flip is a Python package that uses the maximum likelihood method to fit the growth rate based on the velocity and density fields. The first part of the software is the computation of a covariance matrix from a model power spectrum and the considered coordinates. This part is generalized to work for any linear power spectrum models, both for velocities, densities, and cross-terms, and it is optimized with Hankel transform for any model. In the second part, the covariance is used to create a likelihood by multiplying it by velocities or densities. Finally, this package includes some integrated fitters such as Minuit and MCMC (with emcee) to fit the growth rate of structures.
git clone https://github.com/corentinravoux/flip.git
cd flip
pip install .Mandatory: numpy, scipy, matplotlib, cosmoprimo, iminuit, emcee, sympy
Optional: classy, pypower
For an example with velocity fit check out:
For density only:
For a joint fit:
Documentation available on ReadTheDoc
This package was started on the previous work of @bastiencarreres, for now, please cite this article.
A paper describing the concept of the code is in preparation.