Traffic Congestion Propagation Prediction

Predicting Congestion Propagation Based on Traffic Prediction Algorithms

An implementation of existing DCRNN, STGCN, MW-TGC algorithm utilizing dgl and Pytorch.

Getting started

It will be upload requirements.txt file later.

Clone repo and install requirements.txt

git clone https://github.com/~~
cd traffic_forecasting
pip install -r requirements.txt

How to run the code

• preprocessing We have datasets of front, doan, dunsan, and wolpyeong.

1. ts_202107-09 file to csv file.

python ts_to_csv.py --dir tsFolderDirectory #ts folder directory

2. Make the necessary input files in STGCM, and MW-TGC.

python preprocessing.py --loc front # run front

3. Make the necessary input files in DCRNN.

python preprocessing.py --loc front # run front
			--output_dir dataset/ #Output directory

• Usage

  • DCRNN

   python train.py --loc front # run front

  • STGCN

   python main_tmap.py --loc front # run front

  • MW-TGC

   python main.py --loc front # run front