HGAMLP: A Scalable Training Framework for Heterogeneous Graph Neural Networks

Requirements

1. Neural network libraries for GNNs

• pytorch
• pytorch-geometric
• dgl

Data preparation

For experiments in Motivation section and on four medium-scale datasets, please download datasets DBLP.zip, ACM.zip, IMDB.zip, Freebase.zip from the source of HGB benchmark, and extract content from these compresesed files under the folder './data/'.

For experiments on the large dataset ogbn-mag, the dataset will be automatically downloaded from OGB challenge.