Spatio-Temporal Graph Convolutional Networks
About
The PyTorch version of STGCN implemented for the paper Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting.
Paper
https://arxiv.org/abs/1709.04875
Related works
- TCN: An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling
- GLU and GTU: Language Modeling with Gated Convolutional Networks
- ChebNet: Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering
- GCN: Semi-Supervised Classification with Graph Convolutional Networks
Related code
- TCN: https://github.com/locuslab/TCN
- ChebNet: https://github.com/mdeff/cnn_graph
- GCN: https://github.com/tkipf/pygcn
Dataset
Source
- METR-LA: DCRNN author's Google Drive
- PEMS-BAY: DCRNN author's Google Drive
- PeMSD7(M): STGCN author's GitHub repository
Preprocessing
Using the formula from ChebNet：
Model structure
Differents of code between mine and author's
- Fix bugs
- Add Early Stopping approach
- Add Dropout approach
- Offer a different set of hyperparameters
- Offer config files for two different categories graph convolution (ChebyGraphConv and GraphConv)
- Add datasets METR-LA and PEMS-BAY
- Adopt a different data preprocessing method
Requirements
To install requirements:
pip3 install -r requirements.txt