GUIBRUSHR.General_Constants.Classes.BestFitParametersAndResults module

Best-fit parameters and results data container for MCMC retrieval analysis.

This module contains the BestFitParametersAndResults class which stores statistical results from MCMC atmospheric retrieval runs, including chain medians, best-fit parameters, and likelihood values.

class GUIBRUSHR.General_Constants.Classes.BestFitParametersAndResults.BestFitParametersAndResults(number_bad_chains, last_max_LH, current_max_likelihood, lhood_monodimensional, list_best_pars, arr1, best_fit_parameters, value_fix, lista_par_fix, lista_parametri, list_best_pars_labels, burnndx=0, nchains=1, arr_full=None)[source]

Bases: object

Data container for MCMC retrieval statistics and results.

This class stores the statistical outputs from MCMC retrieval runs, including chain quality metrics, best-fit parameters, medians, and likelihood values. It serves as a structured container for passing retrieval results between analysis components.

Variables:

  • number_bad_chains – Number of chains identified as poor/outliers

  • last_max_LH – Maximum log-likelihood from previous iteration

  • current_max_likelihood – Current maximum log-likelihood value

  • lhood_monodimensional – Flattened log-likelihood values from all chains # IDL: chi2_monodimensional

  • list_best_pars – List of best-fit parameter values

  • arr1 – Array of parameter distributions (flattened chains)

  • best_fit_parameters – Median values for each parameter

  • value_fix – Values of fixed (non-fitted) parameters

  • lista_par_fix – List of fixed parameter names

  • lista_parametri – List of all parameter names (fitted + fixed)

  • list_best_pars_labels – Labels corresponding to best-fit parameters

__init__(number_bad_chains, last_max_LH, current_max_likelihood, lhood_monodimensional, list_best_pars, arr1, best_fit_parameters, value_fix, lista_par_fix, lista_parametri, list_best_pars_labels, burnndx=0, nchains=1, arr_full=None)[source]

Initialize the BestFitParametersAndResults data container.

Parameters:

  • number_bad_chains – Number of chains identified as outliers

  • last_max_LH – Previous maximum log-likelihood

  • current_max_likelihood – Current maximum log-likelihood

  • lhood_monodimensional – Flattened log-likelihood array # IDL: chi2_monodimensional

  • list_best_pars – Best-fit parameter values

  • arr1 – Flattened post-burn-in parameter chains, shape (nsteps*nchains, nfit)

  • best_fit_parameters – Parameter median values

  • value_fix – Fixed parameter values

  • lista_par_fix – Fixed parameter names

  • lista_parametri – All parameter names

  • list_best_pars_labels – Best-fit parameter labels

  • burnndx – Burn-in cut index (first step kept in posterior)

  • nchains – Number of good chains used in posterior

  • arr_full – Full chain including burn-in, shape (nfit, nsteps_total, nchains)