Welcome to RayFed documentation!#

RayFed is a multiple parties joint, distributed execution engine based on Ray, to help build your own federated learning frameworks in minutes.

RayFed has the following highlight features:

Ray Native Programming Pattern
Multiple Controller Execution Mode
Very Restricted and Clear Data Perimeters
Very Large Scale Federated Computing and Training

Check out the Getting Started section for further information, including how to install the project.

Note

This project is under active development.

Why RayFed#

TBD

Getting Started#

Please check Getting Started for installation and a quick start guide.

Tutorials#

Federated Learning on RayFed
Split Learning on RayFed
Seamless Federated, Transitioning from Ray to RayFed.

API#

Check out api to write your own RayFed programs.

Architecture and principle of RayFed#

To gain a comprehensive understanding of RayFed’s architecture and principles, we highly recommend reading Advanced Topic