Skip to content

SouthPoleTelescope/spt_candl_forecasts

Repository files navigation

Official SPT mock likelihoods for forecasts with candl

Official SPT mock likelihoods for forecasts with the differentiable CMB likelihood framework candl.

Download

To install the SPT candl mock likelihoods, simply navigate to where you would like to store the data and then run:

git clone https://github.com/SouthPoleTelescope/spt_candl_forecasts.git
cd spt_candl_forecasts
pip install .

This will download the relevant data files. The installation gives you access to handy short cuts that make it easier to initialise the likelihoods. Note that you also need to install candl in order to run the likelihoods, run pip install candl or see the repository for more detailed instructions.

To check that everything is working and to get a detailed summary of all mock likelihoods available:

import spt_candl_forecasts
spt_candl_forecasts.print_all_shortcuts()

Available mock data

Name Description References
SPT-3G Main T&E
SPT-3G Main field temperature and polarization

mock likelihood for 5 years of observation

Vitrier et al. 2025 (submitted)
SPT-3G Summer-a T&E
SPT-3G Summer-b T&E

SPT-3G Summer-c T&E

SPT-3G Summer field temperature and
polarization mock likelihoods for 4 years of

observation

Vitrier et al. 2025 (submitted)
SPT-3G Wide-a T&E
SPT-3G Wide-b T&E
SPT-3G Wide-c T&E
SPT-3G Wide-d T&E
SPT-3G Wide-e T&E
SPT-3G Wide-f T&E
SPT-3G Wide-g T&E
SPT-3G Wide-h T&E

SPT-3G Wide-i T&E

SPT-3G Wide field temperature and polarization

mock likelihoods for 1 year of observation

Vitrier et al. 2025 (submitted)
SPT-3G Main PP
SPT-3G Main field lensing mock likelihood for

5 years of observation

Vitrier et al. 2025 (submitted)
SPT-3G Summer-a PP
SPT-3G Summer-b PP

SPT-3G Summer-c PP

SPT-3G Summer field lensing mock likelihoods

for 4 years of observation

Vitrier et al. 2025 (submitted)
SPT-3G Wide PP
SPT-3G Wide field lensing mock likelihood for

1 year of observation

Vitrier et al. 2025 (submitted)
Planck Planck mock likelihood Vitrier et al. 2025 (submitted)

Additional Info

In order to forecast cosmological parameter constraints from the full Ext-10k survey, the SPT-3G likelihoods of the different fields have to be combined. The basic combination to start with is in temperature and polarization, combining the 13 fields of the Ext-10k survey (1 Main field, 3 Summer fields, and 9 Wide fields).

The temperature and polarization nuisance parameter priors are directly provided in the .yaml file of each field. However, only those of the Main field are de-commented, so that the priors are not counted 13 times. In the case of forecasting constraints from the Wide survey only, the nuisance parameter priors need to be de-commented in one of the Wide fields T&E .yaml files.

Note that a prior on the optical depth to reionization is also provided in the Main field .yaml file so that the Ext-10k Fisher matrix can be computed. Be careful not to double count this prior when running Markov Chain Monte Carlo (MCMC).

Getting Started

We supply files to help you use the SPT mock data with cobaya as well as a tutorial on how to interact with the mock likelihoods. You can find more help and tutorials in the candl documentation.

Notebook

tutorial_notebooks/SPT3G_Ext10k_TnE_tutorial.ipynb: this notebook uses the SPT-3G Ext-10k T&E mock likelihoods and shows you how to initialize the likelihoods and compute the Ext-10k T&E Fisher matrix.

Cobaya

You can find a template Cobaya .yaml file to help you launch chains as well as a ΛCDM proposal matrix in the cobaya/SPT3G_Ext10k_TnE_PP/ folder. This template shows you how to combine 18 likelihoods (13 T&E and 5 lensing) to forecast cosmological parameter constraints from the Ext-10k survey.


NSF USAP DOE ERC CNRS

IAP NEUCosmoS Sorbonne

Chicago Davis McGill

Berkeley Stanford SLAC

Fermilab Argonne

Melbourne Michigan Case Western

University of Illinois CAPS

Boulder Harvard

About

Official SPT mock likelihoods for forecasts with candl

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published