We embeded the vd2020 code (available here) into montepython to create monteLLTB: a cosmological solver and sampler for the $\Lambda$LTB model. Taking advantage of the likelihood and sampler structure of montepython we include the $\Lambda$LTB cosmology by adapting the likelihood computation scheme. We started defining the method ini_LLTB in sampler.py,
which executes the solver vd2020 considering the current sampled point. Then, a call for ini_LLTB is included into the method compute_lkl to pass the $\Lambda$LTB solution to the corresponding likelihood. Note that this is possible since the method of the likehood loglkl now receives a new argument LLTBin, which contains the $\Lambda$LTB solution. We also modified the likelihoods in order to compute the observables according the $\Lambda$LTB predictions. Note that the output of vd2020 is managed by the file LLTB_functions.py, which contains definitions of distances and metric functions. Finally, it is important to mention that we modified vd2020 in order to customize the management of error, output precision and outputted functions. However, the core of the $\Lambda$LTB solver, the implementation to compute
- CLASS: Cosmic Linear Anisotropy Solving System
- Monte Python (version >= v3.3.0) with the Planck 20218 likelihoods
- scikit-learn
To install monteLLTB you should fisrt compile the vd2020 following the instructions on vd2020/README.PDF. Once vd2020 is installed you should modify montepython by doing:
cp -r montepython_files/* /path-to-your-montepython/
then modify the file default.conf.template to include the path to your vd2020 installation (besides the path to class and clik likelihoods). Set default.conf.template as your default configuration.
Repeat the procedure for class by doing:
cp -r class_files/* /path-to-your-class/
Compile class again and have fun with inhomogeneous cosmology!
monteLLTB was first introduced in (Camarena et al. 2021), please cite this paper if you make use of the code.