update model card README.md

README.md

@@ -17,9 +17,9 @@ 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: 0.0242
-- Precision: 0.9935
-- Recall: 0.9934
+- Loss: 0.0189
+- Precision: 0.9974
+- Recall: 0.9974
 
 ## Model description
 
@@ -45,15 +45,22 @@ 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_steps: 100
-- num_epochs: 3
+- num_epochs: 10
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|
-| 0.1176        | 1.0   | 190  | 0.0644          | 0.9836    | 0.9828 |
-| 0.0515        | 2.0   | 380  | 0.0281          | 0.9934    | 0.9934 |
-| 0.0189        | 3.0   | 570  | 0.0242          | 0.9935    | 0.9934 |
+| 0.1023        | 1.0   | 190  | 0.0528          | 0.9829    | 0.9828 |
+| 0.058         | 2.0   | 380  | 0.0375          | 0.9896    | 0.9894 |
+| 0.038         | 3.0   | 570  | 0.0359          | 0.9909    | 0.9907 |
+| 0.0283        | 4.0   | 760  | 0.0467          | 0.9861    | 0.9854 |
+| 0.0326        | 5.0   | 950  | 0.0222          | 0.9934    | 0.9934 |
+| 0.0257        | 6.0   | 1140 | 0.0249          | 0.9934    | 0.9934 |
+| 0.0259        | 7.0   | 1330 | 0.0296          | 0.9921    | 0.9921 |
+| 0.0171        | 8.0   | 1520 | 0.0239          | 0.9948    | 0.9947 |
+| 0.0211        | 9.0   | 1710 | 0.0190          | 0.9974    | 0.9974 |
+| 0.0054        | 10.0  | 1900 | 0.0189          | 0.9974    | 0.9974 |
 
 
 ### Framework versions