Contributing¶

Thank you for your interest in contributing. QuPrep is a focused tool — contributions that keep it simple, correct, and composable are most welcome.

Before you start¶

Check open issues to avoid duplicate work.
For large changes, open an issue first to discuss the approach.
QuPrep's scope is intentionally narrow. Features outside the ingest → clean → reduce → normalize → encode → export pipeline are unlikely to be accepted.

Install uv:

curl -LsSf https://astral.sh/uv/install.sh | sh

Clone and install:

git clone https://github.com/quprep/quprep.git
cd quprep
uv sync --extra dev

Run tests:

uv run pytest

Lint:

uv run ruff check .

Keep PRs focused — one feature or fix per PR.
Add or update tests for any changed behaviour.
Update CHANGELOG.md under [Unreleased].
Docstrings for encoders and normalizers should include the mathematical formulation.
Target the main branch.

All tests must pass (uv run pytest).
New encoders must include property-based tests with hypothesis (e.g. amplitude encoder must always produce unit-norm output).
Target ≥ 90% coverage for new modules.

Python ≥ 3.10, type hints required.
ruff for linting (configured in pyproject.toml).
No external dependencies beyond core (numpy, scipy, pandas, scikit-learn). Framework packages go in optional extras only.

Use the bug report template. Include a minimal reproducible example.

Use GitHub Discussions. Issues are for bugs and feature requests only.

This project follows the Contributor Covenant.