End of training

README.md

@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9771
+- Loss: 1.0015
 - Accuracy: 0.6176
 
 ## Model description

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.2,
-    "eval_accuracy": 0.47058823529411764,
-    "eval_loss": 1.3654829263687134,
-    "eval_runtime": 226.6589,
-    "eval_samples_per_second": 0.075,
+    "epoch": 12.04,
+    "eval_accuracy": 0.6176470588235294,
+    "eval_loss": 1.001463770866394,
+    "eval_runtime": 216.1108,
+    "eval_samples_per_second": 0.079,
     "eval_steps_per_second": 0.009
 }

test_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.2,
-    "eval_accuracy": 0.47058823529411764,
-    "eval_loss": 1.3654829263687134,
-    "eval_runtime": 226.6589,
-    "eval_samples_per_second": 0.075,
+    "epoch": 12.04,
+    "eval_accuracy": 0.6176470588235294,
+    "eval_loss": 1.001463770866394,
+    "eval_runtime": 216.1108,
+    "eval_samples_per_second": 0.079,
     "eval_steps_per_second": 0.009
 }

trainer_state.json

@@ -1,103 +1,193 @@
 {
-  "best_metric": 0.5652173913043478,
-  "best_model_checkpoint": "finetuned-Accident-SingleLabel-Final/checkpoint-16",
-  "epoch": 4.2,
+  "best_metric": 0.6521739130434783,
+  "best_model_checkpoint": "finetuned-Accident-SingleLabel-Final/checkpoint-32",
+  "epoch": 12.04,
   "eval_steps": 500,
-  "global_step": 20,
+  "global_step": 50,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.2,
-      "eval_accuracy": 0.391304347826087,
-      "eval_loss": 1.7558271884918213,
-      "eval_runtime": 181.0738,
+      "epoch": 0.08,
+      "eval_accuracy": 0.13043478260869565,
+      "eval_loss": 1.7643578052520752,
+      "eval_runtime": 180.6913,
       "eval_samples_per_second": 0.066,
       "eval_steps_per_second": 0.006,
       "step": 4
     },
     {
-      "epoch": 1.2,
-      "eval_accuracy": 0.5217391304347826,
-      "eval_loss": 1.6333752870559692,
-      "eval_runtime": 175.4835,
+      "epoch": 1.08,
+      "eval_accuracy": 0.4782608695652174,
+      "eval_loss": 1.644978404045105,
+      "eval_runtime": 176.6998,
       "eval_samples_per_second": 0.068,
       "eval_steps_per_second": 0.006,
       "step": 8
     },
     {
-      "epoch": 2.1,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 1.5371,
+      "epoch": 2.04,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 1.6076,
       "step": 10
     },
     {
-      "epoch": 2.2,
-      "eval_accuracy": 0.5217391304347826,
-      "eval_loss": 1.4526761770248413,
-      "eval_runtime": 175.7603,
+      "epoch": 2.08,
+      "eval_accuracy": 0.5652173913043478,
+      "eval_loss": 1.420976996421814,
+      "eval_runtime": 175.9453,
       "eval_samples_per_second": 0.068,
       "eval_steps_per_second": 0.006,
       "step": 12
     },
     {
-      "epoch": 3.2,
-      "eval_accuracy": 0.5652173913043478,
-      "eval_loss": 1.291401982307434,
-      "eval_runtime": 178.1903,
-      "eval_samples_per_second": 0.067,
+      "epoch": 3.08,
+      "eval_accuracy": 0.6086956521739131,
+      "eval_loss": 1.1925344467163086,
+      "eval_runtime": 176.0087,
+      "eval_samples_per_second": 0.068,
       "eval_steps_per_second": 0.006,
       "step": 16
     },
     {
-      "epoch": 4.2,
-      "learning_rate": 0.0,
-      "loss": 1.1439,
+      "epoch": 4.08,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 1.0244,
       "step": 20
     },
     {
-      "epoch": 4.2,
-      "eval_accuracy": 0.5652173913043478,
-      "eval_loss": 1.2895885705947876,
-      "eval_runtime": 175.7621,
+      "epoch": 4.08,
+      "eval_accuracy": 0.6086956521739131,
+      "eval_loss": 1.1086814403533936,
+      "eval_runtime": 175.4334,
       "eval_samples_per_second": 0.068,
       "eval_steps_per_second": 0.006,
       "step": 20
     },
     {
-      "epoch": 4.2,
-      "step": 20,
-      "total_flos": 3.115272515753083e+17,
-      "train_loss": 1.3405299186706543,
-      "train_runtime": 3828.671,
-      "train_samples_per_second": 0.084,
+      "epoch": 5.08,
+      "eval_accuracy": 0.5652173913043478,
+      "eval_loss": 0.9823909401893616,
+      "eval_runtime": 150.4209,
+      "eval_samples_per_second": 0.08,
+      "eval_steps_per_second": 0.007,
+      "step": 24
+    },
+    {
+      "epoch": 6.08,
+      "eval_accuracy": 0.5217391304347826,
+      "eval_loss": 1.0296951532363892,
+      "eval_runtime": 150.2481,
+      "eval_samples_per_second": 0.08,
+      "eval_steps_per_second": 0.007,
+      "step": 28
+    },
+    {
+      "epoch": 7.04,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 0.9684,
+      "step": 30
+    },
+    {
+      "epoch": 7.08,
+      "eval_accuracy": 0.6521739130434783,
+      "eval_loss": 1.0348024368286133,
+      "eval_runtime": 149.9139,
+      "eval_samples_per_second": 0.08,
+      "eval_steps_per_second": 0.007,
+      "step": 32
+    },
+    {
+      "epoch": 8.08,
+      "eval_accuracy": 0.6521739130434783,
+      "eval_loss": 0.9425884485244751,
+      "eval_runtime": 149.1356,
+      "eval_samples_per_second": 0.08,
+      "eval_steps_per_second": 0.007,
+      "step": 36
+    },
+    {
+      "epoch": 9.08,
+      "learning_rate": 1.1111111111111112e-05,
+      "loss": 0.7826,
+      "step": 40
+    },
+    {
+      "epoch": 9.08,
+      "eval_accuracy": 0.6086956521739131,
+      "eval_loss": 1.0070533752441406,
+      "eval_runtime": 148.8072,
+      "eval_samples_per_second": 0.081,
+      "eval_steps_per_second": 0.007,
+      "step": 40
+    },
+    {
+      "epoch": 10.08,
+      "eval_accuracy": 0.6086956521739131,
+      "eval_loss": 0.9810559749603271,
+      "eval_runtime": 151.9451,
+      "eval_samples_per_second": 0.079,
+      "eval_steps_per_second": 0.007,
+      "step": 44
+    },
+    {
+      "epoch": 11.08,
+      "eval_accuracy": 0.6086956521739131,
+      "eval_loss": 0.9039665460586548,
+      "eval_runtime": 153.6996,
+      "eval_samples_per_second": 0.078,
+      "eval_steps_per_second": 0.007,
+      "step": 48
+    },
+    {
+      "epoch": 12.04,
+      "learning_rate": 0.0,
+      "loss": 0.7829,
+      "step": 50
+    },
+    {
+      "epoch": 12.04,
+      "eval_accuracy": 0.6086956521739131,
+      "eval_loss": 0.8987236022949219,
+      "eval_runtime": 154.4435,
+      "eval_samples_per_second": 0.078,
+      "eval_steps_per_second": 0.006,
+      "step": 50
+    },
+    {
+      "epoch": 12.04,
+      "step": 50,
+      "total_flos": 7.875408929100923e+17,
+      "train_loss": 1.0331808185577394,
+      "train_runtime": 9247.5358,
+      "train_samples_per_second": 0.087,
       "train_steps_per_second": 0.005
     },
     {
-      "epoch": 4.2,
-      "eval_accuracy": 0.47058823529411764,
-      "eval_loss": 1.3361997604370117,
-      "eval_runtime": 250.3731,
-      "eval_samples_per_second": 0.068,
-      "eval_steps_per_second": 0.008,
-      "step": 20
+      "epoch": 12.04,
+      "eval_accuracy": 0.6176470588235294,
+      "eval_loss": 0.9771482944488525,
+      "eval_runtime": 215.9095,
+      "eval_samples_per_second": 0.079,
+      "eval_steps_per_second": 0.009,
+      "step": 50
     },
     {
-      "epoch": 4.2,
-      "eval_accuracy": 0.47058823529411764,
-      "eval_loss": 1.3654829263687134,
-      "eval_runtime": 226.6589,
-      "eval_samples_per_second": 0.075,
+      "epoch": 12.04,
+      "eval_accuracy": 0.6176470588235294,
+      "eval_loss": 1.001463770866394,
+      "eval_runtime": 216.1108,
+      "eval_samples_per_second": 0.079,
       "eval_steps_per_second": 0.009,
-      "step": 20
+      "step": 50
     }
   ],
   "logging_steps": 10,
-  "max_steps": 20,
+  "max_steps": 50,
   "num_train_epochs": 9223372036854775807,
   "save_steps": 500,
-  "total_flos": 3.115272515753083e+17,
+  "total_flos": 7.875408929100923e+17,
   "trial_name": null,
   "trial_params": null
 }