STK::ProjectedVariance Class Reference

A ProjectedVariance is an implementation of the abstract ILinearReduct class. More...

#include <STK_ProjectedVariance.h>

Inheritance diagram for STK::ProjectedVariance:

Collaboration diagram for STK::ProjectedVariance:

List of all members.

Public Member Functions
	ProjectedVariance (Matrix const *p_data)
	~ProjectedVariance ()
	Destructor.
virtual void	run (Integer const &dim)
	Compute the Index.
virtual void	run (Vector const *weights, Integer const &dim)
	Compute the weighted index.
void	svdData ()
	Do svd on the matrix p_data_.
void	svdWData ()
	Do svd on the matrix p_data_ with the weights of the samples.
Protected Member Functions
virtual void	maximizeIndex ()
	Find the axis by maximizing the Index.
virtual void	wmaximizeIndex ()
	Find the axis by maximizing the weighed Index.
void	dataReduced ()
	Compute the new coordinates of samples.
Protected Attributes
Svd	svdData_
	Object which contains the matrix resulting from svd of p_data_.
Private Member Functions
void	computeAxis ()
	Compute the axis using the matrix V_ of the svd of data.

---

Detailed Description

A ProjectedVariance is an implementation of the abstract ILinearReduct class.

ProjectedVariance (PCA) is the best method to reduce the dimension of data. This method computes the principal components. The number of principal components is the same of the number of variables. Then, the best components are selected thanks to the index of inertia.

Definition at line 32 of file STK_ProjectedVariance.h.

---

Constructor & Destructor Documentation

STK::ProjectedVariance::ProjectedVariance

(

Matrix const *

p_data

)

STK::ProjectedVariance::~ProjectedVariance

(

)

Destructor.

---

Member Function Documentation

virtual void STK::ProjectedVariance::run

(

Integer const &

dim

)

[virtual]

Compute the Index.

Parameters:

dim	the dimension of the reduced data set

virtual void STK::ProjectedVariance::run	(	Vector const *	weights,
		Integer const &	dim
	)		[virtual]

Compute the weighted index.

Parameters:

weights	the weights to used
dim	number of Axis to compute

void STK::ProjectedVariance::svdData

(

)

Do svd on the matrix p_data_.

void STK::ProjectedVariance::svdWData

(

)

Do svd on the matrix p_data_ with the weights of the samples.

virtual void STK::ProjectedVariance::maximizeIndex

(

)

[protected, virtual]

Find the axis by maximizing the Index.

Implements STK::ILinearReduct.

virtual void STK::ProjectedVariance::wmaximizeIndex

(

)

[protected, virtual]

Find the axis by maximizing the weighed Index.

void STK::ProjectedVariance::dataReduced

(

)

[protected]

Compute the new coordinates of samples.

void STK::ProjectedVariance::computeAxis

(

)

[private]

Compute the axis using the matrix V_ of the svd of data.

---

Member Data Documentation

Svd STK::ProjectedVariance::svdData_ [protected]

Object which contains the matrix resulting from svd of p_data_.

Definition at line 75 of file STK_ProjectedVariance.h.

---

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

• STK_ProjectedVariance.h