Skip to content

Latest commit

 

History

History
39 lines (27 loc) · 778 Bytes

README.md

File metadata and controls

39 lines (27 loc) · 778 Bytes

STGCN

This is an mxnet version implementation of Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting which has been accepted by IJCAI 2018.

Dataset

dataset comes from STGCN_IJCAI-18.

Requirements

mxnet >= 1.4.1, mxboard and scipy

or

Docker with nvidia-docker

Usage

Unzip datasets before you run the code.

cd datasets
tar -zxvf PeMSD7.tar.gz
python main.py

or use docker:

# build
cd docker
docker build -t stgcn/mxnet:1.4.1_gpu_cu100_mkl_py35 .

# run
cd ..
docker run -d -it --rm --runtime=nvidia -v $PWD:/mxnet --name stgcn stgcn/mxnet:1.4.1_gpu_cu100_mkl_py35 python3 main.py
docker logs stgcn