metadata
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: CodeBERTa-commit-message-autocomplete
results: []
CodeBERTa-commit-message-autocomplete
This model is a fine-tuned version of microsoft/codebert-base-mlm on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.7327
- Accuracy: 0.6612
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: 42
- eval_batch_size: 42
- seed: 42
- gradient_accumulation_steps: 3
- total_train_batch_size: 126
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 50
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 325 | 2.6847 | 0.5185 |
| 3.367 | 2.0 | 650 | 2.4055 | 0.5573 |
| 3.367 | 3.0 | 975 | 2.2742 | 0.5766 |
| 2.4354 | 4.0 | 1300 | 2.1065 | 0.6057 |
| 2.1925 | 5.0 | 1625 | 2.0764 | 0.6053 |
| 2.1925 | 6.0 | 1950 | 2.0169 | 0.6172 |
| 2.0217 | 7.0 | 2275 | 1.9270 | 0.6209 |
| 1.9424 | 8.0 | 2600 | 1.9326 | 0.6318 |
| 1.9424 | 9.0 | 2925 | 1.8849 | 0.6321 |
| 1.8485 | 10.0 | 3250 | 1.8834 | 0.6422 |
| 1.7847 | 11.0 | 3575 | 1.8213 | 0.6481 |
| 1.7847 | 12.0 | 3900 | 1.8674 | 0.6374 |
| 1.719 | 13.0 | 4225 | 1.7865 | 0.6473 |
| 1.6847 | 14.0 | 4550 | 1.8005 | 0.6523 |
| 1.6847 | 15.0 | 4875 | 1.8039 | 0.6516 |
| 1.6274 | 16.0 | 5200 | 1.7457 | 0.6617 |
| 1.5833 | 17.0 | 5525 | 1.7456 | 0.6526 |
| 1.5833 | 18.0 | 5850 | 1.7314 | 0.6626 |
| 1.5485 | 19.0 | 6175 | 1.7605 | 0.6590 |
| 1.5448 | 20.0 | 6500 | 1.7694 | 0.6592 |
| 1.5448 | 21.0 | 6825 | 1.7327 | 0.6612 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu117
- Datasets 2.7.1
- Tokenizers 0.13.2