If you are looking for the distance-modulus covariance matrices from https://arxiv.org/abs/2506.04327, they are here: https://github.com/rubind/roman_fisher_results
A few notes about how to run things:
You will need two GitHubs:
https://github.com/rubind/wfirst-sim
and
https://github.com/rubind/wfirst-sim-data
after cloning:
export WFIRST_SIM_DATA=/path/to/wfirst-sim-data
export WFIRST_SIM=/path/to/wfirst-sim
The simulation package:
wfirst-sim/scripts/stan_cosmo/STEP1_simple_survey.py
Which takes a parameter file (e.g., scripts/stan_cosmo/paramfile_CCS_DDF.csv) and the name of an output pickle file:
python STEP1_simple_survey.py paramfile.csv survey.pickle
And for making some summary plots:
python STEP1A_plot_survey survey.pickle
Then wfirst-sim/scripts/stan_cosmo/STEP2_Analytic_Fisher.py runs with the pickle file and makes a distance-modulus covariance matrix (comb_mat.fits).
python STEP2_Analytic_Fisher.py survey.pickle
wfirst-sim/scripts/stan_cosmo/FoM.py takes that comb_mat.fits and makes FoM values. We’ve been using FoM_0.26.
python FoM.py */comb_mat*.fits
wfirst-sim/scripts/stan_cosmo/STEP1_survey_grid.py generates sets of surveys (it's setup for our Slurm queue) so that's good to look at as well.