Installation and Contribution


LEAF has been used under Python 3.5, and is designed to work with the TensorFlow and NumPy ecosystem. The dependency requirements are:

  • scipy(>=0.13.3)
  • tensorflow(>=1.10) (or tensorflow-gpu)
  • numpy

Additionally, you might need:

  • wget (for fetching data, if not already present)
  • matplotlib(>=2.0.0)


  • Clone the repo from GitHub:
  • Install the libraries listed in requirements.txt
    • pip3 install -r requirements.txt
  • Go to directory of respective dataset for instructions on fetching and generating data
  • models directory contains instructions on running baseline reference implementations


You can contribute to this framework by submitting pull requests to the GitHub repo