English Français

Documentation / Manuel développeur

Modules disponibles

Calibration,  DataServer,  Launcher,  MetaModelOptim,  Modeler,  Optimizer,  ReLauncher,  Reliability,  ReOptimizer,  Sampler,  Sensitivity,  UncertModeler,  XmlProblem,   Uranie / Modeler: URANIE::Modeler::TGPMLCostFunction Class Reference
Uranie / Modeler v4.9.0
/* @license-end */
URANIE::Modeler::TGPMLCostFunction Class Reference

Description of the class TGPMLCostFunction. More...

#include <TGPMLCostFunction.h>

Inheritance diagram for URANIE::Modeler::TGPMLCostFunction:
Collaboration diagram for URANIE::Modeler::TGPMLCostFunction:

Public Member Functions

Constructor and Destructor
 TGPMLCostFunction (TGPBuilder *gpb)
 Standard constructor.
 
virtual ~TGPMLCostFunction ()
 Default destructor.
 
virtual URANIE::Relauncher::TEval * rebuild (int rank, Bool_t chef=kTRUE)
 
virtual Bool_t unbuild ()
 
Estimate
virtual Int_t eval (Double_t *theta, Double_t *res, int=0)
 Evaluate the cost function.
 
virtual Int_t eval (vector< URANIE::DataServer::UEntry * > *, vector< URANIE::DataServer::UEntry * > *, int=0)
 
- Public Member Functions inherited from URANIE::Modeler::TGPCostFunction
 TGPCostFunction (TGPBuilder *gpb)
 Standard constructor.
 
virtual ~TGPCostFunction ()
 Default destructor.
 
TGPBuildergetBuilder ()
 Return a pointer to the GP Builder.
 
Double_t * getCovarianceMatrix ()
 Return a pointer to the GP Builder covariance matrix.
 
Double_t * getCorrelationMatrix ()
 Return a pointer to the GP Builder correlation matrix.
 
Double_t * getObservationVector ()
 Return a pointer to the observations outputs.
 
Int_t getNObs ()
 Return the number of observations.
 
void setBuilder (TGPBuilder *gpb)
 
void setLog ()
 
void unsetLog ()
 
void changeLog ()
 
Bool_t getLog ()
 
virtual void printLog (Option_t *option="")
 

Additional Inherited Members

- Public Attributes inherited from URANIE::Modeler::TGPCostFunction
Bool_t _blog
 Boolean to decide if the log information is shown or not.
 
TGPBuildergpBuilder
 pointer to a gaussian process builder object.
 
- Protected Attributes inherited from URANIE::Modeler::TGPCostFunction
Double_t * _H
 pointer to the deterministic trend matrix (size: _n * _p)
 
Double_t * _K
 pointer to the covariance matrix (size: _n * _n)
 
Double_t * _C
 pointer to the correlation matrix (size: _n * _n)
 
Double_t * _yObs
 pointer to the vector of observations outputs (size: _n)
 
Int_t _n
 number of observations
 
Int_t _p
 number of deterministic trend coefficients
 

Detailed Description

Description of the class TGPMLCostFunction.

This class defines a Maximum Likelyhood cost function used to select the optimal parameters of the correlation function of a Gaussian Process. It is a wrapper to the gpML0 and gpMLh cost functions provided by the "gaussian process Library" (gpL), by Jean-Marc Martinez. NOTE: TGPMLCostFunction is unsuited to bayesian studies. Use TGPReMLCostFunction instead.

Constructor & Destructor Documentation

◆ TGPMLCostFunction()

URANIE::Modeler::TGPMLCostFunction::TGPMLCostFunction ( TGPBuilder gpb)

Standard constructor.

A TGPMLCostFunction object need to be associated with a TGPBuilder object which will provide the covariance matrix and the observations.

Parameters
gpb(TGPBuilder*): pointer to a Gaussian Process Builder object.
Warning
this class can't handle Bayesian studies. Use the TGPReMLCostFunction class instead.

Referenced by ClassImp().

◆ ~TGPMLCostFunction()

virtual URANIE::Modeler::TGPMLCostFunction::~TGPMLCostFunction ( )
virtual

Default destructor.

Referenced by ClassImp().

Member Function Documentation

◆ eval() [1/2]

virtual Int_t URANIE::Modeler::TGPMLCostFunction::eval ( Double_t *  theta,
Double_t *  res,
int  = 0 
)
virtual

Evaluate the cost function.

This function provides compatibility with the Relauncher module. It takes two arrays of double and returns an integer: 1 if all went well, 0 otherwise.

Parameters
theta(Double_t*): a single set of parameters to optimise.
res(Double_t*): output of the cost function.
Warning
If a measurement error is taken into account and its variance is provided by the user, the last element of the array "theta" is the GP variance. If the measurement error variance exists but is unknown, the last element of "theta" is the value of the parameter alpha.

Implements URANIE::Modeler::TGPCostFunction.

Referenced by ClassImp().

◆ eval() [2/2]

virtual Int_t URANIE::Modeler::TGPMLCostFunction::eval ( vector< URANIE::DataServer::UEntry * > *  ,
vector< URANIE::DataServer::UEntry * > *  ,
int  = 0 
)
virtual

◆ rebuild()

virtual URANIE::Relauncher::TEval * URANIE::Modeler::TGPMLCostFunction::rebuild ( int  rank,
Bool_t  chef = kTRUE 
)
virtual

Referenced by ClassImp().

◆ unbuild()

virtual Bool_t URANIE::Modeler::TGPMLCostFunction::unbuild ( )
virtual

Referenced by ClassImp().