- This repository is the Pytorch implmentation of the STGCN that is designed for a graph timeseries forecasting.
- In fact, all codes in this repo are nothing but a clone code of the Keras tutorial — Arash Khodadadi(2021) - "Traffic forecasting using graph neural networks and LSTM".
- The only difference is that I coded it using the Pytorch and I modified some train options...
# DOWNLOAD_DATASET
wget https://github.com/VeritasYin/STGCN_IJCAI-18/raw/master/dataset/PeMSD7_Full.zip- The dataset used in this repo is the "PeMSD7" which was collected from Caltrans Performance Measurement System (PeMS) in real-time by over 39, 000 sensor stations, deployed across the major metropolitan areas of California state highway system.
- If you want to know a theoretical concept, I recommend this paper and GitHub repository.