metadata
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: CommitPredictor
results: []
CommitPredictor
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: 0.5037
- Accuracy: 0.8963
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: 2.722963804424848e-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: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.7479 | 1.0 | 2440 | 0.6350 | 0.8715 |
| 0.6629 | 2.0 | 4880 | 0.5891 | 0.8793 |
| 0.6252 | 3.0 | 7320 | 0.5720 | 0.8829 |
| 0.5904 | 4.0 | 9760 | 0.5426 | 0.8883 |
| 0.5549 | 5.0 | 12200 | 0.5284 | 0.8918 |
| 0.5456 | 6.0 | 14640 | 0.5212 | 0.8927 |
| 0.5452 | 7.0 | 17080 | 0.5206 | 0.8933 |
| 0.5196 | 8.0 | 19520 | 0.5094 | 0.8948 |
| 0.5193 | 9.0 | 21960 | 0.5078 | 0.8950 |
| 0.5193 | 10.0 | 24400 | 0.5037 | 0.8963 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu117
- Datasets 2.7.1
- Tokenizers 0.13.2