CLUS::MulticlassContinuousDistribution< T_Distribution > Class Template Reference

The class is a repository of continuous sistributions that each predict one of the class labels of a discrete variable. More...

#include <multiclasscontinuousdistribution.h>

Inheritance diagram for CLUS::MulticlassContinuousDistribution< T_Distribution >:

[legend]Collaboration diagram for CLUS::MulticlassContinuousDistribution< T_Distribution >:

[legend]


Public Member Functions
	MulticlassContinuousDistribution (int NoClasses, T_Distribution &D)
	Constructor.
virtual void	Infer (const double cdata, const int ddata, double *result)
	Infer will use data to produce noClasses normalized probabilities into result.
virtual void	MultiplicativeInfer (const double cdata, const int ddata, double *result)
	MultiplicativeInfer uses data to produce probabilities and multiplies these probabilities with the ones in result.
virtual void	StartLearning (void)
	Prepares the distributions for learning.
virtual void	LearnSample (const double cdata, const int ddata, int classLabel, double weightSample=1.0)
	Update the sufficient statistics according to the current input.
virtual void	LearnSample (const double cdata, const int ddata, double *classProbabilities, double weightSample=1.0)
virtual void	StopLearning (void)
	Uses the sufficient statistics to compute estimates of the parameters of the distribution.
virtual double	PValueStatisticalTest (void)
virtual bool	IsClassLabelAbsent (int index)
Protected Attributes
Vector< T_Distribution >	distributions

Detailed Description

template<class T_Distribution>
class CLUS::MulticlassContinuousDistribution< T_Distribution >

The class is a repository of continuous sistributions that each predict one of the class labels of a discrete variable.

Definition at line 50 of file multiclasscontinuousdistribution.h.

Constructor & Destructor Documentation

template<class T_Distribution>

CLUS::MulticlassContinuousDistribution< T_Distribution >::MulticlassContinuousDistribution ( int NoClasses,

T_Distribution & D

) [inline]

Constructor.

Parameters:

NoClasses number of classes

D a distribution that contains all the parameters the distributions inside the class have.

Definition at line 62 of file multiclasscontinuousdistribution.h.

Member Function Documentation

template<class T_Distribution>

virtual void CLUS::MulticlassContinuousDistribution< T_Distribution >::Infer ( const double * cdata,

const int * ddata,

double * result

) [inline, virtual]

Infer will use data to produce noClasses normalized probabilities into result.

Reimplemented from CLUS::MulticlassDistribution.
Definition at line 69 of file multiclasscontinuousdistribution.h.

template<class T_Distribution>

virtual bool CLUS::MulticlassContinuousDistribution< T_Distribution >::IsClassLabelAbsent ( int index ) [inline, virtual]

Returns:
true if the classLabel index has no significant apearance

Reimplemented from CLUS::MulticlassDistribution.
Definition at line 133 of file multiclasscontinuousdistribution.h.

template<class T_Distribution>

virtual void CLUS::MulticlassContinuousDistribution< T_Distribution >::LearnSample ( const double * cdata,

const int * ddata,

double * classProbabilities,

double weightSample = 1.0

) [inline, virtual]

Definition at line 103 of file multiclasscontinuousdistribution.h.

template<class T_Distribution>

virtual void CLUS::MulticlassContinuousDistribution< T_Distribution >::LearnSample ( const double * cdata,

const int * ddata,

int classLabel,

double weightSample = 1.0

) [inline, virtual]

Update the sufficient statistics according to the current input.
Should be used if the class label is known for sure.

Parameters:

cdata contains values for the continuous variables

ddata for the discrete ones

classLabel known classification label

weightSample used to give different importance to the samples (magnifying glass effect).

Reimplemented from CLUS::MulticlassDistribution.
Definition at line 98 of file multiclasscontinuousdistribution.h.

template<class T_Distribution>

virtual void CLUS::MulticlassContinuousDistribution< T_Distribution >::MultiplicativeInfer ( const double * cdata,

const int * ddata,

double * result

) [inline, virtual]

MultiplicativeInfer uses data to produce probabilities and multiplies these probabilities with the ones in result.

Reimplemented from CLUS::MulticlassDistribution.
Definition at line 85 of file multiclasscontinuousdistribution.h.

template<class T_Distribution>

double CLUS::MulticlassContinuousDistribution< T_Distribution >::PValueStatisticalTest ( void ) [virtual]

Todo:
fix this: return F(W, noClasses-1, n-noClasses, 3);

Reimplemented from CLUS::MulticlassDistribution.
Definition at line 212 of file simplenormaldistribution.h.

template<class T_Distribution>

virtual void CLUS::MulticlassContinuousDistribution< T_Distribution >::StartLearning ( void ) [inline, virtual]

Prepares the distributions for learning.

Reimplemented from CLUS::MulticlassDistribution.
Definition at line 92 of file multiclasscontinuousdistribution.h.

template<class T_Distribution>

virtual void CLUS::MulticlassContinuousDistribution< T_Distribution >::StopLearning ( void ) [inline, virtual]

Uses the sufficient statistics to compute estimates of the parameters of the distribution.

Reimplemented from CLUS::MulticlassDistribution.
Definition at line 111 of file multiclasscontinuousdistribution.h.

Field Documentation

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

Generated on Mon Jul 21 16:57:50 2003 for SECRET by

1.3.2

CLUS::MulticlassContinuousDistribution< T_Distribution > Class Template Reference

Public Member Functions

Protected Attributes

Detailed Description

template<class T_Distribution> class CLUS::MulticlassContinuousDistribution< T_Distribution >

Constructor & Destructor Documentation

Member Function Documentation

Field Documentation

template<class T_Distribution>
class CLUS::MulticlassContinuousDistribution< T_Distribution >