lti::linearDiscriminantAnalysis< T > Class Template Reference
[LAPack based functors]

Functor for computing a linear discriminant analysis (LDA). More...

#include <ltiLDA.h>

Inheritance diagram for lti::linearDiscriminantAnalysis< T >:

[legend]

Collaboration diagram for lti::linearDiscriminantAnalysis< T >:

[legend]

List of all members.

Classes
class	parameters
	the parameters for the class linearDiscriminantAnalysis More...
Public Member Functions
	linearDiscriminantAnalysis ()
	linearDiscriminantAnalysis (const linearDiscriminantAnalysis &other)
virtual	~linearDiscriminantAnalysis ()
virtual const char *	getTypeName () const
bool	apply (matrix< T > &srcdest, const ivector &labels)
bool	apply (const matrix< T > &src, matrix< T > &dest, const ivector &labels)
bool	apply (const matrix< T > &src, matrix< T > &dest, const matrix< T > &Sb)
bool	apply (const matrix< T > &src, matrix< T > &result)
bool	apply (const vector< T > &src, vector< T > &result)
bool	apply (matrix< T > &srcdest)
void	setDimension (int k)
virtual bool	setParameters (const parameters &theParam)
bool	getEigenValues (vector< T > &result) const
const vector< T > &	getEigenValues () const
bool	getEigenVectors (matrix< T > &result) const
bool	getEigenVectorsInRows (matrix< T > &result) const
linearDiscriminantAnalysis &	copy (const linearDiscriminantAnalysis &other)
virtual functor *	clone () const
const parameters &	getParameters () const
bool	computeSb (const matrix< T > &src, matrix< T > &Sb, const ivector &labels) const
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::linearDiscriminantAnalysis< T >

Functor for computing a linear discriminant analysis (LDA).

It uses functors that require the LAPack.

See also:

LAPack based functors

---

Constructor & Destructor Documentation

template<class T >

lti::linearDiscriminantAnalysis< T >::linearDiscriminantAnalysis

(

)

default constructor

template<class T >

lti::linearDiscriminantAnalysis< T >::linearDiscriminantAnalysis

(

const linearDiscriminantAnalysis< T > &

other

)

copy constructor

Parameters:

other

the object to be copied

template<class T >

virtual lti::linearDiscriminantAnalysis< T >::~linearDiscriminantAnalysis

(

)

[virtual]

destructor

---

Member Function Documentation

template<class T >

bool lti::linearDiscriminantAnalysis< T >::apply

(

matrix< T > &

srcdest

)

On-Place version of the transformation.

Parameters:

srcdest

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

Returns:

true if the LDA could be applied, false otherwise

template<class T >

bool lti::linearDiscriminantAnalysis< T >::apply	(	const vector< T > &	src,
		vector< T > &	result
	)

Uses a previously computed transform in order to reduce the given data set's dimensionality according to the resultDim parameter.

Parameters:

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

Returns:

true if the LDA could be applied, false otherwise

template<class T >

bool lti::linearDiscriminantAnalysis< T >::apply	(	const matrix< T > &	src,
		matrix< T > &	result
	)

Uses a previously computed transform in order to reduce the given data set's dimensionality according to the resultDim parameter.

Parameters:

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

Returns:

true if the LDA could be applied, false otherwise

template<class T >

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

Computes a linear discriminant analysis for the data given in src.

This matrix is assumed to contain the data points for one class. The computed transformation is stored internally and can be used just as a global transformation. The main purpose of this method is to be used for LDA classifiers, which compute class-wise LDA transforms and classify by subsequently applying the LDA's of all classes to the vector, and then choosing the class whose mean is closest to the presented vector. This method can be used to compute the required LDAs.

Parameters:

	src	matrix<T> with the source data.
	dest	matrix<T> where the result will be left.
	Sb	matrix<T> which is the between-class scatter matrix of the entire population.

Returns:

true if the LDA could be computed, false otherwise

template<class T >

bool lti::linearDiscriminantAnalysis< T >::apply	(	const matrix< T > &	src,
		matrix< T > &	dest,
		const ivector &	labels
	)

Computes the linear discriminant analysis for the data given in src.

It internally stores a transformation matrix which is applied to the data. This matrix can be reused by calling one of the functions without the label vector.

Parameters:

	src	matrix<T> with the source data.
	dest	matrix<T> where the result will be left.
	labels	ivector for each row of the source matrix, this vector contains the class label.

Returns:

true if the LDA could be computed, false otherwise

template<class T >

bool lti::linearDiscriminantAnalysis< T >::apply	(	matrix< T > &	srcdest,
		const ivector &	labels
	)

operates on the given parameter.

Parameters:

	srcdest	matrix<T> with the source data. The result will be left here too.
	labels	ivector for each row of the source matrix, this vector contains the class label.

Returns:

true if the LDA could be computed, false otherwise

template<class T >

int lti::linearDiscriminantAnalysis< 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::linearDiscriminantAnalysis< T >::clone

(

)

const [virtual]

returns a pointer to a clone of this functor.

Implements lti::functor.

template<class T >

bool lti::linearDiscriminantAnalysis< T >::computeSb	(	const matrix< T > &	src,
		matrix< T > &	Sb,
		const ivector &	labels
	)			const

Computes the between-class scatter matrix Sb.

template<class T >

linearDiscriminantAnalysis& lti::linearDiscriminantAnalysis< T >::copy

(

const linearDiscriminantAnalysis< 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::linearDiscriminantAnalysis< T >::getEigenValues

(

)

const

Returns the previously computed eigenvalues of the scatter matrix.

Returns:

true if the values could be obtained, false otherwise.

template<class T >

bool lti::linearDiscriminantAnalysis< T >::getEigenValues

(

vector< T > &

result

)

const

Returns the previously computed eigenvalues of the scatter matrix.

Parameters:

result

the vector which will receive the eigenvalues.

Returns:

true if the values could be obtained, false otherwise.

template<class T >

bool lti::linearDiscriminantAnalysis< T >::getEigenVectors

(

matrix< T > &

result

)

const

Returns the previously computed eigenvectors of the scatter 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::linearDiscriminantAnalysis< T >::getEigenVectorsInRows

(

matrix< T > &

result

)

const

Returns the previously computed eigenvectors of the scatter 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::linearDiscriminantAnalysis< T >::getParameters

(

)

const

returns used parameters

Reimplemented from lti::functor.

template<class T >

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

(

)

const [virtual]

returns the name of this type ("linearDiscriminantAnalysis")

Reimplemented from lti::linearAlgebraFunctor.

template<class T >

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

Read the functor from the given ioHandler.

The default implementation is to read just the parameters object.

Since this virtual method needs to know the exact type of the parameters to call the proper read method, it will just assume that the current functor instance has a valid parameter set. If this is not the case, you need to reimplement the read method to set first a dummy parameter object.

Parameters:

	handler	the ioHandler to be used
	complete	if true (the default) the enclosing begin/end will be also written, otherwise only the data block will be written.

Returns:

true if write was successful

Reimplemented from lti::functor.

template<class T >

void lti::linearDiscriminantAnalysis< T >::setDimension

(

int

k

)

Set the dimension to which the vectors should be reduced.

template<class T >

virtual bool lti::linearDiscriminantAnalysis< T >::setParameters

(

const parameters &

theParam

)

[virtual]

set functor's parameters.

This member makes a copy of theParam: the functor will keep its own copy of the parameters! The difference to the standard is that this one passes the useSVD flag on to the matrix inverter.

Returns:

true if successful, false otherwise

Reimplemented from lti::functor.

template<class T >

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

return the last message set with setStatusString().

This will never return 0. If no status-string has been set yet an empty string (pointer to a string with only the char(0)) will be returned. set a status string.

Parameters:

msg

the const string to be reported next time by getStatusString(). The given string will be copied. This message will be usually set within the apply methods to indicate an error cause.

Note that the change of the status string is not considered as a change in the functor status. append a message to the current status string. Take care to reset the status string by calling setStatusString() for each call of an apply() or similar method. appendStatusString() should only be used after setStatusString() has been called.

Parameters:

msg

message to be appended to the current status string.

Note that the change of the status string is not considered as a change in the functor status. append an integer value to the current status string. Take care to reset the status string by calling setStatusString() for each call of an apply() or similar method. appendStatusString() should only be used after setStatusString() has been called.

Parameters:

msg

integer value to be appended to the current status string.

Note that the change of the status string is not considered as a change in the functor status. append a double value to the current status string. Take care to reset the status string by calling setStatusString() for each call of an apply() or similar method. appendStatusString() should only be used after setStatusString() has been called.

Parameters:

msg

double value to be appended to the current status string.

Note that the change of the status string is not considered as a change in the functor status. Append the status string of another functor to this functors status string. To achieve better readability of the resulting message a new line is started with the other functor's name and the message.

Parameters:

other

functor whose status string is to be append to this status string.

Note that the change of the status string is not considered as a change in the functor status. write the functor in the given ioHandler. The default implementation is to write just the parameters object.

Parameters:

	handler	the ioHandler to be used
	complete	if true (the default) the enclosing begin/end will be also written, otherwise only the data block will be written.

Returns:

true if write was successful

Reimplemented from lti::functor.

---

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

• ltiLDA.h