This project contains the Python code for the PhD thesis titled "Joint epidemic and spatio-temporal approach for modelling disease outbreaks" at Lancaster University supervised by Dr. Chris Jewell and Prof. Dr. Peter Neal.

It provides an extension to the Python probabilistic programming library PyMC3 with a sampler to effectively impute missing infection and removal time data. Furthermore, it integrates a mechanistic modeling approach into an epidemiological modeling setting.

Analysis of PHE covid19 data for weekly SPI-M reports

Repository containing the Covid19UK pipeline

Add-on classes for Tensorflow Probability provided by GEM. This is a transitional repo, intended to be incorporated into GEM eventually.

This project has been migrated to https://gitlab.com/gem-epidemics/gemlib

Note on DataOps for CHICAS

Discrete time Markov epidemic model for COVID-19. This paper describes the MCMC sampler we use to impute censored event times.

Sequential testing study for COVID-19

Making Uber H3 hexagons for polygon areas, with sample UK data.

Grabs ISO8601 date-named folders from an SFTP server.

sourceR R package to fit source attribution models