Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time

victor - Python code for cross-correlation function modelling and fitting


victor is a Python package for modelling and likelihood analysis of the cross-correlation function of density regions (e.g., voids or density-split centres) with galaxies. It can be used for posterior sampling using MCMC through an interface with cobaya.



The only pre-requisites are Python (version>=3.7.4) and the Python package manager pip (version>=20.0).

Installing with pip

To install victor, do

python -m pip install git+

This will install the package with the minimal required packages (numpy, scipy, matplotlib, astropy, h5py and PyYAML). Some features of the code require additional packages:

  • camb (for more accurate calculation of the matter power spectrum in certain models)
  • cobaya (for posterior sampling via MCMC)
  • GetDist (analysis/visualisation of the results of posterior sampling)

To install with these extra dependencies, do

python -m pip install git+[all]

Here the [all] will install all extra dependencies (camb, cobaya and GetDist); replacing this with [mcmc] will install only cobaya and GetDist for MCMC analyses, and replacing it with [camb] will install only camb.

Note: To enable MPI parallelisation with cobaya (highly recommended), you may wish to separately install the package mpi4py (or use the corresponding module on your cluster). To do this, follow the instructions here before installing.

Installing in development mode

If you want edit the code and to collaborate on further development of victor the best way to do this is using git. Fork the victor repo on GitHub then clone your fork using

git clone

and then install in editable mode using

python -m pip install -e .[all]

where the .[full] at the end is optional and will install all the extra dependencies as above. (See (here)[] if you use zsh instead of bash.)

Alternatively, simply do

python -m pip install -e git+[all]

(but this means you won't be able to share your changes via pull requests).


Full documentation (including API) is a work in progress and will be updated soon. In the meantime, look at victor_usage_demo.ipynb for worked examples of typical use cases, and model_options_demo.ipynb for a fairly extensive summary of various model evaluation options.


victor is free software distributed under a GNU GPLv3 license. For details see the LICENSE.


Python code for likelihood analysis and MCMC posterior sampling of void-galaxy cross-correlation data







No releases published


No packages published