End of training

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: 0.0441
-- Accuracy: 0.987
+- Loss: 0.0525
+- Accuracy: 0.9819
 
 ## Model description
 
@@ -38,11 +38,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0002
-- train_batch_size: 32
-- eval_batch_size: 32
+- train_batch_size: 8
+- eval_batch_size: 8
 - seed: 1337
-- gradient_accumulation_steps: 4
-- total_train_batch_size: 128
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
@@ -52,11 +52,11 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.0923        | 0.99  | 62   | 0.0608          | 0.9805   |
-| 0.0594        | 2.0   | 125  | 0.0612          | 0.978    |
-| 0.0596        | 2.99  | 187  | 0.0538          | 0.98     |
-| 0.0378        | 4.0   | 250  | 0.0440          | 0.9825   |
-| 0.0246        | 4.96  | 310  | 0.0441          | 0.987    |
+| 0.1366        | 1.0   | 100  | 0.1011          | 0.9569   |
+| 0.0648        | 2.0   | 200  | 0.0648          | 0.9788   |
+| 0.0816        | 3.0   | 300  | 0.0532          | 0.9775   |
+| 0.0351        | 4.0   | 400  | 0.0518          | 0.9794   |
+| 0.02          | 5.0   | 500  | 0.0525          | 0.9819   |
 
 
 ### Framework versions