Model save

README.md

@@ -18,8 +18,8 @@ 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: 7.6141
-- Accuracy: 0.0010
+- Loss: 1.2462
+- Accuracy: 0.4286
 
 ## Model description
 
@@ -47,14 +47,14 @@ The following hyperparameters were used during training:
 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- training_steps: 1786
+- training_steps: 7
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 30.6603       | 1.0   | 1786 | 7.6141          | 0.0010   |
+| 5.9518        | 1.0   | 7    | 1.2462          | 0.4286   |
 
 
 ### Framework versions

trainer_state.json

@@ -1,41 +1,41 @@
 {
-  "best_metric": 0.0010214504596527069,
-  "best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/10 fold timesformer/VideoMAE_wlasl__codeCheck/checkpoint-1786",
+  "best_metric": 0.42857142857142855,
+  "best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/10 fold timesformer/VideoMAE_wlasl__codeCheck/checkpoint-7",
   "epoch": 1.0,
   "eval_steps": 500,
-  "global_step": 1786,
+  "global_step": 7,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 20.611459732055664,
-      "learning_rate": 1.2445550715619166e-07,
-      "loss": 30.6603,
-      "step": 1786
+      "grad_norm": 56.47215270996094,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 5.9518,
+      "step": 7
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.0010214504596527069,
-      "eval_loss": 7.61414909362793,
-      "eval_runtime": 297.8226,
-      "eval_samples_per_second": 13.149,
-      "eval_steps_per_second": 6.574,
-      "step": 1786
+      "eval_accuracy": 0.42857142857142855,
+      "eval_loss": 1.2461984157562256,
+      "eval_runtime": 2.4468,
+      "eval_samples_per_second": 5.722,
+      "eval_steps_per_second": 2.861,
+      "step": 7
     },
     {
       "epoch": 1.0,
-      "step": 1786,
-      "total_flos": 1.8121006360874189e+19,
-      "train_loss": 30.660261233202686,
-      "train_runtime": 1787.4428,
-      "train_samples_per_second": 7.994,
-      "train_steps_per_second": 0.999
+      "step": 7,
+      "total_flos": 6.978086161927373e+16,
+      "train_loss": 5.951799665178571,
+      "train_runtime": 25.0388,
+      "train_samples_per_second": 2.237,
+      "train_steps_per_second": 0.28
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1786,
+  "max_steps": 7,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 9223372036854775807,
   "save_steps": 500,
@@ -60,7 +60,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.8121006360874189e+19,
+  "total_flos": 6.978086161927373e+16,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null