Flower a Agreeable Federated Studying Framework
A unified approach to federated studying, analytics, and evaluation.
Scalability
Flower became constructed to enable proper-world methods with a sizable quantity of purchasers. Researchers aged Flower to urge workloads with tens of thousands and thousands of purchasers.
ML Framework Agnostic
Flower is cherish minded with most reward and future machine studying frameworks. You cherish Keras? Colossal. You plan shut PyTorch? Awesome. Raw NumPy, no automated differentiation? You rock!
Cloud, Cell, Edge & Beyond
Flower permits research on every form of servers and gadgets, collectively with mobile. AWS, GCP, Azure, Android, iOS, Raspberry Pi, Nvidia Jetson, all cherish minded with Flower.
Study to Manufacturing
Flower permits solutions to open as research tasks and then step by step pass in direction of production deployment with low engineering effort and confirmed infrastucture.
Platform Honest
Flower is interoperable with a vary of operating methods and hardware platforms to work smartly in heterogeneous edge instrument environments.
Usability
Or not it’s easy to fetch started. 20 lines of Python is ample to fetch a elephantine federated studying diagram. Take a look at the code examples to fetch started alongside with your favourite framework.
Installation E-book
The Flower documentation has detailed instructions on what it be well-known to set up Flower and the kind you set up it. Spoiler alert: you just need pip! Take a look at out our installation info.
PyTorch, TensorFlow, 🤗, …?
Develop you make employ of PyTorch, TensorFlow, scikit-learn, MXNet, or Hugging Face? Then simply apply our quickstart examples that can enable you to federate your reward ML tasks.
