Hypergraph-MLP

This is the repo for our ICASSP 2024 paper: Hypergraph-MLP: Learning on Hypergraphs without Message Passing.

Overview

A quick summary of different folders:

• 'baselines_hypergnn' contains the source code for our baseline hypergraph neural networks.

• 'ours' contains the source code for our Hypergraph-MLP.

Recommend Environment:

conda create -n "hgmlp" python=3.7
conda activate hgmlp

conda install pytorch==1.4.0 torchvision==0.5.0 cudatoolkit=10.0 -c pytorch

pip install torch-scatter==2.0.4 -f https://pytorch-geometric.com/whl/torch-1.4.0+cu100.html
pip install torch-sparse==0.6.0 -f https://pytorch-geometric.com/whl/torch-1.4.0+cu100.html
pip install torch-cluster==1.5.2 -f https://pytorch-geometric.com/whl/torch-1.4.0+cu100.html
pip install torch-geometric==1.6.3 -f https://pytorch-geometric.com/whl/torch-1.4.0+cu100.html
pip install ipdb
pip install tqdm
pip install scipy
pip install matplotlib

Data Preparation:

To generate a dataset for training using PyG or DGL, please set up the following three directories:

p2root: './data/pyg_data/hypergraph_dataset_updated/'
p2raw: './data/AllSet_all_raw_data/'
p2dgl_data: './data/dgl_data_raw/'

Next, unzip the raw data zip file into p2raw. The raw data zip file can be found in this link.

Acknowledgement

This code is based on the official code of AllSet (Paper; Github). Sincere appreciation is extended for their valuable contributions.

Citation

If you use this code, please cite our paper:

@inproceedings{tang2024hypergraph,
  title={Hypergraph-MLP: Learning on Hypergraphs without Message Passing},
  author={Tang, Bohan and Chen, Siheng and Dong, Xiaowen},
  booktitle={ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  year={2024},
  organization={IEEE}
}