End of training

README.md

@@ -16,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [huggingface/CodeBERTa-small-v1](https://huggingface.co/huggingface/CodeBERTa-small-v1) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3131
-- Accuracy: 0.9207
-- Best Accuracy: 0.9207
+- Loss: 0.1651
+- Accuracy: 0.9439
+- Best Accuracy: 0.9439
 
 ## Model description
 
@@ -37,23 +37,24 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2.34314e-05
+- learning_rate: 1.238e-05
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.05
-- training_steps: 813
+- training_steps: 915
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | Best Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|
-| 0.4448        | 0.17  | 163  | 0.2735          | 0.9066   | 0.9066        |
-| 0.2779        | 0.33  | 326  | 0.2817          | 0.9185   | 0.9185        |
-| 0.1733        | 0.5   | 489  | 0.3446          | 0.9027   | 0.9185        |
-| 0.1861        | 0.66  | 652  | 0.3131          | 0.9207   | 0.9207        |
+| 0.4914        | 0.19  | 183  | 0.2747          | 0.8956   | 0.8956        |
+| 0.2639        | 0.37  | 366  | 0.3623          | 0.8925   | 0.8956        |
+| 0.2105        | 0.56  | 549  | 0.2257          | 0.9224   | 0.9224        |
+| 0.1669        | 0.74  | 732  | 0.1651          | 0.9439   | 0.9439        |
+| 0.1037        | 0.93  | 915  | 0.1676          | 0.9408   | 0.9439        |
 
 
 ### Framework versions