update model card README.md

README.md

@@ -21,7 +21,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7841318907779495
+      value: 0.7302889760970389
 ---
 
 <!-- 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 [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4549
-- Accuracy: 0.7841
+- Loss: 0.5574
+- Accuracy: 0.7303
 
 ## Model description
 
@@ -52,23 +52,50 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 128
-- eval_batch_size: 128
+- train_batch_size: 256
+- eval_batch_size: 256
 - seed: 42
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 512
+- total_train_batch_size: 1024
 - 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: 30
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.585         | 0.99  | 34   | 0.5135          | 0.7553   |
-| 0.529         | 1.99  | 68   | 0.4736          | 0.7687   |
-| 0.4975        | 2.98  | 102  | 0.4549          | 0.7841   |
+| 0.6271        | 0.99  | 98   | 0.6035          | 0.6926   |
+| 0.6156        | 1.99  | 197  | 0.5844          | 0.7006   |
+| 0.6148        | 3.0   | 296  | 0.5758          | 0.7104   |
+| 0.6055        | 4.0   | 395  | 0.5853          | 0.7015   |
+| 0.5938        | 4.99  | 493  | 0.5858          | 0.7104   |
+| 0.5878        | 5.99  | 592  | 0.5630          | 0.7210   |
+| 0.5873        | 7.0   | 691  | 0.5620          | 0.7236   |
+| 0.5947        | 8.0   | 790  | 0.5670          | 0.7196   |
+| 0.5866        | 8.99  | 888  | 0.5592          | 0.7265   |
+| 0.5807        | 9.99  | 987  | 0.5574          | 0.7254   |
+| 0.5764        | 11.0  | 1086 | 0.5655          | 0.7245   |
+| 0.5729        | 12.0  | 1185 | 0.5611          | 0.7237   |
+| 0.577         | 12.99 | 1283 | 0.5702          | 0.7189   |
+| 0.5702        | 13.99 | 1382 | 0.5588          | 0.7259   |
+| 0.5717        | 15.0  | 1481 | 0.5565          | 0.7244   |
+| 0.5646        | 16.0  | 1580 | 0.5536          | 0.7303   |
+| 0.5591        | 16.99 | 1678 | 0.5525          | 0.7345   |
+| 0.5586        | 17.99 | 1777 | 0.5565          | 0.7286   |
+| 0.5668        | 19.0  | 1876 | 0.5520          | 0.7304   |
+| 0.5617        | 20.0  | 1975 | 0.5557          | 0.7289   |
+| 0.5546        | 20.99 | 2073 | 0.5561          | 0.7325   |
+| 0.5579        | 21.99 | 2172 | 0.5537          | 0.7314   |
+| 0.5604        | 23.0  | 2271 | 0.5545          | 0.7290   |
+| 0.5563        | 24.0  | 2370 | 0.5591          | 0.7288   |
+| 0.5634        | 24.99 | 2468 | 0.5546          | 0.7307   |
+| 0.5563        | 25.99 | 2567 | 0.5557          | 0.7303   |
+| 0.5563        | 27.0  | 2666 | 0.5571          | 0.7276   |
+| 0.5544        | 28.0  | 2765 | 0.5551          | 0.7298   |
+| 0.5491        | 28.99 | 2863 | 0.5596          | 0.7282   |
+| 0.5461        | 29.77 | 2940 | 0.5574          | 0.7303   |
 
 
 ### Framework versions