Skip to content

The official implementation of Convergent Graph Solvers (CGS)

License

Notifications You must be signed in to change notification settings

Junyoungpark/CGS

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Convergent Graph Solvers

The official code repository of Convergent Graph Solvers (CGS)

Paper

For more details, please see our paper Convergent Graph Solvers which has been accepted at ICLR 2022. If this code is useful for your work, please consider to cite our paper:

@inproceedings{
    park2022convergent,
    title={Convergent Graph Solvers},
    author={Junyoung Park and Jinhyun Choo and Jinkyoo Park},
    booktitle={International Conference on Learning Representations},
    year={2022},
    url={https://openreview.net/forum?id=ItkxLQU01lD}
}

Requirements

  • pytorch
  • dgl 0.7.0
  • openpnm 2.8.2
  • hydra 1.1

Quick start

1 .To train CGS for porous network problems (Section 6.1)

python pn_train.py
  1. To train CGS for graph value iteration problems (Section 6.2)
python gvi_train.py
  1. To train CGS for graph benchmark problems (Section 6.3)
python benchmark_train.py

About

The official implementation of Convergent Graph Solvers (CGS)

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages