What is the PaddleFL Tool Assembly?
PaddleFL is an open-source federated learning framework developed by the PaddlePaddle team. It supports both horizontal and vertical FL strategies and leverages PaddlePaddle for distributed training and Kubernetes for deployment. PaddleFL is designed for applications in computer vision, NLP, and recommendation systems, addressing data isolation and secure data knowledge sharing.
What can you use the PaddleFL Tool Assembly for?
- Horizontal and vertical federated learning
- Distributed training with PaddlePaddle
- Secure data knowledge sharing in CV, NLP, and recommendation systems
Limitations/Remarks
- Ongoing development for vertical FL and deployment schemes
- Optimization and feature expansion in progress
Specific Use Cases/Applications
- Computer vision, NLP, and recommendation systems
- Secure, privacy-preserving data collaboration
How to access the PaddleFL Tool Assembly?
- GitHub Repository
- Install via pip or from source
References
Original Framework Contributors
- PaddlePaddle Team
- PaddleFL Community
For a full list, see the PaddleFL GitHub contributors.
More information
Training
Affiliations Contributors
Editor
KU Leuven / University of UHasselt