Distribution Knowledge Embedding for Graph Pooling

This is the code for our paper "Distribution Knowledge Embedding for Graph Pooling". It is based on the code from SOPool. Many thanks!

Created by Kaixuan Chen (chenkx@zju.edu.cn, chenkx.jsh@aliyun.com)

Download & Citation

If you find our code useful for your research, please kindly cite our paper.

@article{chen2023distribution,
  title={Distribution Knowledge Embedding for Graph Pooling},
  author={Chen, Kaixuan and Song, Jie and Liu, Shunyu and Yu, Na and Feng, Zunlei and Han, Gengshi and Song, Mingli},
  journal={IEEE Transactions on Knowledge and Data Engineering},
  volume={35},
  number={8},
  pages={7898--7908},
  year={2023},
  publisher={IEEE}
}

System requirement

Programming language

Python 3.6

Python Packages

PyTorch > 1.0.0, tqdm, networkx, numpy

Run the code

We provide scripts to run the experiments.

For DKEPool module tested on MUTAG dataset, run

open run_mutag.sh file
set DKE_TYPE="${5-0}"

sh run_mutag.sh

For robust DKEPool module tested on MUTAG dataset, run