View on GitHub

General contributions guidelines

We always welcome contributions to Py-Feat and recommend you follow these basic steps to do. We highly recommend using Visual Studio Code as we include recommended editor extensions and settings in this repo. We use pytest for testing and ruff for linting and formatting:

1. Fork the repository on GitHub.
2. Install Feat on your machine, by git clone your fork
3. Install the development dependencies which will also install the package dependencies: pip install -r requirements-dev.txt
4. Install Py-Feat in development mode so that any changes you make to source files are automatically reflected in scripts/notebooks: pip install -e .
5. Add you code contributions and/or changes
6. Create or update the appropriate tests in feat/tests/.
7. Run a single test to make sure your new functionality works: pytest -k 'name_of_your_test'
8. Alternatively (or additionally) run the full Py-Feat test suite: pytest
9. Add any applicable licenses to LICENSE.txt
10. When your tests pass create a pull-request against the master/main branch on github!

Tutorial contribution guidelines

Py-Feat's docs are built with marimo-book (Material for MkDocs under the hood). To add a new tutorial or page: 1. Add a marimo notebook (.py) or a Markdown file under docs/ — tutorials live in docs/basic_tutorials/, prose pages in docs/pages/. 2. Add an entry for your file to the table of contents in docs/book.yml. 3. Run cd docs && marimo-book build to render the site (output in docs/_site/).

For executable tutorials marked mode: cached, run marimo-book render on a machine with the models/data, then commit the refreshed docs/_rendered/ so CI can build the page without executing it.

Note

Our documentation building pipeline does not execute jupyter notebooks. It just renders their input and output as pages. So make sure you locally execute cells that you want output for before committing your changes

For instructions on how to add new detectors see here