Awesome
Graph WaveNet for Deep Spatial-Temporal Graph Modeling
This is the original pytorch implementation of Graph WaveNet in the following paper: [Graph WaveNet for Deep Spatial-Temporal Graph Modeling, IJCAI 2019] (https://arxiv.org/abs/1906.00121). A nice improvement over GraphWavenet is presented by Shleifer et al. paper code.
<p align="center"> <img width="350" height="400" src=./fig/model.png> </p>Requirements
- python 3
- see
requirements.txt
Data Preparation
Step1: Download METR-LA and PEMS-BAY data from Google Drive or Baidu Yun links provided by DCRNN.
Step2: Process raw data
# Create data directories
mkdir -p data/{METR-LA,PEMS-BAY}
# METR-LA
python generate_training_data.py --output_dir=data/METR-LA --traffic_df_filename=data/metr-la.h5
# PEMS-BAY
python generate_training_data.py --output_dir=data/PEMS-BAY --traffic_df_filename=data/pems-bay.h5
Train Commands
python train.py --gcn_bool --adjtype doubletransition --addaptadj --randomadj