update model card README.md

README.md

@@ -21,7 +21,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.854
+      value: 0.96
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,8 +31,8 @@ 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 food101 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.6021
-- Accuracy: 0.854
+- Loss: 0.1448
+- Accuracy: 0.96
 
 ## Model description
 
@@ -59,17 +59,18 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 512
 - 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: 5
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| No log        | 1.0   | 9    | 3.5563          | 0.508    |
-| 4.1714        | 2.0   | 18   | 2.8400          | 0.842    |
-| 3.1127        | 3.0   | 27   | 2.6021          | 0.854    |
+| No log        | 1.0   | 9    | 0.5069          | 0.896    |
+| 2.1627        | 2.0   | 18   | 0.1891          | 0.946    |
+| 0.3451        | 3.0   | 27   | 0.1448          | 0.96     |
+| 0.2116        | 4.0   | 36   | 0.1509          | 0.958    |
+| 0.1711        | 5.0   | 45   | 0.1498          | 0.958    |
 
 
 ### Framework versions