Instructions to use dev-analyzer/commit-message-model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use dev-analyzer/commit-message-model with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="dev-analyzer/commit-message-model")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("dev-analyzer/commit-message-model") model = AutoModelForSequenceClassification.from_pretrained("dev-analyzer/commit-message-model", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| Model for classifying git commit messages, bertweet-base fine-tuned. | |
| Git commit messages are categorized into the following categories: | |
| 1. Adaptive - Adaptive activities are functional activities and involve making modifications | |
| to the software to ensure it remains compatible with new environments. Examples of these are | |
| feature additions and test cases. | |
| 2. Perfective - Perfective activities encompass modifications aimed at improving the software's | |
| overall quality, structure, and efficiency, such as refactoring, renaming, cleaning up | |
| redundant segments, and improving algorithms and performance. | |
| 3. Corrective - Corrective activities address and correct software problems such as bugs, | |
| defects, errors, and faults that negatively impact the system. | |
| 4. Administrative - Administrative activities include working on documentation such as | |
| README.md files, javadocs, or commenting the code. | |
| 5. Other - This activity includes Git operations such as merges and pull requests. It also includes | |
| vague and unspecified tasks. |