Python wrapper for the CosmoLSS likelihood. It can be used as a template for wrapping other CosmoMC likelihoods.
- Download the install script
CosmoLSSinstaller.shto you MontePython root folder. - If your C and Fortran compiler are different from gcc, change CC=gcc and FC=gcc in the last line of the install script. Example: CC=gcc-6 FC=gcc-6 make
- Run
bash CosmoLSSinstaller.sh. Beware that it will overwrite the CosmoLSS directory in montepython/likelihoods if it exist and also the CosmoLSS directory in data.
The CosmoLSS likelihood depends mainly on three things:
- Several helper classes defined in CosmoMC
- a set of functions provided by CAMB, like cosmological distances.
- data that are computed by CAMB but made available through the
CosmoMC class
TCosmoTheoryPredictionslike the matter power spectrum.
Item 1 is handled by using part of Antony Lewis forutils
package (https://github.com/cmbant/forutils), which is a subset of
CosmoMC for general purpose.
Item 2 and 3 are taken care of by creating Fortran interpolation
objects of all the relevant data from CLASS.
The cosmological parameters in CosmoMC are stored in the class
CMBParams, which in this implementation became a derived type
which must be updated every time the Python likelihood is called.
In the future, if all CosmoMC likelihoods restricted themselves to using only
the two classes CMBParams and
TCosmoTheoryPredictions, it might be worth wrapping these two
classes in a more complete way.