Contributing

We welcome contributions to CausalBoundingEngine!

Quick Start

  1. Fork & Clone

    Fork the repository on GitHub and clone your fork:

    git clone https://github.com/tmaringgele/CausalBoundingEngine.git
cd CausalBoundingEngine

  2. Install for Development

    pip install -e .
pip install pytest  # For running tests

  3. Make Your Changes

    • Create a new branch: git checkout -b feature-name

    • Make your changes

    • Add tests if needed

    • Run tests: pytest

  4. Submit Pull Request

    • Push your branch: git push origin feature-name

    • Open a Pull Request on GitHub

    • Describe your changes clearly

That’s it! We’ll review your PR and provide feedback.

Development Guidelines

Code Style

  • Follow existing code patterns

  • Add docstrings to new functions/classes

  • Keep changes focused and minimal

Testing

  • Add tests for new functionality

  • Ensure existing tests pass: pytest

  • Test with different scenarios when possible

Documentation

  • Update relevant documentation

  • Add examples for new features

  • Update algorithm references if adding new methods

Types of Contributions

Bug Fixes

  • Report bugs via GitHub issues

  • Include minimal reproduction example

  • Fix and submit PR with test case

New Algorithms

  • Inherit from Algorithm base class

  • Implement _compute_ATE and/or _compute_PNS

  • Add to appropriate scenario’s AVAILABLE_ALGORITHMS

  • Include proper citations and references

New Scenarios

  • Inherit from Scenario base class

  • Define AVAILABLE_ALGORITHMS mapping

  • Add to scenarios module

  • Update documentation

Documentation

  • Fix typos, improve clarity

  • Add examples and use cases

  • Update references and citations

Questions?

  • Open a GitHub issue for bugs or feature requests

  • Check existing issues before creating new ones

  • Be clear and provide examples when possible

Thanks for contributing! 🎉

Areas needing optimization & feature ideas:

  1. Killing dependencies: Get rid of dependencies (especially R/Java) or at least make everything use the same solver.

  2. Custom DAG: Causaloptim, Zaffalonbounds and Autobound are able to use a custom DAG as input, but this is not yet implemented in the Python interface.

  3. Mediation: Causaloptim, Zaffalonbounds and Autobound may be applicable to mediation scenarios.

  4. Cool New Algorithms: Implement new algorithms that are not yet supported.

  5. More Symbolic bounds: Low hanging fruit: implement more symbolic bounds (like tianpearl or manski) for other scenarios.

  6. Extract symbols from causaloptim: Currently, causaloptim does not expose the symbols used in the bounds (but rather directly evaluates them), so it is not possible to use them in the Python interface. This would be a great addition.

Getting Help

Where to Get Help

  1. GitHub Issues: Bug reports and feature requests

  2. Documentation: Check existing docs first

  3. Code Review: Ask for feedback during PR process

  4. E-Mail: For a mortal emergency, you can contact me via email at tobias.maringgele@tum.de

Communication Guidelines

  • Be respectful and constructive

  • Provide minimal reproducible examples for bugs

  • Search existing issues before creating new ones

  • Use descriptive titles and clear descriptions

  • Follow up on your contributions

Recognition

Contributors are recognized in:

  • AUTHORS.md: All contributors listed

  • CHANGELOG: Major contributions noted in releases

  • Documentation: Algorithm implementers credited in docs

  • Citations: Academic contributors cited in papers

Thank you for contributing to CausalBoundingEngine! Your efforts help make causal inference more accessible to researchers and practitioners worldwide.