bilby.core.likelihood.GeorgeLikelihood

class bilby.core.likelihood.GeorgeLikelihood(kernel, mean_model, t, y, yerr=1e-06)

Bases: _GPLikelihood

__init__(kernel, mean_model, t, y, yerr=1e-06)

Basic Gaussian Process likelihood interface for celerite and george. For celerite documentation see: https://celerite.readthedocs.io/en/stable/ For george documentation see: https://george.readthedocs.io/en/latest/

Parameters:

kernel: george.kernels.Kernel

celerite or george kernel. See the respective package documentation about the usage.

mean_model: george.modeling.Model

Mean model

t: array_like

The times or x values of the data set.

y: array_like

The y values of the data set.

yerr: float, int, array_like, optional

The error values on the y-values. If a single value is given, it is assumed that the value applies for all y-values. Default is 1e-6, effectively assuming that no y-errors are present.

__call__(*args, **kwargs)

Call self as a function.

Methods

__init__(kernel, mean_model, t, y[, yerr])	Basic Gaussian Process likelihood interface for celerite and george.
log_likelihood()	Calculate the log-likelihood for the Gaussian process given the current parameters.
log_likelihood_ratio()	Difference between log likelihood and noise log likelihood
noise_log_likelihood()	Returns:
set_parameters(parameters)	Safely set a set of parameters to the internal instances of the gp and mean_model, as well as the parameters dict.

Attributes

marginalized_parameters
meta_data

log_likelihood()

Calculate the log-likelihood for the Gaussian process given the current parameters.

Returns:

float: The log-likelihood value.

log_likelihood_ratio()

Difference between log likelihood and noise log likelihood

Returns:

float

noise_log_likelihood()

Returns:

float

set_parameters(parameters)

Safely set a set of parameters to the internal instances of the gp and mean_model, as well as the parameters dict.

Parameters:

parameters: dict, pandas.DataFrame

The set of parameters we would like to set.