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 [microsoft/git-base](https://huggingface.co/microsoft/git-base) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 7.2329
-- Wer Score: 337.0
+- Loss: 0.0157
+- Wer Score: 3.4082
 
 ## Model description
 
@@ -38,11 +38,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 2
-- eval_batch_size: 2
+- train_batch_size: 8
+- eval_batch_size: 8
 - seed: 42
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 4
+- total_train_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 20
@@ -51,8 +51,22 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Wer Score |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|
-| No log        | 8.0   | 20   | 7.7933          | 340.0     |
-| No log        | 16.0  | 40   | 7.2329          | 337.0     |
+| No log        | 1.21  | 20   | 7.3815          | 194.2041  |
+| No log        | 2.42  | 40   | 5.4366          | 0.8878    |
+| No log        | 3.64  | 60   | 3.5512          | 0.7755    |
+| No log        | 4.85  | 80   | 1.8784          | 1.2551    |
+| No log        | 6.06  | 100  | 0.7146          | 21.0918   |
+| No log        | 7.27  | 120  | 0.2174          | 1.4184    |
+| No log        | 8.48  | 140  | 0.0743          | 12.8673   |
+| No log        | 9.7   | 160  | 0.0360          | 0.5510    |
+| No log        | 10.91 | 180  | 0.0265          | 0.4184    |
+| No log        | 12.12 | 200  | 0.0207          | 0.3878    |
+| No log        | 13.33 | 220  | 0.0185          | 0.5       |
+| No log        | 14.55 | 240  | 0.0174          | 4.3571    |
+| No log        | 15.76 | 260  | 0.0166          | 1.2653    |
+| No log        | 16.97 | 280  | 0.0158          | 4.0918    |
+| No log        | 18.18 | 300  | 0.0158          | 3.7959    |
+| No log        | 19.39 | 320  | 0.0157          | 3.4082    |
 
 
 ### Framework versions