EasyFL is an easy-to-use federated learning (FL) platform based on PyTorch. It aims to enable users with various levels of expertise to experiment and prototype FL applications with little or no coding.
What is EasyFL?
EasyFL is designed to democratize federated learning, allowing users from different backgrounds to engage in:
- FL Research on algorithms and systems
- Proof-of-concept (POC) for new FL applications
- Prototyping industrial applications
- Learning FL implementations
Currently, EasyFL focuses on horizontal FL, supporting both cross-silo and cross-device FL. You can explore more about federated learning from these resources.
Major Features
Easy to Start
EasyFL is simple to install and learn. No complex dependency requirements exist, meaning you can get started on your personal computer with just three lines of code! Check out the Quick Start for more details.
Out-of-the-box Functionalities
EasyFL comes packed with numerous out-of-the-box functionalities, including:
- Datasets
- Models
- FL algorithms
With straightforward configurations, users can simulate different FL scenarios using popular datasets. EasyFL supports both statistical and system heterogeneity simulation.
Flexible, Customizable, and Reproducible
EasyFL is highly customizable, meaning you can migrate existing Computer Vision (CV) or Natural Language Processing (NLP) applications into a federated mode simply by using PyTorch codes that you are already familiar with.
Multiple Training Modes
EasyFL supports:
- Standalone training
- Distributed training
- Remote training
With a single codebase, you can accelerate FL training using distributed training on multiple GPUs or deploy it to Kubernetes with Docker for remote training.
Getting Started
For installation instructions, refer to the Get Started page. If you want to dive right in, the Quick Run guide provides the easiest way to use EasyFL.
Advanced Usage and Tutorials
For more advanced usage information, you can find tutorials on:
- High-level APIs
- Configurations
- Datasets
- Models
- Customize Server and Client
- Distributed Training
- Remote Training
Projects and Papers
The EasyFL team has released source code for several papers, including:
- Federated Multiple Task Learning: Code for MAS: Towards Resource-Efficient Federated Multiple-Task Learning (ICCV2023)
- FedSSL: Code for Divergence-aware Federated Self-Supervised Learning (ICLR2022)
We have been deeply engaged in federated learning research for several years, and you can discover even more of our publications.
Troubleshooting
Whether you’re facing installation issues or usage questions, here’s how to resolve common problems:
- Installation Fails: Ensure you have all system dependencies required by PyTorch. You can check the installation guide for detailed steps.
- Code Doesn’t Run: Verify that you have the appropriate environment set up. Review the high-level APIs tutorial for examples.
- Learning FL Concepts: If you’re new to FL, start with the tutorial on resources available on GitHub.
For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.
Conclusion
At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.