/TMLR-CLP

[TMLR] Simplifying Node Classification on Heterophilous Graphs with Compatible Label Propagation (https://arxiv.org/abs/2205.09389)

Primary LanguagePython

Compatible Label Propagation

Required packages

The code has been tested running under Python 3.7.3. with the following packages installed (along with their dependencies):

  • numpy == 1.16.5
  • pandas == 0.25.1
  • scikit-learn == 0.21.2
  • networkx == 2.3
  • community == 0.13
  • pytorch == 1.1.0
  • torch_geometric == 1.3.2

Data requirement

All eight datasets we used in the paper are all public datasets which can be downloaded from the internet.

Code execution

Two demo file is given to show the execution of CLP. Many optional hyper-parameters to try with, such as alpha, echo-free propagation mechanism, etc.

Cite

Please cite our paper if it is helpful in your own work:

@article{ZIP22,
title = {Simplifying Node Classification on Heterophilous Graphs with Compatible Label Propagation},
author = {Zhiqiang Zhong and Sergei Ivanov and Jun Pang},
journal = {Transactions on Machine Learning Research (TMLR)},
year = {2022},
}