Model save

README.md

@@ -3,6 +3,8 @@ license: apache-2.0
 base_model: google/vit-base-patch16-224
 tags:
 - generated_from_trainer
+metrics:
+- accuracy
 model-index:
 - name: vit-bird
   results: []
@@ -14,6 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 # vit-bird
 
 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: 1.6612
+- Accuracy: 0.7653
 
 ## Model description
 
@@ -32,17 +37,24 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 100
+- learning_rate: 5e-05
+- train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 1
+- num_epochs: 3
 
 ### Training results
 
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 3.3816        | 1.0   | 187  | 3.1595          | 0.5787   |
+| 1.9573        | 2.0   | 375  | 1.9335          | 0.7390   |
+| 1.3958        | 2.99  | 561  | 1.6612          | 0.7653   |
 
 
 ### Framework versions