rlds-cosmo 0.1.0

Cosmological application layer for Regime-Limited Dynamical Systems (RLDS)

rlds-cosmo

RLDS-COSMO is the cosmological application layer of the Regime-Limited Dynamical Systems (RLDS) framework. It provides a reproducible high-redshift GRB forecast toolkit for testing the RLDS dark-sector closure against CPL/ΛCDM baselines.

Position in the RLDS software family

• rlds-mat is the mathematical core: RLDS reduction criteria, auxiliary-function geometry, attractor verification, and stability analysis.
• rlds-cosmo is the first application layer: shooting self-consistency for dark-sector cosmology, high-redshift GRB forecast machinery, and a Cobaya-compatible theory wrapper.

RLDS-MAT DOI: https://doi.org/10.5281/zenodo.20342965
RLDS overview DOI: https://doi.org/10.5281/zenodo.19670133

Headline GRB forecast

For the current RLDS posterior, the package reproduces the high-redshift GRB signature:

• median distance-modulus residual in 4 < z < 8: about −40 mmag over the posterior;
• best-fit example: about −32 mmag in 4 < z < 8;
• detection forecast at σ_μ = 0.25 mag: approximately 2.3σ for N_GRB = 200 and >3σ for N_GRB = 500 for the posterior median signal;
• Cobaya-compatible theory wrapper for use in standard cosmological pipelines.

The sign convention is

Δμ = μ_CPL+Planck − μ_RLDS

so a negative value means the RLDS prediction is brighter at high redshift than the CPL+Planck counter-fit.

Installation

From a local checkout:

pip install -e .

After PyPI publication:

pip install rlds-cosmo

Quick start

from rlds_cosmo import ShootingRLDS, best_fit_params, predict_grb_residual

model = ShootingRLDS(**best_fit_params())

print(f"Omega_m = {model.Omega_m:.4f}")
print(f"w_eff(z=1) = {model.w_eff(1.0):.5f}")

result = predict_grb_residual(model)
print(f"<Δμ> z=4-8 = {result['avg_4_8_mmag']:+.1f} mmag")
print(f"CPL counter-fit: w0={result['w0_cpl']:+.3f}, wa={result['wa_cpl']:+.3f}")

Expected best-fit-scale output:

Omega_m = 0.2998
w_eff(z=1) = -1.00149
<Δμ> z=4-8 = -32.1 mmag

Command line

rlds-cosmo info
rlds-cosmo quickstart
rlds-cosmo headline --samples 100 --output grb_headline.png

The headline command regenerates the four-panel GRB figure from posterior samples.

What is included

Module	Purpose
rlds_cosmo.model	Core ShootingRLDS cosmology and observables
rlds_cosmo.posterior	Bundled posterior chain and propagation helpers
rlds_cosmo.forecast	GRB residual forecast and CPL counter-fit machinery
rlds_cosmo.cobaya_module	Cobaya Theory plugin
rlds_cosmo.lcdm	Reference ΛCDM utilities

Cobaya integration

theory:
  rlds_cosmo.cobaya_module.RLDSTheory:
    speed: 50

params:
  K:       {prior: {min: 0.005, max: 0.25}, ref: 0.012}
  H0:      {prior: {min: 60.0,  max: 75.0}, ref: 67.44}
  R_c:     {prior: {min: 0.30,  max: 0.60}, ref: 0.51}
  Delta_R: {prior: {min: 0.03,  max: 0.20}, ref: 0.15}
  alpha:   {prior: {min: 3.01,  max: 3.30}, ref: 3.010}
  Omega_m_derived: {derived: true}

See examples/02_cobaya_demo.py and examples/rlds_demo.yaml.

Reproducing the GRB figure

python examples/01_grb_headline.py

or, after installation:

rlds-cosmo headline --samples 200 --output grb_headline.png

Runtime depends on the number of posterior samples. A 100–200 sample run is enough for a fast reproducibility check; larger runs are used for publication figures.

Important naming note

The distribution name is:

rlds-cosmo

The Python import name is:

import rlds_cosmo

The import package is intentionally not named rlds, to avoid conflicts with unrelated packages and to mirror the existing rlds_mat core package.

Citing

If you use this code, cite the RLDS overview record:

Kubanski, A. (2026). RLDS dark-sector research program. Zenodo. DOI: 10.5281/zenodo.19670133

When referring to the mathematical core, cite RLDS-MAT:

RLDS-MAT DOI: 10.5281/zenodo.20342965

License

MIT. See LICENSE.