Extension of ILikelihood with explicit parameter-block access. More...

#include <IProfileableLikelihood.h>

Inheritance diagram for IProfileableLikelihood:

[legend]

Collaboration diagram for IProfileableLikelihood:

Public Member Functions
virtual std::size_t	p_dimension () const =0
	Returns the dimension of the fitted-parameter block.

virtual std::size_t	eta_dimension () const =0
	Returns the dimension of the nuisance-parameter block.

virtual std::vector< double >	central_p () const =0
	Returns the central values of the fitted parameters.

virtual std::vector< double >	central_eta () const =0
	Returns the central values of the nuisance parameters.

virtual std::vector< double >	predict (const std::vector< double > &p, const std::vector< double > &eta) const =0
	Evaluates the model prediction for split parameters.

virtual std::vector< double >	residuals (const std::vector< double > &p, const std::vector< double > &eta) const =0
	Computes observable residuals for split parameters.

virtual double	nll_from_split (const std::vector< double > &p, const std::vector< double > &eta) const =0
	Evaluates the negative log-likelihood from split parameters.

virtual RealMatrix	observable_curvature (const std::vector< double > &residuals) const =0
	Computes the observable-term curvature matrix.

virtual RealMatrix	nuisance_curvature (const std::vector< double > &eta) const =0
	Computes the nuisance-term curvature matrix.

Public Member Functions inherited from ILikelihood
virtual	~ILikelihood ()=default
	Virtual destructor for safe polymorphic deletion.

virtual double	nll (const std::vector< double > &theta) const =0
	Evaluates the negative log-likelihood at a given parameter point.

virtual std::vector< fit_app::ParameterDefinition >	get_param_defs () const =0
	Returns the metadata describing the likelihood parameters.

virtual std::size_t	dim () const =0
	Returns the total dimension of the parameter vector.

Detailed Description

Extension of ILikelihood with explicit parameter-block access.

Implementations must provide utilities to evaluate the likelihood from split parameter blocks, compute model predictions and residuals, and expose local curvature matrices for the observable and nuisance terms.

Definition at line 29 of file IProfileableLikelihood.h.

Member Function Documentation

◆ central_eta()

virtual std::vector< double > IProfileableLikelihood::central_eta ( ) const

pure virtual

Returns the central values of the nuisance parameters.

Returns: Vector containing the nominal values of $\eta$ .

Implemented in BaseLikelihood.

◆ central_p()

virtual std::vector< double > IProfileableLikelihood::central_p ( ) const

pure virtual

Returns the central values of the fitted parameters.

Returns: Vector containing the nominal values of .

Implemented in BaseLikelihood.

◆ eta_dimension()

virtual std::size_t IProfileableLikelihood::eta_dimension ( ) const

pure virtual

Returns the dimension of the nuisance-parameter block.

Returns: Number of scalar parameters in $\eta$ .

Implemented in BaseLikelihood.

◆ nll_from_split()

virtual double IProfileableLikelihood::nll_from_split	(	const std::vector< double > &	p,
		const std::vector< double > &	eta
	)		const

pure virtual

Evaluates the negative log-likelihood from split parameters.

Parameters

p	Fitted-parameter block.
eta	Nuisance-parameter block.

Returns: Negative log-likelihood value at $(p, \eta)$ .

Implemented in BaseLikelihood.

◆ nuisance_curvature()

virtual RealMatrix IProfileableLikelihood::nuisance_curvature ( const std::vector< double > & eta ) const

pure virtual

Computes the nuisance-term curvature matrix.

Parameters

eta	Nuisance-parameter vector.

Returns: Curvature matrix associated with the nuisance likelihood term.

Implemented in BaseLikelihood, and ChiSquaredLikelihood.

◆ observable_curvature()

virtual RealMatrix IProfileableLikelihood::observable_curvature ( const std::vector< double > & residuals ) const

pure virtual

Computes the observable-term curvature matrix.

Parameters

residuals Residual vector in observable space.

Returns: Curvature matrix associated with the observable likelihood term.

Implemented in BaseLikelihood, and ChiSquaredLikelihood.

◆ p_dimension()

virtual std::size_t IProfileableLikelihood::p_dimension ( ) const

pure virtual

Returns the dimension of the fitted-parameter block.

Returns: Number of scalar parameters in .

Implemented in BaseLikelihood.

◆ predict()

virtual std::vector< double > IProfileableLikelihood::predict	(	const std::vector< double > &	p,
		const std::vector< double > &	eta
	)		const

pure virtual

Evaluates the model prediction for split parameters.

Parameters

p	Fitted-parameter block.
eta	Nuisance-parameter block.

Returns: Model prediction vector associated with p and eta.

Implemented in BaseLikelihood.

◆ residuals()

virtual std::vector< double > IProfileableLikelihood::residuals	(	const std::vector< double > &	p,
		const std::vector< double > &	eta
	)		const

pure virtual

Computes observable residuals for split parameters.

Residuals are generally defined as model predictions minus the experimental observable central values.

Parameters

p	Fitted-parameter block.
eta	Nuisance-parameter block.

Returns: Residual vector in observable space.

Implemented in BaseLikelihood.

The documentation for this class was generated from the following file:

/home/runner/work/Hyperiso/Hyperiso/Hyperiso/Hyperiso/core/src/Statistic/domain/Likelihood/IProfileableLikelihood.h

Public Member Functions

Detailed Description

Member Function Documentation

◆ central_eta()

◆ central_p()

◆ eta_dimension()

◆ nll_from_split()

◆ nuisance_curvature()

◆ observable_curvature()

◆ p_dimension()

◆ predict()

◆ residuals()