TorchFL: A Performant Library for Bootstrapping Federated Learning Experiments

TL;DR

TorchFL is a versatile, PyTorch-based library that simplifies setting up, executing, and profiling federated learning experiments across various hardware, making FL research more accessible and efficient.

Contribution

The paper introduces TorchFL, a comprehensive library that streamlines federated learning experimentation with customizable components and multi-hardware support.

Findings

Enables easy setup of FL experiments with minimal coding.

Supports multiple hardware accelerators for efficient training.

Provides detailed performance profiling and logging.

Abstract

With the increased legislation around data privacy, federated learning (FL) has emerged as a promising technique that allows the clients (end-user) to collaboratively train deep learning (DL) models without transferring and storing the data in a centralized, third-party server. We introduce TorchFL, a performant library for (i) bootstrapping the FL experiments, (ii) executing them using various hardware accelerators, (iii) profiling the performance, and (iv) logging the overall and agent-specific results on the go. Being built on a bottom-up design using PyTorch and Lightning, TorchFL provides ready-to-use abstractions for models, datasets, and FL algorithms, while allowing the developers to customize them as and when required. This paper aims to dig deeper into the architecture and design of TorchFL, elaborate on how it allows researchers to bootstrap the federated learning experience, and provide experiments and code snippets for the same. With the ready-to-use implementation of state-of-the-art DL models, datasets, and federated learning support, TorchFL aims to allow researchers with little to no engineering background to set up FL experiments with minimal coding and infrastructure overhead.