Skip to content

Replication of Chai et al. 2018 paper: Bike Flow Prediction with Multi-Graph Convolutional Networks

Notifications You must be signed in to change notification settings

callumHub/BikeGraphLSTMReplication

Repository files navigation

Working Version of Chai et al. Bike flow prediction with GNNs

https://github.com/Di-Chai/GraphCNN-Bike/ had missing data for many years. This project has recovered suitable data to replicate their experiment. Recovery process documented at https://github.com/Di-Chai/GraphCNN-Bike/issues

TO RUN

  1. Download the recovered data from: https://zenodo.org/records/11039616
  2. preprocess all of the data following directions in the PrepareData Directory. (This has already been done)
  3. Next, change what model script you will run in Main.py
  4. After setting up a python venv. Set the number of jobs in Main.py to your CPU count.
  5. python Main.py (5.) You'll want to run this in the background:
nohup python Main.py >> output_file.txt 2>&1 &

About

Replication of Chai et al. 2018 paper: Bike Flow Prediction with Multi-Graph Convolutional Networks

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages