/ReGENN

Recurrent Graph Evolution Neural Network (ReGENN) using Graph Soft Evolution (GSE)

Primary LanguagePythonGNU Affero General Public License v3.0AGPL-3.0

ReGENN for Panel Time-Series Forecasting using GSE

This repository is no longer maintained.

Acronyms

  • Recurrent Graph Evolution Neural Network (ReGENN)
  • Graph Soft Evolution (GSE)

Requirements

Please clone @gabrielspadon's Virtual Environment by running the following command:

conda env create -f py37.yaml

Instructions

For detailed information, please run:

python main.py --help

Citation

Please cite the following paper:

@Article{Spadon2022:ReGENN,
    author  = {Spadon, Gabriel and Hong, Shenda and Brandoli, Bruno and Matwin, Stan and Rodrigues-Jr, Jose F. and Sun, Jimeng},
    journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence}, 
    title   = {Pay Attention to Evolution: Time Series Forecasting With Deep Graph-Evolution Learning}, 
    year    = {2022},
    volume  = {44},
    number  = {9},
    pages   = {5368-5384},
    doi     = {10.1109/TPAMI.2021.3076155}}
}