Skip to content

A-Good-System-for-Smart-Cities/KiTE-utils

Repository files navigation

KiTE CI Documentation Test .github/workflows/update-docs.ymlStreamlit App

Summary

This package is a tool that validates and calibrates supervised classification models against bias. We hope to empower users to audit models and develop diagnostic plots that help identify and quantify bias in supervised ML models.

Let's get Started!

Type this into your Python environment!

pip install git+https://github.com/A-Good-System-for-Smart-Cities/KiTE-utils.git

We welcome feedback!

Please submit any feedback, questions, or issues in the Issues Tab of this Repository. One of our team members will promptly respond to help you out!

Contributing Summary

  • To format code, run black .
  • To lint code, run flake8 .
  • To install requirements, run pip install -r requirements.txt
  • To run tests, run pytest -s KiTE

Contributing

  1. Fork this Repo
  2. Clone the Repo onto your computer -- You may need to setup an SSH Key on your device.
  • Run pip install -r requirements.txt to get all the packages you need.
  1. Create a branch (git checkout -b new-feature)
  2. Make Changes
  3. Run necessary quality assurance tools
  1. Add your changes (git commit -am "Commit Message" or git add <whatever files/folders you want to add> followed by git commit -m "Commit Message")
  2. Push your changes to the repo (git push origin new-feature)
  3. Create a pull request

You can locally build the package with pip install -e . and run unit tests with pytest -s KiTE.


Code Quality Tools

black formatter automatically formats code

  1. Run pip install black to get the package.
  2. After making changes, run black ./.

flake8 lints code

Notifies you if any code is not currently up to Python standards.

  1. Run pip install flake8 to get the package.
  2. After making changes, run flake8.

pytest runs unit tests

  1. Run pip install pytest to get the package.
  2. After making changes, you can run pytest -s KiTE to run all unit tests

Update Documentation

We use pdoc to create KiTE's documentation.

  1. pip install pdoc into your working environment
  2. Preview edits with pdoc -t doc_template --docformat numpy KiTE.
    • This will locally host the documentation site (that will update as you make edits)
  3. Edit documentation by adding/modifying docstrings in KiTE