Substra documentation

Substra is an open source federated learning framework. It provides a flexible python interface and a web app to run federated learning training at scale.

Substra main usage is a production one. It has already been deployed and used by hospitals and biotech companies (see the MELLODDY project for instance). Yet Substra can also be used on a single machine on a virtually splitted dataset for two use cases:

• debugging code before launching experiments on a real network

• performing FL simulations

Substra was created by Owkin and is now hosted by the Linux Foundation for AI and Data.

Interfaces

Substra has three user interfaces:

• Substra: a low-level Python library (also called SDK). Substra is used to create datasets, algorithms and machine learning tasks on the platform.

• Substrafl: a high-level federated learning Python library based on Substra. Substrafl is used to run complex federated learning experiments at scale.

• a web application used to monitor training and explore their results.

Installation

Install substra and substrafl with the following command:

$ pip install substrafl

Substra is a dependency of Substrafl, so it will be automatically installed.

Links

Some quick links:

• MNIST federated learning example

• Substrafl overview

• Compatibility table

• How to deploy Substra for Site Reliability Engineers

• Community