TGM: a Modular and Efficient Library for Machine Learning on Temporal Graphs
Abstract
Well-designed open-source software drives progress in Machine Learning (ML) research. While static graph ML enjoys mature frameworks like PyTorch Geometric and DGL, ML for temporal graphs (TG), networks that evolve over time, lacks comparable infrastructure. Existing TG libraries are often tailored to specific architectures, hindering support for diverse models in this rapidly evolving field. Additionally, the divide between continuous- and discrete-time dynamic graph methods (CTDG and DTDG) limits direct comparisons and idea transfer. To address these gaps, we introduce Temporal Graph Modelling (TGM), a research-oriented library for ML on temporal graphs, the first to unify CTDG and DTDG approaches. TGM offers first-class support for dynamic node features, time-granularity conversions, and native handling of link-, node-, and graph-level tasks. Empirically, TGM achieves an average 7.8x speedup across multiple models, datasets, and tasks compared to the widely used DyGLib, and an average 175x speedup on graph discretization relative to available implementations. Beyond efficiency, we show in our experiments how TGM unlocks entirely new research possibilities by enabling dynamic graph property prediction and time-driven training paradigms, opening the door to questions previously impractical to study. TGM is available at https://github.com/tgm-team/tgm
Cite
@article{arxiv.2510.07586,
title = {TGM: a Modular and Efficient Library for Machine Learning on Temporal Graphs},
author = {Jacob Chmura and Shenyang Huang and Tran Gia Bao Ngo and Ali Parviz and Farimah Poursafaei and Jure Leskovec and Michael Bronstein and Guillaume Rabusseau and Matthias Fey and Reihaneh Rabbany},
journal= {arXiv preprint arXiv:2510.07586},
year = {2025}
}
Comments
21 pages, 5 figures, 14 tables