This repository contains code for Bayesian calibration of the stochastic extension of the Harris Wilson ordinary differential equations. The differential equations are defined based on a singly or doubly constrained spatial interaction model of urban transport. The code was based on this repository and was created for the purposes of my MRes thesis at the University of Cambridge. My thesis can be found on my website.
Before running any python script please run chmod +x ./setup.sh and then ./setup.sh. This bash script installs python requirements, sets up necessary directories etc.
This repository is a work in progress. The following tasks have to be completed:
- Make doubly constrained spatial interaction model class compatible to the singly constrained model class.
- Complete
generate_figures.shscript. - Improve comments on model classes and argsparse descriptions in inference classes.
- Write tests for all classes.
- Release outputs in nice format.
More updates soon!
Give the repository a star and fork it if you like it!