Model save

README.md

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.4366
-- Accuracy: 0.435
+- Loss: 1.0286
+- Accuracy: 0.865
 
 ## Model description
 
@@ -46,15 +46,16 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 3
+- num_epochs: 4
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| No log        | 0.98  | 8    | 1.5488          | 0.305    |
-| 1.6543        | 1.96  | 16   | 1.4751          | 0.31     |
-| 1.5           | 2.95  | 24   | 1.4366          | 0.435    |
+| No log        | 0.98  | 8    | 1.2950          | 0.71     |
+| 1.383         | 1.96  | 16   | 1.1531          | 0.815    |
+| 1.2243        | 2.95  | 24   | 1.0636          | 0.86     |
+| 1.1189        | 3.93  | 32   | 1.0286          | 0.865    |
 
 
 ### Framework versions