/DistributedGCN

Primary LanguageJupyter Notebook

Distributed Training of Graph Convolutional Networks

This is a basic example of the training procedure used in the paper:

S. Scardapane; I. Spinelli, P. Di Lorenzo Distributed Training of Graph Convolutional Networks, IEEE Transactions on Signal and Information Processing over Networks, 2021.

Installation

To install all required dependencies run:

pip install -r requirements.txt

Run

The "self-contained" notebook shows the distributed training for the task of node classification on CoRa.

Cite

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

@ARTICLE{scardapane2021dgcn,
  author={Scardapane, Simone and Spinelli, Indro and Di Lorenzo, Paolo},
  journal={IEEE Transactions on Signal and Information Processing over Networks}, 
  title={Distributed Training of Graph Convolutional Networks}, 
  year={2021},
  volume={7},
  number={},
  pages={87-100},
  doi={10.1109/TSIPN.2020.3046237}}