lti::serialPCA< T > Class Template Reference
[Linear Algebra]

Functor for sequentially computing a principal component analysis. More...

#include <ltiSerialPCA.h>

Inheritance diagram for lti::serialPCA< T >:

[legend]

Collaboration diagram for lti::serialPCA< T >:

[legend]

List of all members.

Classes
class	parameters
	the parameters for the class serialPCA More...
Public Member Functions
	serialPCA ()
	serialPCA (const serialPCA &other)
virtual	~serialPCA ()
virtual const char *	getTypeName () const
bool	consider (const matrix< T > &data)
bool	consider (const vector< T > &data)
bool	reset ()
bool	apply (const matrix< T > &src, matrix< T > &dest)
bool	apply (matrix< T > &srcdest)
bool	apply (const vector< T > &src, vector< T > &result)
bool	transform (const vector< T > &src, vector< T > &result)
bool	transform (const matrix< T > &src, matrix< T > &result)
bool	getTransformMatrix (matrix< T > &result) const
const matrix< T > &	getTransformMatrix () const
bool	getEigenValues (vector< T > &result) const
const vector< T > &	getEigenValues () const
bool	getEigenVectors (matrix< T > &result) const
bool	getEigenVectorsInRows (matrix< T > &result) const
const matrix< T > &	getEigenVectors () const
void	setDimension (int k)
serialPCA &	copy (const serialPCA &other)
virtual functor *	clone () const
const parameters &	getParameters () const
virtual bool	updateParameters ()
virtual bool	read (ioHandler &handler, const bool complete=true)
virtual bool	write (ioHandler &handler, const bool complete=true) const
Protected Member Functions
int	checkDim ()

---

Detailed Description

template<class T>
class lti::serialPCA< T >

Functor for sequentially computing a principal component analysis.

It receives a set of input vectors in form of a matrix (each row of the matrix corresponds to an input vector), which will be transformed with PCA.

The first time you use the apply()-method, the transformation matrix will be computed. You can use this transformation matrix with other data sets using the transform() methods.

Please note that the eigenvector matrices will contain the eigenvector in the COLUMNS and not in the rows, as could be expected. This avoids the requirement of transposing matrices in the vector transformations (see also lti::eigenSystem<T>).

See also:

lti::principalComponents

---

Constructor & Destructor Documentation

template<class T >

lti::serialPCA< T >::serialPCA

(

)

default constructor

template<class T >

lti::serialPCA< T >::serialPCA

(

const serialPCA< T > &

other

)

copy constructor

Parameters:

other

the object to be copied

template<class T >

virtual lti::serialPCA< T >::~serialPCA

(

)

[virtual]

destructor

---

Member Function Documentation

template<class T >

bool lti::serialPCA< T >::apply	(	const vector< T > &	src,
		vector< T > &	result
	)			[inline]

Transforms a single vector according to a previously computed transformation matrix.

References lti::serialPCA< T >::transform().

template<class T >

bool lti::serialPCA< T >::apply

(

matrix< T > &

srcdest

)

On-Place version of the transformation.

Computes the principal components of the previously considered data points and applies it to the given data matrix. The result is the transformed matrix. This method uses the eigenvector functor given in the parameters. By default, it uses ltilib's own jacobi functor. However, you can speed it up considerably by using the eigensystem functor in the lamath directory.

Warning:

This method does not consider the given srcdest data to compute the transformation, unless it is the very first time you call this functor. The normal way of using apply() is previously having called the consider() method.

Parameters:

srcdest

matrix<T> with the source data, which will also contain the result.

Returns:

a reference to srcdest.

template<class T >

bool lti::serialPCA< T >::apply	(	const matrix< T > &	src,
		matrix< T > &	dest
	)

Computes the principal components of the previously considered data points and applies it to the given data matrix.

The result is the transformed matrix. data and result must not be references to the same matrix. This method uses the eigenvector functor given in the parameters. By default, it uses ltilib's own jacobi functor. However, you can speed it up considerably by using the eigensystem functor in the lamath directory.

Warning:

This method does not consider the given src data to compute the transformation, unless it is the very first time you call this functor. The normal way of using apply() is previously having called the consider() method.

Parameters:

	src	matrix<T> with the source data.
	dest	matrix<T> with the result data.

Returns:

true if the PCA could be computed, false otherwise

template<class T >

int lti::serialPCA< T >::checkDim

(

)

[protected]

Determines the intrinsic dimensionality of the data set if the user specify autoDim, otherwise return parameters::resultDim.

The member usedDimensionality will be set with the returned value

template<class T >

virtual functor* lti::serialPCA< T >::clone

(

)

const [virtual]

returns a pointer to a clone of this functor.

Implements lti::functor.

template<class T >

bool lti::serialPCA< T >::consider

(

const vector< T > &

data

)

Consider a vector in the estimation of the covariance/correlation matrix.

You can later call the transform() method or the apply method to transform the data.

Each time you call this method, the current transformation matrix is invalidated and later recomputed when you call a transform method. Then, all data considered so far will be used in the covariance/correlation matrix estimation.

Parameters:

data

matrix<T> with the source data.

Returns:

true if the PCA could be computed, false otherwise

template<class T >

bool lti::serialPCA< T >::consider

(

const matrix< T > &

data

)

Consider the given data for the computation of a principal components transformation matrix.

You can later call the transform() method or the apply method to transform the data.

Each time you call this method, the current transformation matrix is invalidated and later recomputed when you call a transform method. Then, all data considered so far will be used in the covariance/correlation matrix estimation.

Parameters:

data

matrix<T> with the source data.

Returns:

true if the PCA could be computed, false otherwise

template<class T >

serialPCA& lti::serialPCA< T >::copy

(

const serialPCA< T > &

other

)

copy data of "other" functor.

Parameters:

other

the functor to be copied

Returns:

a reference to this functor object

Reimplemented from lti::functor.

template<class T >

const vector<T>& lti::serialPCA< T >::getEigenValues

(

)

const

Returns the previously computed eigenvalues of the covariance matrix.

Returns:

a const reference to the last computed eigenvalues

template<class T >

bool lti::serialPCA< T >::getEigenValues

(

vector< T > &

result

)

const

Returns the previously computed eigenvalues of the covariance matrix.

Parameters:

result

the vector which will receive the eigenvalues.

Returns:

true if the values could be obtained, false otherwise.

template<class T >

const matrix<T>& lti::serialPCA< T >::getEigenVectors

(

)

const

Returns the previously computed eigenvectors of the covariance matrix.

Returns:

a const reference to the last computed eigenvectors

template<class T >

bool lti::serialPCA< T >::getEigenVectors

(

matrix< T > &

result

)

const

Returns the previously computed eigenvectors of the covariance matrix.

Parameters:

result

the matrix which will receive the eigenvectors. Each column of the matrix contains one eigenvector.

Returns:

true if the vectors could be obtained, false otherwise

template<class T >

bool lti::serialPCA< T >::getEigenVectorsInRows

(

matrix< T > &

result

)

const

Returns the previously computed eigenvectors of the covariance matrix.

This method will call the normal getEigenVectors() methods and after that will transpose the obtained matrix, i.e. it is faster to get the eigenvectors in the columns.

Parameters:

result

the matrix which will receive the eigenvectors. Each row of the matrix contains one eigenvector.

Returns:

true if the vectors could be obtained, false otherwise

template<class T >

const parameters& lti::serialPCA< T >::getParameters

(

)

const

returns used parameters

Reimplemented from lti::functor.

template<class T >

const matrix<T>& lti::serialPCA< T >::getTransformMatrix

(

)

const

Returns the previously computed transform matrix.

Returns:

a const reference to the last computed or used transformation matrix.

template<class T >

bool lti::serialPCA< T >::getTransformMatrix

(

matrix< T > &

result

)

const

Returns the previously computed transform matrix.

Parameters:

result

the matrix which will receive the transformation matrix.

Returns:

true if the matrix could be computed, false otherwise.

template<class T >

virtual const char* lti::serialPCA< T >::getTypeName

(

)

const [virtual]

returns the name of this type ("serialPCA")

Reimplemented from lti::linearAlgebraFunctor.

template<class T >

virtual bool lti::serialPCA< T >::read	(	ioHandler &	handler,
		const bool	complete = true
	)			[virtual]

Reads this functor from the given handler.

Reimplemented from lti::functor.

template<class T >

bool lti::serialPCA< T >::reset

(

)

Reset the data.

Removes all data considered until now and set the functor as in the initial empty state.

template<class T >

void lti::serialPCA< T >::setDimension

(

int

k

)

Set the dimension to which the vectors should be reduced.

template<class T >

bool lti::serialPCA< T >::transform	(	const matrix< T > &	src,
		matrix< T > &	result
	)

Transform an entire matrix according to a previously computed transformation matrix.

Unfortunately, we must choose a name different from apply.

Parameters:

	src	the data matrix
	result	the matrix which will receive the transformed data

Returns:

a reference to result

template<class T >

bool lti::serialPCA< T >::transform	(	const vector< T > &	src,
		vector< T > &	result
	)

Transforms a single vector according to a previously computed transformation matrix.

Parameters:

	src	the data vector
	result	the vector which will receive the transformed data

Returns:

a reference to result

Referenced by lti::serialPCA< T >::apply().

template<class T >

virtual bool lti::serialPCA< T >::updateParameters

(

)

[virtual]

Update functor's parameters.

Initialize some internal data according to the parameters.

Returns:

true if successful, false otherwise

Reimplemented from lti::functor.

template<class T >

virtual bool lti::serialPCA< T >::write	(	ioHandler &	handler,
		const bool	complete = true
	)			const [virtual]

Writes this functor to the given handler.

Reimplemented from lti::functor.

---

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

• ltiSerialPCA.h