Contributing

Contributions of all kinds are welcome — new statistical tests, NLP improvements, better documentation, bug fixes, and example notebooks.

---

Development Setup

# 1. Fork the repository on GitHub, then clone your fork
git clone https://github.com/YOUR_USERNAME/HypoTestX.git
cd HypoTestX

# 2. Create a virtual environment
python -m venv venv

# Linux / macOS
source venv/bin/activate

# Windows
venv\Scripts\activate

# 3. Install in editable mode with all dev dependencies
pip install -e ".[dev]"

# 4. Install pre-commit hooks (black, isort, flake8)
pre-commit install

# 5. Run the test suite to verify everything works
pytest

---

Running Tests

# Full suite
pytest

# With coverage report
pytest --cov=hypotestx --cov-report=term-missing

# Specific test file
pytest tests/test_tests/test_parametric.py

# Specific test function
pytest tests/test_tests/test_parametric.py::test_two_sample_ttest_basic

---

Contribution Areas

Area	Details
New statistical tests	Add to hypotestx/tests/ and expose through hypotestx/__init__.py
NLP / routing improvements	Extend _TESTS_BY_KEYWORD in fallback.py, or improve the LLM prompts in prompts.py
Visualizations	Add new chart types to hypotestx/explore/visualize.py
Educational content	Improve docstrings with examples, add entries to hypotestx/education/
Domain packages	Add specialist tests to hypotestx/domains/
Documentation	Update or add pages in docs/source/
Bug fixes	Check open issues on GitHub

---

Code Style

PEP 8

The project uses black (line length 88) for formatting and isort for import order. Pre-commit hooks enforce these automatically.

# Format manually
black hypotestx/
isort hypotestx/

Type hints

All public functions must have complete type annotations:

def ttest_2samp(
    group1: list[float],
    group2: list[float],
    alpha: float = 0.05,
    alternative: str = "two-sided",
    equal_var: bool = False,
) -> HypoResult:

Docstrings

Use Google-style docstrings for all public functions and classes:

def my_test(data: list[float], mu: float = 0.0) -> HypoResult:
    """
    One-line summary.

    Longer description if needed.

    Args:
        data: The sample data as a list of floats.
        mu: The null hypothesis mean. Defaults to 0.

    Returns:
        A HypoResult object.

    Example:
        >>> result = my_test([1.2, 3.4, 2.1], mu=2.0)
        >>> result.is_significant
        False
    """

Test coverage

New code must have 95 %+ test coverage. Place tests in the corresponding tests/ subdirectory:

• tests/test_tests/ — for hypotestx/tests/

• tests/test_core/ — for hypotestx/core/

• tests/test_stats/ — for hypotestx/stats/

---

Pull Request Process

1. Branch from main with a descriptive name:

   git checkout -b feat/two-way-anova
   git checkout -b fix/fallback-column-matching
   

2. Write tests first — confirm the test fails, then implement the feature.

3. Keep commits atomic — one logical change per commit.

4. Run the full test suite before opening a PR:

   pytest --cov=hypotestx
   flake8 hypotestx/
   mypy hypotestx/
   

5. Open a Pull Request against main and fill in the PR template.

6. At least one review approval is required before merging.

---

Filing a Bug Report

Please use the GitHub issue tracker and include:

• HypoTestX version (pip show hypotestx)

• Python version

• Minimal reproducible example

• Full traceback