Contributing¶

We welcome contributions to scikit-feature! This document provides guidelines for contributing.

Development setup¶

# Clone the repository
git clone https://github.com/NoRaincheck/scikit-feature.git
cd scikit-feature

# Install dependencies and the package in editable mode
uv sync --group dev
uv pip install -e .

Code style¶

We use Ruff for linting and formatting and ty for type checking:

# Format code
make format

# Check linting
make lint

# Type-check code
make typecheck

Testing¶

Tests are located in the tests/ directory, one file per algorithm. Run them with pytest:

make test
# or
uv run pytest tests/

Adding a new algorithm¶

Create your algorithm file under the appropriate category in skfeature/function/<category>/
Follow the existing pattern for the scoring function signature so it stays compatible with sklearn.feature_selection.SelectKBest
Add comprehensive docstrings with inputs, outputs, and references
Write unit tests in the corresponding test file under tests/
Update this documentation (see below)

Documentation¶

The documentation is built with Material for MkDocs. Navigation lives in mkdocs.yml and the pages live under docs/.

# Install the docs dependencies
uv sync --group docs

# Serve the docs locally with live reload
make serve-docs
# open http://localhost:8000

# Build the static site
make build-docs

When adding an algorithm:

Add a reference page under docs/api/<category>/
Link it in the category table in docs/api/algorithms-overview.md
Register it in the nav section of mkdocs.yml

Pull request process¶

Fork the repository
Create a feature branch (git checkout -b feature/my-feature)
Make your changes with tests and documentation
Run linting, formatting, type checking, and tests (make format && make lint && make typecheck && make test)
Commit your changes with clear messages
Push to your fork and open a Pull Request

License¶

By contributing, you agree that your contributions will be licensed under the GNU General Public License v2.0.