carrey7 / stag-gcn

Spatiotemporal Adaptive Gated Graph Convolution Network for Urban Traffic Flow Forecasting

STAG-GCN

Spatiotemporal Adaptive Gated Graph Convolution Network for Urban Traffic Flow Forecasting [paper]

Requirements

• pytorch >= 1.4.0
• numpy >= 1.18.1
• scikit-learn >= 0.21.0
• pytorch geometric >= 1.4.1
• pyaml
• scipy
• tqdm

Data

The data in paper can be download here: GAIA Open Dataset

Graph Construction

Run the following command to generate semantic neighbor adjacency matrix.

# Achieve DTW distance matrix
python tools/DTW_embedding.py
# Set threshold to generate semantic neighbor adjacency matrix
python tools/DTW_matrix_analysis.py

Model Training & Testing

# Training process
python train.py --config_filename='config.yaml'
# Testing process
python test.py --config_filename='config.yaml'

Citation

If you find this repository, e.g., the paper, code and the datasets, useful in your research, please cite the following paper:

@inproceedings{DBLP:conf/cikm/LuGJFZ20,
  author    = {Bin Lu and
               Xiaoying Gan and
               Haiming Jin and
               Luoyi Fu and
               Haisong Zhang},
  title     = {Spatiotemporal Adaptive Gated Graph Convolution Network for Urban
               Traffic Flow Forecasting},
  booktitle = {{CIKM} '20: The 29th {ACM} International Conference on Information
               and Knowledge Management, Virtual Event, Ireland, October 19-23, 2020},
  pages     = {1025--1034},
  publisher = {{ACM}},
  year      = {2020}
}