Skip to main content

PyRossGeo is a python library for spatially resolved numerical simulation of infectious diseases

Project description


PyRossGeo: spatially resolved infectious disease models in Python Binder CI PyPI Python Version

About | Model | Installation | Documentation | License | Contact


PyRossGeo is a numerical library for spatially resolved mathematical modelling of infectious diseases. The library has a Python interface, but is coded in C using Cython. See below for more details on the model.

PyRoss is a companion library that offers tools for both deterministic and stochastic simulation of macroscopic compartmental models, as well as a complete suite of inference and optimal control tools.

Please open an issue, or join our public slack, if you have any queries, in preference to e-mailing us. For urgent enquiries, please contact Lukas Kikuchi at

The authors are part of the Rapid Assistance in Modelling the Pandemic (RAMP) taskforce at the University of Cambridge. In alphabetical order, we are: Jakub Dolezal, Tim Ekeh, Lukas Kikuchi, Hideki Kobayashi, Paul Rohrbach, Rajesh Singh and Fernando Caballero.


The PyRossGeo uses a spatially resolved infectious disease model. The model is distinct from other network-SIR models in that it explicitly considers movement between geographical nodes, by modelling the commuting patterns of the population.

Locally at each geographical node, we simulate compartmental epidemiological dynamics with an age-contact structure. The resident population at each node can move between nodes via the commuter network. The epidemics of the commute itself is modelled using the "commuterverse": People moving between geographical nodes must spend the requisite amount of time (corresponding to the distance travelled) with their fellow commuters in a commuter node. See the figure below for an example of a commuter network:

The local infective dynamics at a node-level is customizable, and any variant of the compartmental epidemiological models (e.g. SIR, SEIR, SEAIR, etc.) can be coded using a configuration file.

The model has been tested with synthetic data on London, at an MSOA (Middle Super Output Area) level. We used ~1000 geographical nodes, with a commuter network of ~300'000 edges, constructed using the 2011 UK Census data. We are currently developing a test for a UK-wide simulation at an LAD (Local Authority District) level.

For a more detailed description of the model, please read this.


You can take PyRossGeo for a spin without installation: Binder. Please be patient while Binder loads.

From a checkout of this repository

Install PyRossGeo and required dependencies using

>> git clone
>> cd PyRossGeo
>> pip install -r requirements.txt
>> python install

Via Anaconda

Install PyRossGeo and its dependencies in a pyrossgeo environment

>> git clone
>> cd PyRossGeo
>> make env
>> conda activate pyrossgeo
>> make

Via pip

Install the latest PyPI version

>> pip install pyrossgeo

PyRossGeo requires the following software


See here for documentation, tutorials and example notebooks.


We believe that openness and sharing improves the practice of science and increases the reach of its benefits. This code is released under the MIT license. Our choice is guided by the excellent article on Licensing for the scientist-programmer.


For inquiries about PyRossGeo, please join the #pyrossgeo channel of our public slack here.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pyrossgeo-1.0.8.tar.gz (1.0 MB view hashes)

Uploaded source

Built Distribution

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page