Base class for canned Gaussian likelihoods. More...

#include <LikelihoodBase.h>

Inheritance diagram for QUESO::LikelihoodBase< V, M >:

Public Member Functions
virtual void	evaluateModel (const V &domainVector, const V domainDirection, V &modelOutput, V gradVector, M hessianMatrix, V hessianEffect) const
	Deprecated. Evaluates the user's model at the point `domainVector`. More...

virtual void	evaluateModel (const V &domainVector, V &modelOutput) const
	Evaluates the user's model at the point `domainVector`. More...

virtual double	actualValue (const V &domainVector, const V , V , M , V ) const
	Actual value of the scalar function. More...

Constructor/Destructor methods.
	LikelihoodBase (const char *prefix, const VectorSet< V, M > &domainSet, const V &observations)
	Default constructor. More...

virtual	~LikelihoodBase ()=0
	Destructor, pure to make this class abstract. More...

Public Member Functions inherited from QUESO::BaseScalarFunction< V, M >
void	setFiniteDifferenceStepSize (double fdStepSize)
	Sets the step size for finite differencing gradients. More...

void	setFiniteDifferenceStepSize (unsigned int i, double fdStepSize)

	BaseScalarFunction (const char *prefix, const VectorSet< V, M > &domainSet)
	Default constructor. More...

virtual	~BaseScalarFunction ()
	Destructor. More...

const VectorSet< V, M > &	domainSet () const
	Access to the protected attribute `m_domainSet:` domain set of the scalar function. More...

virtual double	lnValue (const V &domainVector, const V domainDirection, V gradVector, M hessianMatrix, V hessianEffect) const
	Logarithm of the value of the scalar function. Deprecated. More...

virtual double	lnValue (const V &domainVector) const
	Returns the logarithm of the function at `domainVector`. More...

virtual double	lnValue (const V &domainVector, V &gradVector) const
	Returns the logarithm of the function and its gradient at `domainVector`. More...

virtual double	lnValue (const V &domainVector, V &gradVector, const V &domainDirection, V &hessianEffect) const

Protected Attributes
const V &	m_observations

Protected Attributes inherited from QUESO::BaseScalarFunction< V, M >
const BaseEnvironment &	m_env

std::string	m_prefix

const VectorSet< V, M > &	m_domainSet
	Domain set of the scalar function. More...

Detailed Description

template<class V = GslVector, class M = GslMatrix>
class QUESO::LikelihoodBase< V, M >

Base class for canned Gaussian likelihoods.

This class is an abstract base class for 'canned' Gaussian likelihoods. All this class does is add a pure virtual function called evaluateModel that the user will implement to interact with the forward code.

Definition at line 49 of file LikelihoodBase.h.

Constructor & Destructor Documentation

template<class V , class M >

QUESO::LikelihoodBase< V, M >::LikelihoodBase	(	const char *	prefix,
		const VectorSet< V, M > &	domainSet,
		const V &	observations
	)

Default constructor.

The vector of observations must be passed. This will be used when evaluating the likelihood functional

Definition at line 35 of file LikelihoodBase.C.

   : BaseScalarFunction<V, M>(prefix, domainSet),
     m_observations(observations)
 {
 }

template<class V , class M >

QUESO::LikelihoodBase< V, M >::~LikelihoodBase ( )

pure virtual

Destructor, pure to make this class abstract.

Definition at line 44 of file LikelihoodBase.C.

45 {

46 }

Member Function Documentation

template<class V = GslVector, class M = GslMatrix>

virtual double QUESO::LikelihoodBase< V, M >::actualValue	(	const V &	domainVector,
		const V *	,
		V *	,
		M *	,
		V *
	)		const

inlinevirtual

Actual value of the scalar function.

Implements QUESO::BaseScalarFunction< V, M >.

Definition at line 98 of file LikelihoodBase.h.

References QUESO::BaseScalarFunction< V, M >::lnValue().

100 { return std::exp(this->lnValue(domainVector)); }

QUESO::BaseScalarFunction::lnValue

virtual double lnValue(const V &domainVector, const V *domainDirection, V *gradVector, M *hessianMatrix, V *hessianEffect) const

Logarithm of the value of the scalar function. Deprecated.

Definition: ScalarFunction.C:129

template<class V , class M >

void QUESO::LikelihoodBase< V, M >::evaluateModel	(	const V &	domainVector,
		const V *	domainDirection,
		V &	modelOutput,
		V *	gradVector,
		M *	hessianMatrix,
		V *	hessianEffect
	)		const

virtual

Deprecated. Evaluates the user's model at the point domainVector.

This method is deprecated. The user instead should subclass evaluateModel(const V & domainVector, V & modelOutput).

Definition at line 49 of file LikelihoodBase.C.

Referenced by QUESO::LikelihoodBase< V, M >::evaluateModel().

 {
   std::stringstream ss;
   ss << "ERROR: evaluateModel() not implemented! This interface is deprecated."
      << std::endl
      << "Prefer implementing evaluateModel(const V & domainVector, V & modelOutput)"
      << std::endl;
 
   queso_error_msg(ss.str());
 }

template<class V = GslVector, class M = GslMatrix>

virtual void QUESO::LikelihoodBase< V, M >::evaluateModel	(	const V &	domainVector,
		V &	modelOutput
	)		const

inlinevirtual

Evaluates the user's model at the point domainVector.

Subclass implementations will fill up the modelOutput vector with output from the model. This represents a vector of synthetic observations that will be compared to actual observations when computing the likelihood functional.

The first n components of domainVector are the model parameters. The rest of domainVector contains the hyperparameters, if any. For example, in GaussianLikelihoodFullCovarainceRandomCoefficient, the last component of domainVector contains the multiplicative coefficient of the observational covariance matrix. In this case, the user need not concern themselves with this parameter as it is handled not in the model evaluation but by the likelihood evaluation.

By default, to ensure backward compatibility, this method will call the deprecated evaluateModel( const V & domainVector, const V * domainDirection, V & modelOutput, V * gradVector, M * hessianMatrix, V * hessianEffect ).

Definition at line 94 of file LikelihoodBase.h.

References QUESO::LikelihoodBase< V, M >::evaluateModel().

95 { this->evaluateModel(domainVector,NULL,modelOutput,NULL,NULL,NULL); }

QUESO::LikelihoodBase::evaluateModel

virtual void evaluateModel(const V &domainVector, const V *domainDirection, V &modelOutput, V *gradVector, M *hessianMatrix, V *hessianEffect) const

Deprecated. Evaluates the user's model at the point domainVector.

Definition: LikelihoodBase.C:49