yangbing668 / ST3DNet

Deep Spatial–Temporal 3D Convolutional Neural Networks for Traffic Data Forecasting, TITS 2019

ST3DNet

Deep Spatial–Temporal 3D Convolutional Neural Networks for Traffic Data Forecasting

Reference

@article{guo2019deep,
  title={Deep Spatial-Temporal 3D Convolutional Neural Networks for Traffic Data Forecasting},
  author={Guo, Shengnan and Lin, Youfang and Li, Shijie and Chen, Zhaoming and Wan, Huaiyu},
  journal={IEEE Transactions on Intelligent Transportation Systems},
  year={2019},
  publisher={IEEE}
}

Datasets

BikeNYC is one of the datasets we used in the paper, it suffices to reproduce the results what we have reported in the paper.

Step 1: Download BikeNYC dataset provided by DeepST

Step 2: process dataset

python prepareData.py

Train and Test

python trainNY.py