pyGSL: A Graph Structure Learning Toolkit

TL;DR

pyGSL is a comprehensive Python toolkit that enables efficient, scalable, and flexible graph structure learning with diverse datasets and a unified interface, facilitating rapid development and comparison of state-of-the-art models.

Contribution

It introduces a GPU-friendly, unified framework for graph structure learning models in PyTorch, supporting large-scale tasks and easy integration into larger systems.

Findings

Efficient GPU-optimized implementations for large networks

Unified interface for various graph learning algorithms

Diverse datasets enable consistent model evaluation

Abstract

We introduce pyGSL, a Python library that provides efficient implementations of state-of-the-art graph structure learning models along with diverse datasets to evaluate them on. The implementations are written in GPU-friendly ways, allowing one to scale to much larger network tasks. A common interface is introduced for algorithm unrolling methods, unifying implementations of recent state-of-the-art techniques and allowing new methods to be quickly developed by avoiding the need to rebuild the underlying unrolling infrastructure. Implementations of differentiable graph structure learning models are written in PyTorch, allowing us to leverage the rich software ecosystem that exists e.g., around logging, hyperparameter search, and GPU-communication. This also makes it easy to incorporate these models as components in larger gradient based learning systems where differentiable estimates of graph structure may be useful, e.g. in latent graph learning. Diverse datasets and performance metrics allow consistent comparisons across models in this fast growing field. The full code repository can be found on https://github.com/maxwass/pyGSL.