8243243243243243,
+  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-114",
+  "epoch": 3.22972972972973,
+  "global_step": 148,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.09,
-      "learning_rate": 4.166666666666667e-05,
-      "loss": 2.3105,
+      "epoch": 0.07,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 2.3253,
       "step": 10
     },
     {
-      "epoch": 0.18,
-      "learning_rate": 4.595959595959596e-05,
-      "loss": 2.173,
+      "epoch": 0.14,
+      "learning_rate": 4.81203007518797e-05,
+      "loss": 2.2018,
       "step": 20
     },
     {
-      "epoch": 0.27,
-      "learning_rate": 4.0909090909090915e-05,
-      "loss": 2.1044,
+      "epoch": 0.2,
+      "learning_rate": 4.43609022556391e-05,
+      "loss": 2.1321,
       "step": 30
     },
     {
-      "epoch": 0.34,
-      "eval_accuracy": 0.49145299145299143,
-      "eval_loss": 1.7608355283737183,
-      "eval_runtime": 20.0376,
-      "eval_samples_per_second": 11.678,
-      "eval_steps_per_second": 1.497,
+      "epoch": 0.26,
+      "eval_accuracy": 0.5,
+      "eval_loss": 1.7063186168670654,
+      "eval_runtime": 6.3814,
+      "eval_samples_per_second": 11.596,
+      "eval_steps_per_second": 1.567,
       "step": 38
     },
     {
-      "epoch": 1.02,
-      "learning_rate": 3.5858585858585855e-05,
-      "loss": 1.9043,
+      "epoch": 1.01,
+      "learning_rate": 4.0601503759398494e-05,
+      "loss": 1.8954,
       "step": 40
     },
     {
-      "epoch": 1.11,
-      "learning_rate": 3.080808080808081e-05,
-      "loss": 1.4396,
+      "epoch": 1.08,
+      "learning_rate": 3.6842105263157895e-05,
+      "loss": 1.4623,
       "step": 50
     },
     {
-      "epoch": 1.2,
-      "learning_rate": 2.575757575757576e-05,
-      "loss": 1.0538,
+      "epoch": 1.15,
+      "learning_rate": 3.3082706766917295e-05,
+      "loss": 1.1013,
       "step": 60
     },
     {
-      "epoch": 1.29,
-      "learning_rate": 2.070707070707071e-05,
-      "loss": 0.875,
+      "epoch": 1.22,
+      "learning_rate": 2.9323308270676693e-05,
+      "loss": 0.7513,
       "step": 70
     },
     {
-      "epoch": 1.34,
-      "eval_accuracy": 0.7350427350427351,
-      "eval_loss": 0.7819657325744629,
-      "eval_runtime": 20.4305,
-      "eval_samples_per_second": 11.453,
-      "eval_steps_per_second": 1.468,
+      "epoch": 1.26,
+      "eval_accuracy": 0.6756756756756757,
+      "eval_loss": 0.8875435590744019,
+      "eval_runtime": 6.6832,
+      "eval_samples_per_second": 11.073,
+      "eval_steps_per_second": 1.496,
       "step": 76
     },
     {
-      "epoch": 2.04,
-      "learning_rate": 1.565656565656566e-05,
-      "loss": 0.8563,
+      "epoch": 2.03,
+      "learning_rate": 2.556390977443609e-05,
+      "loss": 0.7555,
       "step": 80
     },
     {
-      "epoch": 2.13,
-      "learning_rate": 1.0606060606060607e-05,
-      "loss": 0.6119,
+      "epoch": 2.09,
+      "learning_rate": 2.1804511278195487e-05,
+      "loss": 0.531,
       "step": 90
     },
     {
-      "epoch": 2.22,
-      "learning_rate": 5.555555555555556e-06,
-      "loss": 0.5474,
+      "epoch": 2.16,
+      "learning_rate": 1.8045112781954888e-05,
+      "loss": 0.5464,
       "step": 100
     },
     {
-      "epoch": 2.31,
-      "learning_rate": 5.050505050505052e-07,
-      "loss": 0.5103,
+      "epoch": 2.23,
+      "learning_rate": 1.4285714285714285e-05,
+      "loss": 0.3882,
       "step": 110
     },
     {
-      "epoch": 2.32,
-      "eval_accuracy": 0.8247863247863247,
-      "eval_loss": 0.6440751552581787,
-      "eval_runtime": 20.0847,
-      "eval_samples_per_second": 11.651,
-      "eval_steps_per_second": 1.494,
-      "step": 111
-    },
-    {
-      "epoch": 2.32,
-      "step": 111,
-      "total_flos": 1.0966150712918016e+18,
-      "train_loss": 1.2986480625900063,
-      "train_runtime": 215.193,
-      "train_samples_per_second": 4.127,
-      "train_steps_per_second": 0.516
-    },
-    {
-      "epoch": 2.32,
-      "eval_accuracy": 0.8194444444444444,
-      "eval_loss": 0.5837831497192383,
-      "eval_runtime": 17.7976,
-      "eval_samples_per_second": 12.136,
-      "eval_steps_per_second": 1.517,
-      "step": 111
-    },
-    {
-      "epoch": 2.32,
-      "eval_accuracy": 0.8194444444444444,
-      "eval_loss": 0.5837830901145935,
-      "eval_runtime": 17.7379,
-      "eval_samples_per_second": 12.177,
-      "eval_steps_per_second": 1.522,
-      "step": 111
+      "epoch": 2.26,
+      "eval_accuracy": 0.8243243243243243,
+      "eval_loss": 0.46954473853111267,
+      "eval_runtime": 6.7034,
+      "eval_samples_per_second": 11.039,
+      "eval_steps_per_second": 1.492,
+      "step": 114
+    },
+    {
+      "epoch": 3.04,
+      "learning_rate": 1.0526315789473684e-05,
+      "loss": 0.271,
+      "step": 120
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 6.766917293233083e-06,
+      "loss": 0.2625,
+      "step": 130
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 3.007518796992481e-06,
+      "loss": 0.2671,
+      "step": 140
+    },
+    {
+      "epoch": 3.23,
+      "eval_accuracy": 0.8243243243243243,
+      "eval_loss": 0.3919770121574402,
+      "eval_runtime": 6.5967,
+      "eval_samples_per_second": 11.218,
+      "eval_steps_per_second": 1.516,
+      "step": 148
+    },
+    {
+      "epoch": 3.23,
+      "step": 148,
+      "total_flos": 1.460491890402263e+18,
+      "train_loss": 1.0175014428190283,
+      "train_runtime": 239.3713,
+      "train_samples_per_second": 4.946,
+      "train_steps_per_second": 0.618
+    },
+    {
+      "epoch": 3.23,
+      "eval_accuracy": 0.8698224852071006,
+      "eval_loss": 0.4558912515640259,
+      "eval_runtime": 15.206,
+      "eval_samples_per_second": 11.114,
+      "eval_steps_per_second": 1.447,
+      "step": 148
+    },
+    {
+      "epoch": 3.23,
+      "eval_accuracy": 0.8698224852071006,
+      "eval_loss": 0.4558912515640259,
+      "eval_runtime": 14.3413,
+      "eval_samples_per_second": 11.784,
+      "eval_steps_per_second": 1.534,
+      "step": 148
     }
   ],
-  "max_steps": 111,
+  "max_steps": 148,
   "num_train_epochs": 9223372036854775807,
-  "total_flos": 1.0966150712918016e+18,
+  "total_flos": 1.460491890402263e+18,
   "trial_name": null,
   "trial_params": null
 }