/gcn_Co-Training_Self-Training

Implementation of Graph Convolutional Networks in TensorFlow

Primary LanguagePythonMIT LicenseMIT

Introduction

This is a implementation of algorithms for semi-supervised classification mentioned in our paper:

Qimai Li, Zhichao Han, Xiao-Ming Wu, Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning (AAAI 2018) BLOG

Requirements

  • Python 3
  • tensorflow
  • networkx
  • scikit-learn

Run

python train.py

Configuration

See config.py for details. If you encounter any problem, please open an issue in this github page.

Cite

Please cite our paper if you use this code in your own work:

@inproceedings{li2018deeper,
        title = "{Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning}",
       author = {{Li}, Q. and {Han}, Z. and {Wu}, X.-M.},
    booktitle = {The Thirty-Second AAAI Conference on Artificial Intelligence},
         year = {2018},
 organization = {AAAI},
}