CDC Paper II Referee Code

This repository contains the lightweight numerical code and support data for the Physics of the Dark Universe Paper II draft. It is intentionally limited to referee-facing reproducibility materials.

Included

• cdc_upgrade/paper2_check_claims.py: checks the paper-facing numerical claims against the committed data and, optionally, recomputes the DESI DR2 + compressed Planck geometric table.
• cdc_upgrade/paper2_geometric_reproduce.py: standalone reproduction of the background-level DESI DR2 BAO + compressed Planck fit used for Table 3 in the draft.
• cdc_upgrade/python/cdc_boltzmann_reference.py: stable reference solver for the CDC background and quasi-static growth checks.
• cdc_upgrade/mcmc/: Cobaya/CAMB approximation wrapper and run helpers for background-consistent likelihood experiments.
• cdc_upgrade/postprocess/: chain-table, convergence, growth, and SVG post-processing helpers.
• data/cdc_geometric_results.json: geometric BAO+CMB table quoted in the paper.
• data/cdc_perturbation_data.json: perturbation data used in the draft-level figures and tables.

Excluded

The repository deliberately excludes manuscript source/PDF files, LaTeX build products, large MCMC chain outputs, installed likelihood packages, caches, and temporary render images. The Paper III full CLASS/CAMB perturbation backend is not part of this Paper II repository.

Install

python -m pip install -r requirements.txt

Cobaya likelihood data and CAMB/Cobaya runtime setup are external dependencies and should be installed following Cobaya documentation.

Typical Smoke Checks

python cdc_upgrade/python/cdc_boltzmann_reference.py --help
python cdc_upgrade/postprocess/plot_cdc_results.py --help
python cdc_upgrade/postprocess/cdc_growth_from_chain.py --help
python cdc_upgrade/paper2_check_claims.py

Full Paper II Number Check

The full geometric minimization takes longer than the smoke checks:

python cdc_upgrade/paper2_check_claims.py --recompute-geometric

This recomputes the DESI DR2 Table-IV BAO + compressed Planck 2018 distance-prior table and checks it against data/cdc_geometric_results.json.