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-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.7213
-- Accuracy: 0.4483
+- Loss: 1.0475
+- Accuracy: 0.7586
 
 ## Model description
 
@@ -38,11 +38,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 1e-05
-- train_batch_size: 8
-- eval_batch_size: 8
+- train_batch_size: 1
+- eval_batch_size: 1
 - seed: 42
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 32
+- total_train_batch_size: 4
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
@@ -52,14 +52,14 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 2.0681        | 0.93  | 10   | 1.9095          | 0.2874   |
-| 1.831         | 1.95  | 21   | 1.8602          | 0.3793   |
-| 1.7667        | 2.98  | 32   | 1.8181          | 0.4023   |
-| 1.7262        | 4.0   | 43   | 1.7910          | 0.4598   |
-| 1.8574        | 4.93  | 53   | 1.7632          | 0.4598   |
-| 1.6504        | 5.95  | 64   | 1.7332          | 0.4828   |
-| 1.6352        | 6.98  | 75   | 1.7169          | 0.4828   |
-| 1.5997        | 7.44  | 80   | 1.7213          | 0.4483   |
+| 1.8893        | 1.0   | 86   | 1.8192          | 0.4253   |
+| 1.6679        | 2.0   | 172  | 1.5983          | 0.4828   |
+| 1.4013        | 3.0   | 258  | 1.3888          | 0.6437   |
+| 1.2347        | 4.0   | 344  | 1.2665          | 0.6782   |
+| 1.0931        | 5.0   | 430  | 1.1818          | 0.6782   |
+| 0.9913        | 6.0   | 516  | 1.1167          | 0.7241   |
+| 0.9458        | 7.0   | 602  | 1.0819          | 0.7241   |
+| 0.909         | 8.0   | 688  | 1.0475          | 0.7586   |
 
 
 ### Framework versions