Base class for all Mixture densities. More...

#include <STK_IMixtureDensity.h>

Inheritance diagram for STK::IMixtureDensity< Derived >:

Public Types
typedef hidden::MixtureTraits< Derived >::Array	Array

typedef hidden::MixtureTraits< Derived >::Parameters	Parameters

typedef hidden::Traits< Array >::Type	Type

Public Member Functions
	~IMixtureDensity ()
	destructor

int	nbCluster () const

int	nbSample () const

Real	lnNbSample () const

Array const *const &	p_data () const

Parameters const &	param () const

void	setData (Array const &data)
	Set the data set.

void	setData (Array const &data, int nbRow, int nbCol, bool byRow=true)
	Set the data set and give dimensions.

bool	initializeStep ()
	This function will be called at the beginning of the estimation process once the model is created and data is set.

void	setParametersStep ()
	set the parameters obtained with the intermediate results and release the intermediate results.

void	finalizeStep ()
	This function will be called once the model is estimated.

template<class Weights >
Type	sample (int i, int j, Weights const &tk) const

Public Member Functions inherited from STK::IRecursiveTemplate< Derived >
Derived &	asDerived ()
	static cast : return a reference of this with a cast to the derived class.

Derived const &	asDerived () const
	static cast : return a const reference of this with a cast to the derived class.

Derived *	asPtrDerived ()
	static cast : return a ptr on a `Derived` of this with a cast to the derived class.

Derived const *	asPtrDerived () const
	static cast : return a ptr on a constant `Derived` of this with a cast to the derived class.

Derived *	clone () const
	create a leaf using the copy constructor of the Derived class.

Derived *	clone (bool isRef) const
	create a leaf using the copy constructor of the Derived class and a flag determining if the clone is a reference or not.

Protected Member Functions
	IMixtureDensity (int nbCluster)
	Default constructor.

	IMixtureDensity (IMixtureDensity const &model)
	copy constructor.

void	initializeModel ()
	Initialize the model before its first use.

bool	initializeStepImpl ()
	default implementation of initializeStepImpl (do nothing and return true)

void	finalizeStepImpl ()
	default implementation of finalizeStepImpl (do nothing)

void	setNbSample (int nbSample)
	Set the number of sample of the model (needed by kernel models)

Protected Member Functions inherited from STK::IRecursiveTemplate< Derived >
	IRecursiveTemplate ()
	constructor.

	~IRecursiveTemplate ()
	destructor.

Protected Attributes
Parameters	param_
	parameters of the derived mixture model.

Private Attributes
int	nbCluster_
	number of cluster.

int	nbSample_
	total available samples

Array const *	p_dataij_
	pointer on the data set

Detailed Description

template<class Derived>
class STK::IMixtureDensity< Derived >

Base class for all Mixture densities.

Let X be an arbitrary measurable space and let $\mathbf{x} =\{{\mathbf x}_1,...,{\mathbf x}_n\}$ be n independent vectors in X. If each ${\mathbf x}_i$ arises from a probability distribution with density

$f({\mathbf x}_i|\theta) = \sum_{k=1}^K p_k h({\mathbf x}_{i}| \lambda_{k},\alpha)$

where the $p_k$ 's are the mixing proportions, $h(\cdot| \lambda_{k},\alpha)$ denotes a distribution parameterized by $\lambda_k$ and $\alpha$ , it is said that we observe a mixture model.

This interface class is the base class for all distributions part of a mixture model.

See also: IMixture, IMixtureBridge, MixtureComposer

At this level the Parameters struct is created.

A call to setData trigger a call to initializeModel which itself trigger a call to initializeModelImpl. This last method can be re-implemented by derived class.

The pseudo virtual methods (needed by IMixtureBridge) to implement in concrete derived classes are

// return an imputed value for (i,j). In case of
template<class Weights>
Type impute(int i, int j, Weights const& pk) const;
// return a random value for (i,j)
Type rand(int i, int j, int k) const;
Real lnComponentProbability(int i, int k) const;
void randomInit( CArrayXX const* const& p_tik, CPointX const* const& p_tk) ;
bool run( CArrayXX const* const& p_tik, CPointX const* const& p_tk) ;
int computeNbFreeParameters() const;
void initializeModelImpl();

The pseudo virtual methods to re-implement if needed in derived classes are

// default implementation "do nothing" provided to all these methods
bool initializeStepImpl(); // return true by default
void finalizeStepImpl();

Definition at line 99 of file STK_IMixtureDensity.h.

Member Typedef Documentation

◆ Array

template<class Derived >

typedef hidden::MixtureTraits<Derived>::Array STK::IMixtureDensity< Derived >::Array

Definition at line 102 of file STK_IMixtureDensity.h.

◆ Parameters

template<class Derived >

typedef hidden::MixtureTraits<Derived>::Parameters STK::IMixtureDensity< Derived >::Parameters

Definition at line 103 of file STK_IMixtureDensity.h.

◆ Type

template<class Derived >

typedef hidden::Traits<Array>::Type STK::IMixtureDensity< Derived >::Type

Definition at line 104 of file STK_IMixtureDensity.h.

Constructor & Destructor Documentation

◆ IMixtureDensity() [1/2]

template<class Derived >

STK::IMixtureDensity< Derived >::IMixtureDensity ( int nbCluster )

protected

Default constructor.

Parameters

nbCluster the number of cluster

Definition at line 219 of file STK_IMixtureDensity.h.

                                         : param_(nbCluster)
                                         , nbCluster_(nbCluster)
                                         , nbSample_(0)
                                         , p_dataij_(0)
{}

◆ IMixtureDensity() [2/2]

template<class Derived >

STK::IMixtureDensity< Derived >::IMixtureDensity ( IMixtureDensity< Derived > const & model )

protected

copy constructor.

The Parameter class is copied using the copy constructor.
The pointer on the data set is copied as-is. Check if you should not change it on the copied object.
Parameters

model the model to copy

Definition at line 232 of file STK_IMixtureDensity.h.

                                         : param_(model.param_)
                                         , nbCluster_(model.nbCluster_)
                                         , nbSample_(model.nbSample_)
                                         , p_dataij_(model.p_dataij_)
{}

◆ ~IMixtureDensity()

template<class Derived >

STK::IMixtureDensity< Derived >::~IMixtureDensity ( )

inline

destructor

Definition at line 121 of file STK_IMixtureDensity.h.

121{}

Member Function Documentation

◆ finalizeStep()

template<class Derived >

void STK::IMixtureDensity< Derived >::finalizeStep ( )

inline

This function will be called once the model is estimated.

perform specific model finalization stuff

Definition at line 165 of file STK_IMixtureDensity.h.

165{ this->asDerived().finalizeStepImpl();}

STK::IRecursiveTemplate::asDerived

Derived & asDerived()

static cast : return a reference of this with a cast to the derived class.

Definition STK_IRecursiveTemplate.h:68

References STK::IRecursiveTemplate< Derived >::asDerived().

◆ finalizeStepImpl()

template<class Derived >

void STK::IMixtureDensity< Derived >::finalizeStepImpl ( )

inlineprotected

default implementation of finalizeStepImpl (do nothing)

Definition at line 197 of file STK_IMixtureDensity.h.

197{/* do nothing*/}

◆ initializeModel()

template<class Derived >

void STK::IMixtureDensity< Derived >::initializeModel ( )

protected

Initialize the model before its first use.

This function is triggered when data set is set. In this interface, the initializeModel() method

set the number of samples of the mixture model
call the derived class implemented method
initializeModelImpl()

for initialization of the specific model parameters if needed.

Definition at line 272 of file STK_IMixtureDensity.h.

{
  // set dimensions
  this->setNbSample(p_dataij_->sizeRows());
  // call specific model initialization stuff
  this->asDerived().initializeModelImpl();
}

◆ initializeStep()

template<class Derived >

bool STK::IMixtureDensity< Derived >::initializeStep ( )

inline

This function will be called at the beginning of the estimation process once the model is created and data is set.

Note: a stk++ mixture create and initialize all the containers when the data is set. Thus the default behavior is return true.

Definition at line 155 of file STK_IMixtureDensity.h.

155{ return this->asDerived().initializeStepImpl();}

References STK::IRecursiveTemplate< Derived >::asDerived().

◆ initializeStepImpl()

template<class Derived >

bool STK::IMixtureDensity< Derived >::initializeStepImpl ( )

inlineprotected

default implementation of initializeStepImpl (do nothing and return true)

Definition at line 195 of file STK_IMixtureDensity.h.

195{ return true;/* do nothing*/}

◆ lnNbSample()

template<class Derived >

Real STK::IMixtureDensity< Derived >::lnNbSample ( ) const

inline

Returns: the Log of the total available observations

Definition at line 128 of file STK_IMixtureDensity.h.

129 { return (nbSample_ <= 0) ? -Arithmetic<Real>::infinity() : std::log((Real)nbSample_);}

References STK::IMixtureDensity< Derived >::nbSample_.

◆ nbCluster()

template<class Derived >

int STK::IMixtureDensity< Derived >::nbCluster ( ) const

inline

Returns: the number of cluster

Definition at line 124 of file STK_IMixtureDensity.h.

124{ return nbCluster_;}

References STK::IMixtureDensity< Derived >::nbCluster_.

◆ nbSample()

template<class Derived >

int STK::IMixtureDensity< Derived >::nbSample ( ) const

inline

Returns: the total available observations

Definition at line 126 of file STK_IMixtureDensity.h.

126{ return nbSample_;}

References STK::IMixtureDensity< Derived >::nbSample_.

Referenced by STK::IMixtureDensity< Derived >::setNbSample().

◆ p_data()

template<class Derived >

Array const *const & STK::IMixtureDensity< Derived >::p_data ( ) const

inline

Returns: a pointer on the current data set

Definition at line 131 of file STK_IMixtureDensity.h.

131{ return p_dataij_;}

References STK::IMixtureDensity< Derived >::p_dataij_.

◆ param()

template<class Derived >

Parameters const & STK::IMixtureDensity< Derived >::param ( ) const

inline

Returns: a constant reference on the current parameter struct

Definition at line 133 of file STK_IMixtureDensity.h.

133{return param_;}

References STK::IMixtureDensity< Derived >::param_.

◆ sample()

template<class Derived >

template<class Weights >

Type STK::IMixtureDensity< Derived >::sample	(	int	i,
		int	j,
		Weights const &	tk
	)		const

inline

Returns: a simulated value for the jth variable of the ith sample

Parameters

i,j	indexes of the data to impute
tk	the probabilities of each class for the ith individual

Definition at line 172 of file STK_IMixtureDensity.h.

173 { return this->asDerived().rand(i, j, Law::Categorical::rand(tk));}

STK::Law::Categorical::rand

virtual int rand() const

Definition STK_Law_Categorical.h:160

References STK::IRecursiveTemplate< Derived >::asDerived(), and STK::Law::Categorical::rand().

◆ setData() [1/2]

template<class Derived >

void STK::IMixtureDensity< Derived >::setData ( Array const & data )

Set the data set.

Setting a (new) data set will trigger the initializeModel() method.

Parameters

data	the data set to set

Definition at line 244 of file STK_IMixtureDensity.h.

{
  p_dataij_ = &data;
  initializeModel();
}

◆ setData() [2/2]

template<class Derived >

void STK::IMixtureDensity< Derived >::setData	(	Array const &	data,
		int	nbRow,
		int	nbCol,
		bool	byRow = `true`
	)

Set the data set and give dimensions.

Setting a (new) data set will trigger the initializeModel() method.

Parameters

data	the data set to set
nbRow	number of rows of the data set
nbCol	number of columns of the data set
byRow	is data vectorized by row or by column ?

Note: this method is used for matrix valued data sets where data are vectorized

Definition at line 254 of file STK_IMixtureDensity.h.

{
  p_dataij_ = &data;
  initializeModel();
}

◆ setNbSample()

template<class Derived >

void STK::IMixtureDensity< Derived >::setNbSample ( int nbSample )

inlineprotected

Set the number of sample of the model (needed by kernel models)

Parameters

nbSample number of sample of the model

Note: to use when data set is not available but only the dimensions

Definition at line 203 of file STK_IMixtureDensity.h.

203{ nbSample_ = nbSample;}

STK::IMixtureDensity::nbSample

int nbSample() const

Definition STK_IMixtureDensity.h:126

References STK::IMixtureDensity< Derived >::nbSample(), and STK::IMixtureDensity< Derived >::nbSample_.

◆ setParametersStep()

template<class Derived >

void STK::IMixtureDensity< Derived >::setParametersStep ( )

inline

set the parameters obtained with the intermediate results and release the intermediate results.

Definition at line 158 of file STK_IMixtureDensity.h.

    {
      param_.setParametersStep();
      this->asDerived().setParametersImpl();
    }

References STK::IRecursiveTemplate< Derived >::asDerived(), and STK::IMixtureDensity< Derived >::param_.

Member Data Documentation

◆ nbCluster_

template<class Derived >

int STK::IMixtureDensity< Derived >::nbCluster_

private

number of cluster.

Definition at line 207 of file STK_IMixtureDensity.h.

Referenced by STK::IMixtureDensity< Derived >::nbCluster().

◆ nbSample_

template<class Derived >

int STK::IMixtureDensity< Derived >::nbSample_

private

total available samples

Definition at line 209 of file STK_IMixtureDensity.h.

Referenced by STK::IMixtureDensity< Derived >::lnNbSample(), STK::IMixtureDensity< Derived >::nbSample(), and STK::IMixtureDensity< Derived >::setNbSample().

◆ p_dataij_

template<class Derived >

Array const* STK::IMixtureDensity< Derived >::p_dataij_

private

pointer on the data set

Definition at line 211 of file STK_IMixtureDensity.h.

Referenced by STK::IMixtureDensity< Derived >::p_data(), and STK::KmmBase< Derived >::setData().

◆ param_

template<class Derived >

Parameters STK::IMixtureDensity< Derived >::param_

protected

parameters of the derived mixture model.

Should be an instance of the STK::ModelParameters struct.

See also: STK::ModelParameters

Definition at line 180 of file STK_IMixtureDensity.h.

Referenced by STK::IMixtureDensity< Derived >::param(), and STK::IMixtureDensity< Derived >::setParametersStep().

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

STK_IMixtureDensity.h

Public Types

Public Member Functions

Protected Member Functions

Protected Attributes

Private Attributes

Detailed Description

Member Typedef Documentation

◆ Array

◆ Parameters

◆ Type

Constructor & Destructor Documentation

◆ IMixtureDensity() [1/2]

◆ IMixtureDensity() [2/2]

◆ ~IMixtureDensity()

Member Function Documentation

◆ finalizeStep()

◆ finalizeStepImpl()

◆ initializeModel()

◆ initializeStep()

◆ initializeStepImpl()

◆ lnNbSample()

◆ nbCluster()

◆ nbSample()

◆ p_data()

◆ param()

◆ sample()

◆ setData() [1/2]

◆ setData() [2/2]

◆ setNbSample()

◆ setParametersStep()

Member Data Documentation

◆ nbCluster_

◆ nbSample_

◆ p_dataij_

◆ param_