Flower by Flower Labs is an open-source federated learning framework designed to enable distributed machine learning across multiple devices or servers without sharing raw data. It provides a unified approach to federated learning, analytics, and evaluation, allowing users to federate any workload, any ML framework, and any programming language. Flower supports various machine learning libraries including PyTorch, TensorFlow, Hugging Face, JAX, Pandas, fastai, PyTorch Lightning, MXNet, scikit-learn, and XGBoost. The framework enables researchers and developers to bring existing machine learning workloads into a federated setting, providing a flexible platform for implementing federated learning systems where multiple parties collaboratively train machine learning models on decentralized data.
