FedEasy : Federated learning with ease
Majid Kundroo, Ghani Haider, Nguyen Lu Dang Khoa, Abdul Wahab Mamond, Tae‐Hong Kim
Abstract
Federated learning (FL) has emerged as a promising paradigm for training machine learning models on distributed data while preserving privacy and adhering to regulations. However, existing FL frameworks often require extensive code modifications, creating challenges for researchers. In this paper, we introduce FedEasy , a user-friendly and scalable FL framework built on Flower and PyTorch. FedEasy adopts a configuration-based approach, enabling seamless customization of datasets, data distributions, and FL algorithms through a editable configuration file. It supports multi-node simulation, integrated results logging, and high scalability. FedEasy lowers the entry barrier for FL experimentation while addressing key limitations of existing frameworks.