Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
125 lines (86 loc) · 4.02 KB

CONTRIBUTING.md

File metadata and controls

125 lines (86 loc) · 4.02 KB
Raw

Contributing to EliminationSearchCV

Thank you for taking the time to contribute! 🎉
This is an early-stage project and contributions of any size are welcome — from a typo fix to a new feature.

Table of Contents

  1. Getting Started
  2. Reporting Bugs
  3. Suggesting Features
  4. Development Setup
  5. Branch & Commit Conventions
  6. Pull Request Checklist
  7. Running Benchmarks
  8. Code Style

Getting Started

  1. Star the repo ⭐ — it helps the project grow and signals to others that it's worth checking out.
  2. Read the CHANGELOG to understand what has been done and what is planned.
  3. Check open Issues before starting new work — your idea may already be tracked.
  4. Open an issue before starting significant work (new features, refactors) to align on design and avoid duplicated effort.

Reporting Bugs

Use the Bug Report issue template on GitHub. Please include:

  • A minimal, reproducible example (MRE)
  • Python version, OS, and scikit-learn version (pip show scikit-learn)
  • Expected vs actual behavior
  • Full traceback if applicable

Suggesting Features

Use the Feature Request issue template. Describe:

  • What problem you are trying to solve
  • How you expect the API to look (pseudocode is fine)
  • Whether you are willing to implement it yourself

Development Setup

# 1. Fork and clone your fork
git clone https://github.com/<your-username>/elimination-search-cv.git
cd elimination-search-cv

# 2. Create a virtual environment
python -m venv .venv
source .venv/bin/activate        # Windows: .venv\Scripts\activate

# 3. Install in editable mode
pip install -e .

# 4. Install benchmark dependencies (optional)
pip install scikit-learn numpy pandas

Branch & Commit Conventions

Type Branch prefix Example
New feature feat/ feat/n-jobs-parallel
Bug fix fix/ fix/single-param-grid-crash
Documentation docs/ docs/update-readme-benchmarks
Refactor refactor/ refactor/scoring-utils
Benchmark bench/ bench/add-svm-model

Commit messages should be short and imperative, e.g.:

  • feat: add best_score_ and best_estimator_ attributes
  • fix: handle cv=1 edge case in create_cv_data_sets
  • docs: update benchmark table with v0.0.1 results

Pull Request Checklist

Before opening a PR, make sure:

  • Your branch is up to date with main
  • The code is clean and follows existing style (see Code Style)
  • Docstrings are updated for any changed/added functions or classes
  • If you added a public attribute or parameter, the README Result Attributes / Constructor Parameters table is updated
  • If this is a bug fix, a note is included in CHANGELOG.md
  • If this is a significant new feature, a note is included in CHANGELOG.md

Running Benchmarks

# Quick configurable benchmark (recommended for development)
python benchmarks/benchmark_fast.py

# Full benchmark (5 models × 3 datasets × Light+Full grids)
python benchmarks/benchmark.py

Benchmark results are saved to benchmarks/marks/. If your change affects search speed or score quality, please include benchmark output in your PR description.

Code Style

  • Python 3.8+ compatible — avoid f-strings with = (3.8 syntax), walrus operator, and other 3.9+ features unless adding a version guard.
  • Type hints on all public functions and __init__ parameters.
  • Docstrings in Google-style for all public classes and functions.
  • Comments — prefer block comments above non-obvious logic over inline comments.
  • No external dependencies beyond scikit-learn and numpy in the core src/ package.

By contributing, you agree that your contributions will be licensed under the MIT License.