|
--- |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: CommitPredictor |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# CommitPredictor |
|
|
|
This model is a fine-tuned version of [microsoft/codebert-base-mlm](https://huggingface.co/microsoft/codebert-base-mlm) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.7476 |
|
- Accuracy: 0.6825 |
|
|
|
## Model description |
|
|
|
More information needed |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 2e-05 |
|
- train_batch_size: 21 |
|
- eval_batch_size: 21 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 3 |
|
- total_train_batch_size: 63 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 50 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:-----:|:---------------:|:--------:| |
|
| 2.655 | 1.0 | 599 | 2.1802 | 0.5963 | |
|
| 2.2909 | 2.0 | 1198 | 2.0666 | 0.6172 | |
|
| 2.1673 | 3.0 | 1797 | 2.0162 | 0.6197 | |
|
| 2.0743 | 4.0 | 2396 | 1.9740 | 0.6283 | |
|
| 1.9895 | 5.0 | 2995 | 1.9439 | 0.6338 | |
|
| 1.8841 | 6.0 | 3594 | 1.9338 | 0.6291 | |
|
| 1.8547 | 7.0 | 4193 | 1.7883 | 0.6559 | |
|
| 1.7867 | 8.0 | 4792 | 1.8879 | 0.6436 | |
|
| 1.7491 | 9.0 | 5391 | 1.8640 | 0.6445 | |
|
| 1.7008 | 10.0 | 5990 | 1.7935 | 0.6591 | |
|
| 1.631 | 11.0 | 6589 | 1.7864 | 0.6556 | |
|
| 1.6094 | 12.0 | 7188 | 1.7964 | 0.6541 | |
|
| 1.5755 | 13.0 | 7787 | 1.7675 | 0.6652 | |
|
| 1.5787 | 14.0 | 8386 | 1.8498 | 0.6515 | |
|
| 1.5235 | 15.0 | 8985 | 1.7363 | 0.6674 | |
|
| 1.4996 | 16.0 | 9584 | 1.7428 | 0.6641 | |
|
| 1.4571 | 17.0 | 10183 | 1.7004 | 0.6790 | |
|
| 1.4617 | 18.0 | 10782 | 1.7714 | 0.6635 | |
|
| 1.4219 | 19.0 | 11381 | 1.8232 | 0.6563 | |
|
| 1.3959 | 20.0 | 11980 | 1.7245 | 0.6752 | |
|
| 1.3801 | 21.0 | 12579 | 1.7234 | 0.6750 | |
|
| 1.3549 | 22.0 | 13178 | 1.6884 | 0.6817 | |
|
| 1.3227 | 23.0 | 13777 | 1.7566 | 0.6687 | |
|
| 1.3455 | 24.0 | 14376 | 1.7102 | 0.6745 | |
|
| 1.3239 | 25.0 | 14975 | 1.7388 | 0.6730 | |
|
| 1.3066 | 26.0 | 15574 | 1.7391 | 0.6790 | |
|
| 1.2598 | 27.0 | 16173 | 1.6754 | 0.6869 | |
|
| 1.2552 | 28.0 | 16772 | 1.6499 | 0.6798 | |
|
| 1.2431 | 29.0 | 17371 | 1.7397 | 0.6740 | |
|
| 1.2115 | 30.0 | 17970 | 1.7096 | 0.6745 | |
|
| 1.1842 | 31.0 | 18569 | 1.7159 | 0.6751 | |
|
| 1.1799 | 32.0 | 19168 | 1.7341 | 0.6788 | |
|
| 1.1755 | 33.0 | 19767 | 1.7557 | 0.6652 | |
|
| 1.1704 | 34.0 | 20366 | 1.7147 | 0.6771 | |
|
| 1.1427 | 35.0 | 20965 | 1.7631 | 0.6670 | |
|
| 1.1464 | 36.0 | 21564 | 1.7083 | 0.6750 | |
|
| 1.1179 | 37.0 | 22163 | 1.6978 | 0.6718 | |
|
| 1.1247 | 38.0 | 22762 | 1.7205 | 0.6757 | |
|
| 1.1204 | 39.0 | 23361 | 1.7403 | 0.6663 | |
|
| 1.0939 | 40.0 | 23960 | 1.6621 | 0.6852 | |
|
| 1.0904 | 41.0 | 24559 | 1.7671 | 0.6667 | |
|
| 1.0815 | 42.0 | 25158 | 1.7304 | 0.6789 | |
|
| 1.0879 | 43.0 | 25757 | 1.7346 | 0.6858 | |
|
| 1.0718 | 44.0 | 26356 | 1.7841 | 0.6691 | |
|
| 1.0599 | 45.0 | 26955 | 1.7482 | 0.6742 | |
|
| 1.0815 | 46.0 | 27554 | 1.6738 | 0.6823 | |
|
| 1.0812 | 47.0 | 28153 | 1.7573 | 0.6799 | |
|
| 1.0529 | 48.0 | 28752 | 1.6627 | 0.6849 | |
|
| 1.0675 | 49.0 | 29351 | 1.6641 | 0.6785 | |
|
| 1.0593 | 50.0 | 29950 | 1.7476 | 0.6825 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.25.1 |
|
- Pytorch 1.13.0+cu117 |
|
- Datasets 2.7.1 |
|
- Tokenizers 0.13.2 |
|
|
