cobaya_mock_cmb

Mock CMB likelihood class for Cobaya sampler, and several specific experiment examples.

Introduced in Rashkovetskyi et al 2021, so please cite it if you use the code.

Python package contents (classes)

• MockCMBLikelihood - base mock CMB likelihood class. Ported MontePython's 3.3 Likelihood_mock_cmb with the help of Cobaya's example of likelihood class, most loops replaced with Numpy vectorized operations. Unlensed Cl's supported, but note that Cobaya theory class wrappers didn't provide them until Jan 2021.
• MockSO - Simons Observatory (SO) model following Sailer, Schaan and Ferraro 2020, based on MontePython config by Julian Munoz.
• MockSOBaseline - Simons Observatory (SO) with TT, EE deproj0 noise curves (baseline sensitivity).
• MockSOGoal - Simons Observatory (SO) with TT, EE deproj0 noise curves (goal sensitivity).
• MockCMBS4 - CMB-S4 model following the science book, based on MontePython config by Julian Munoz.
• MockCMBS4sens0 - CMB-S4 with TT, EE deproj0 noise curves.
• MockPlanck - Planck model following Munoz et al 2016 with f_sky=0.2 (fraction independent of SO and CMB-S4).

External contents

• make_fiducial.py - example script to generate fiducial power spectra for the experiments
• mock_test.yaml - example config to run Cobaya with all the likelihoods (cobaya-run mock_test.yaml in this directory) to make sure they work. One can't simply combine all in real runs, because they cover the same sky.