This is the code for our publication "Understanding and Simplifying Architecture Search in Spatio-Temporal Graph Neural Networks" accepted by Transactions on Machine Learning Research (TMLR) 2023.

Dataset

• PeMS03/04/07/08

These four datasets are available from Spatial-Temporal Synchronous Graph Convolutional Networks: A New Framework for Spatial-Temporal Network Data Forecasting

• NE-BJ

This dataset is available from Dynamic Graph Convolutional Recurrent Network for Traffic Prediction: Benchmark and Solution

The default data path is ../../traffic/XXX as by data.py.

Environment

We use minimal environment requirement as below. Our code is tested on CUDA 11.1

numpy
pandas
torch==1.9.0

Run

cd SimpleSTG
python run.py

More configurations other than default can be found in run.py.