Simple and Efficient Heterogeneous Graph Neural Network (SeHGNN)

The camera-ready paper for AAAI 23 can be found at: http://arxiv.org/abs/2207.02547

Requirements

1. Neural network libraries for GNNs

• pytorch
• pytorch-geometric

Please check your cuda version first and install the above libraries matching your cuda. If possible, we recommend to install the latest versions of these libraries.

• dgl

If you want to generate ComplEx embeddings for ogbn-mag, we recommend to install dgl<1.0.

2. Other dependencies

Install other requirements:

pip install -r requirements.txt
git clone https://github.com/Yangxc13/sparse_tools.git --depth=1
cd sparse_tools
python setup.py develop
cd ..

Data preparation

• HGB datasets for node classification

These datasets include four medium-scale datasets. Please download them (DBLP.zip, ACM.zip, IMDB.zip, Freebase.zip) from HGB repository and extract content under the folder './data/'.

• Ogbn-mag

It is a large dataset from OGB challenge. Thus dataset will be automatically downloaded for the first time running.

If you encounter any issues, please feel free to reach out to me at yangxc96@gmail.com. The previous email address, yangxiaocheng@ict.ac.cn, is no longer in use.