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 [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6103
-- Accuracy: 0.7891
+- Loss: 0.5538
+- Accuracy: 0.7616
 
 ## Model description
 
@@ -36,23 +36,27 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0002
+- learning_rate: 0.0003
 - train_batch_size: 64
 - eval_batch_size: 64
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- num_epochs: 4
-- mixed_precision_training: Native AMP
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 8
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| No log        | 1.0   | 151  | 0.5499          | 0.7335   |
-| No log        | 2.0   | 302  | 0.5113          | 0.7472   |
-| No log        | 3.0   | 453  | 0.5043          | 0.7584   |
-| 0.4311        | 4.0   | 604  | 0.6103          | 0.7891   |
+| No log        | 1.0   | 10   | 0.5944          | 0.7483   |
+| No log        | 2.0   | 20   | 0.5640          | 0.7483   |
+| No log        | 3.0   | 30   | 0.5582          | 0.7483   |
+| No log        | 4.0   | 40   | 0.5585          | 0.7483   |
+| No log        | 5.0   | 50   | 0.5598          | 0.7483   |
+| No log        | 6.0   | 60   | 0.5484          | 0.7483   |
+| No log        | 7.0   | 70   | 0.5524          | 0.7417   |
+| No log        | 8.0   | 80   | 0.5538          | 0.7616   |
 
 
 ### Framework versions