End of training

README.md

@@ -17,8 +17,8 @@ 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: 0.2787
-- Accuracy: 0.9286
+- Loss: 0.5973
+- Accuracy: 0.8286
 
 ## Model description
 
@@ -45,21 +45,19 @@ The following hyperparameters were used during training:
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
 - training_steps: 1200
-- mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.9197        | 0.25  | 300  | 1.8284          | 0.3286   |
-| 0.5231        | 1.25  | 600  | 0.5825          | 0.7286   |
-| 0.1344        | 2.25  | 900  | 0.5880          | 0.8143   |
-| 0.5295        | 3.25  | 1200 | 0.2787          | 0.9286   |
+| 1.8559        | 0.25  | 300  | 0.9768          | 0.6286   |
+| 1.3879        | 1.25  | 600  | 1.3276          | 0.5857   |
+| 0.3694        | 2.25  | 900  | 1.2342          | 0.7      |
+| 1.0305        | 3.25  | 1200 | 0.5973          | 0.8286   |
 
 
 ### Framework versions
 
-- Transformers 4.40.1
-- Pytorch 2.0.1+cu117
-- Datasets 2.19.1
+- Transformers 4.40.2
+- Pytorch 2.2.1+cu121
 - Tokenizers 0.19.1