fitting (pmrf.fitting)

The fitting module, for fitting ParamRF models to measured data.

All fitters in this module inherit from pmrf.fitting.BaseFitter. Fitting is done by initializing a Fitter class with the model and fitting features, and then calling pmrf.fitting.BaseFitter.run() with measured data. Fitting goals include the features to fit to (e.g. S11), the fit frequency, the cost function (for frequentist fitters) or the likelihood function (for Bayesian fitters).

When calling run, all key-word arguments are forwarded to the underlying backend (for example, PolyChord or SciPy). This allows full configuration of the optimization algorithm, while also providing a convenience wrapper for simple use.

Results are returned in the form of pmrf.fitting.FitResults. This contains the details about the initial configuration, as well as the sampled parameters and features.

Classes

BaseFitter(model, *[, frequency, features])	Base class for all ParamRF fitters (frequentist and Bayesian).
FitResults([initial_model, frequency, ...])	Container for the results of a model fitting process.
FrequentistFitter(model, *[, cost_kind, ...])	A base class for frequentist (classical) optimization methods.
BayesianFitter(model, *[, likelihood_kind, ...])	A base class for Bayesian inference fitters.
BlackJAXNSFitter(model, *[, ...])	Bayesian fitter using the BlackJAX nested slice sampling algorithm.
NumPyroFitter(model, *[, likelihood_kind, ...])	Base class for fitters utilizing the NumPyro probabilistic programming library.
NumPyroNSFitter(model, *[, likelihood_kind, ...])	Bayesian fitter using NumPyro Nested Sampling.
NumPyroMCMCFitter(model, *[, ...])	Bayesian fitter using NumPyro Markov Chain Monte Carlo (MCMC) sampling.
PolyChordFitter(model, *[, likelihood_kind, ...])	Bayesian fitter using the PolyChord nested sampling algorithm.
SciPyMinimizeFitter(model, *[, cost_kind, ...])	Frequentist fitter using the SciPy minimize backend with JAX acceleration.