Skip to content

zzyy0929/KDD23-CauSTG

Repository files navigation

CauSTG 基于 Graph WaveNet for Deep Spatial-Temporal Graph Modeling 完成

Train Commands

  1. 基于划分数据训练k个模型
  2. 对每个模型的参数做MinPooling,得到新的模型参数
  3. 对新的模型参数做微调
sh train_env.sh

The implementation of "Maintaining the Status Qua: Capturing Invariant Relations for OOD Spatiotemporal Learning" accepted by SIGKDD conference 2023. This case is implemented on Metr-LA. Please unzip the file data.zip and pycache.zip.

About

The appendix and core code of model CauSTG, for accepted paper in KDD 2023.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published