Substra
Open-source federated learning software for training and validating models across distributed datasets, with a Python interface and local simulations.
Substra is open-source federated learning software for teams that need to train and validate machine learning models across distributed datasets. It supports production use across a network, as well as simulations on a single machine for testing and debugging federated learning code.
The software provides a Python interface and a web application for running federated training at scale. The Substra/substra package is the low-level Python library that connects code to a Substra network. For developers assessing it, that distinction matters: this package is the programmatic interface to the wider system.
Its use extends beyond research examples. Hospitals and biotech companies have deployed Substra, including in the MELLODDY project. That makes it relevant to teams evaluating federated learning for collaborative medical or biotech work, alongside researchers who want to simulate distributed training locally before working with a network.
Substra uses the Apache 2.0 license. Owkin originally developed it and remains its main contributor; the project is hosted by the Linux Foundation for AI and Data.
Logotyper
