shap-enhanced 0.0.1a2

SHAP-style explainability with enhanced methods for sequential and tabular models

SHAP-Enhanced: Advanced Explainability Toolkit

Overview

SHAP-Enhanced is a research-focused Python library providing a unified, extensible platform for developing, benchmarking, and analyzing advanced SHAP (SHapley Additive exPlanations) variants for tabular, sequential, and sparse data.

• Implements state-of-the-art and experimental SHAP explainers with a clean, consistent interface.
• Supports time series, LSTM, attention models, sparse (binary/one-hot) data, hierarchical and multi-baseline explainers, and more.
• Built for scientific benchmarking, model debugging, and real-world explainability studies.

This framework is created following PEP8 standards, Sphinx documentation format, and is released under the GNU Affero General Public License v3.0.

Installation

You can install the package locally:

git clone https://github.com/niyangbai/enhanced_shap.git
cd shap-enhanced
pip install -r requirements.txt

Contribution

We welcome contributions to SHAP-Enhanced! To contribute, please follow these steps:

1. Fork the Repository: Click the "Fork" button on the GitHub repository page.
2. Clone Your Fork: Clone your forked repository to your local machine:

   git clone https://github.com/niyangbai/enhanced_shap.git
   cd enhanced_shap
   

3. Create a Branch: Create a new branch for your feature or bug fix:

   git checkout -b feature-or-bugfix-name
   

4. Make Changes: Implement your changes, ensuring they follow PEP8 standards and include proper Sphinx-style docstrings.
5. Write Tests: Add or update unit tests in the tests/ directory to cover your changes.
6. Run Tests: Ensure all tests pass before submitting your changes:

   pytest
   

7. Commit Changes: Commit your changes with a descriptive message:

   git add .
   git commit -m "Description of your changes"
   

8. Push Changes: Push your branch to your forked repository:

   git push origin feature-or-bugfix-name
   

9. Submit a Pull Request: Open a pull request to the main repository, describing your changes in detail.

General Guidelines:

• Follow the PEP8 coding style.
• Document all public methods, classes, and modules using Sphinx-style docstrings.
• Ensure your code is properly typed and includes type hints.
• Write clear and concise commit messages.
• Be respectful and collaborative in code reviews.

License

This project is licensed under the GNU Affero General Public License v3.0 (AGPL-3.0).
See the LICENSE file for details.

Note: Any modified versions must also be made publicly available under the same license if deployed.

About

• Master Thesis — Enhanced SHAP for Sequential and Sparse Data in Predictive Process Monitoring
• University: Friedrich-Alexander-Universität Erlangen-Nürnberg

Contact

For inquiries, please reach out via GitHub Issues.