Setup

Based on: How to contribute (opens in a new tab)

The major difference between tech companies and open source that in open source you don’t work on the main repository directly.

• why?

  Projects like scikit-learn have hundreds of pull request monthly and an equal number of branches from many different contributors. If people were directly developing on scikit-learn, it will end up with an intractable number of branches on the main repo, and we don’t want that.

Instead, you must develop your branches on your local fork, before submitting your pull request (PR) from your fork to the main branch of the original repo:

Let’s setup your local working environment.

1. Fork the original repository

   

2. Clone the project on your local machine

   

   On your terminal:

   git clone <Your Project URL>

3. Add the original scikit-learn upstream

   git remote add upstream git@github.com:scikit-learn/scikit-learn.git

   Check that git remote -v display the following:

   origin  git@github.com:YourLogin/scikit-learn.git (fetch)
   origin  git@github.com:YourLogin/scikit-learn.git (push)
   upstream        git@github.com:scikit-learn/scikit-learn.git (fetch)
   upstream        git@github.com:scikit-learn/scikit-learn.git (push)

   It will allow you to pull the latest change from the original repo, while pushing your commits to your fork.

4. Install mamba from forge (opens in a new tab)

   Mamba is a fast front-end from conda.

   Conda is useful because it links all dependencies to the same backend, whereas pip setup adhoc backends for each library.

5. Install mamba compilers

   mamba install compilers

6. Create and activate a mamba environment

   mamba create -n sk
   mamba activate sk

7. Install scikit dependencies

   mamba install cython scipy numpy joblib

8. Install test dependencies

   mamba install pytest pytest-cov flake8 mypy numpydoc black==22.3.0

9. Install pre-commit

   pip install pre-commit
   pre-commit install

10. Build and install sklearn locally

    pip install --no-build-isolation -e . -v 

    When developing Cython files, you need to recompile them by running this command before testing

11. Check your installation

    python -c "import sklearn; sklearn.show_versions()"

If the command is successful, go to the next tutorial to start contributing!