update model card README.md

README.md

@@ -16,8 +16,8 @@ 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: 0.4361
-- Accuracy: 0.8968
+- Loss: 0.4287
+- Accuracy: 0.8387
 
 ## Model description
 
@@ -37,27 +37,27 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 1
-- eval_batch_size: 1
+- train_batch_size: 4
+- eval_batch_size: 4
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- training_steps: 1200
+- training_steps: 300
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.9142        | 0.25  | 300  | 1.9627          | 0.2714   |
-| 1.5385        | 1.25  | 600  | 1.0013          | 0.6857   |
-| 0.1054        | 2.25  | 900  | 0.9924          | 0.7143   |
-| 1.3117        | 3.25  | 1200 | 0.2302          | 0.9429   |
+| 1.6613        | 0.25  | 75   | 1.3506          | 0.5286   |
+| 0.4175        | 1.25  | 150  | 0.5572          | 0.7286   |
+| 0.3592        | 2.25  | 225  | 0.2464          | 0.9286   |
+| 0.1935        | 3.25  | 300  | 0.2603          | 0.9      |
 
 
 ### Framework versions
 
 - Transformers 4.24.0
-- Pytorch 1.12.1+cu113
+- Pytorch 1.8.0+cu111
 - Datasets 2.7.1
 - Tokenizers 0.13.2