Main Page Class Hierarchy Alphabetical List Compound List Compound Members

prapi::SubSetSelector Class Reference

#include <FeatureSelector.h>

Inheritance diagram for prapi::SubSetSelector:

Detailed Description

A common base class for methods searching for sub-optimal subsets (of features, for example).

Public Methods

SubSetSelector (GoodnessMeasure &measure)

Create a new SubSetSelector with the given goodness measure.

virtual util::List< int > optimize (int totalCount, int desiredCount=-1)=0

Find a sub-optimal subset of items.

void stop ()

Stop the optimization.

double measureGoodness (const util::List< int > &enabledIndices)

Measure the goodness of a subset (of features).

void setGoodnessThreshold (double threshold)

Set the goodness value after which optimization is stopped.

double getGoodnessThreshold () const

Get the goodness value after which optimization is stopped.

Protected Attributes

bool _bStopped

A flag indicating whether the optimization should be continued.

GoodnessMeasure & _measure

The goodness measure.

double _dGoodnessThreshold

The goodness threshold.

Constructor & Destructor Documentation

prapi::SubSetSelector::SubSetSelector ( GoodnessMeasure & measure ) [inline]

Create a new SubSetSelector with the given goodness measure.
The goodness measure is consulted whenever a set of features needs to be evaluated.

Member Function Documentation

double prapi::SubSetSelector::measureGoodness ( const util::List< int > & enabledIndices )

Measure the goodness of a subset (of features).
This methods consults the internal goodness measure to perform the actual measurement.

virtual util::List<int> prapi::SubSetSelector::optimize ( int totalCount,

int desiredCount = -1

) [pure virtual]

Find a sub-optimal subset of items.
The implementation of this method is method-dependent. By convention, each implementation should fire a SelectionEvent every time an optimization round is finished.
Parameters:

totalCount the total number of items out of which the subset is to be selected.

the desired number of enabled items after optimization. If this value is set to a non-positive value, optimization is stopped only after the goodness threshold is exceeded.

Implemented in prapi::SequentialSelector, and prapi::BeamSelector.

void prapi::SubSetSelector::setGoodnessThreshold ( double threshold ) [inline]

Set the goodness value after which optimization is stopped.
To disable the threshold, set it to a non-positive number.

void prapi::SubSetSelector::stop ( ) [inline]

Stop the optimization.
This method is useful with fancy stopping criteria. If the optimization is not to be stopped after a certain number of iterations but according to some other rule, this method can be called from the measureGoodness() method of the GoodnessMeasure. This method sets the _bStopped flag to true.

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

FeatureSelector.h
FeatureSelector.cc

Public Methods
	SubSetSelector (GoodnessMeasure &measure)
	Create a new SubSetSelector with the given goodness measure.
virtual util::List< int >	optimize (int totalCount, int desiredCount=-1)=0
	Find a sub-optimal subset of items.
void	stop ()
	Stop the optimization.
double	measureGoodness (const util::List< int > &enabledIndices)
	Measure the goodness of a subset (of features).
void	setGoodnessThreshold (double threshold)
	Set the goodness value after which optimization is stopped.
double	getGoodnessThreshold () const
	Get the goodness value after which optimization is stopped.
Protected Attributes
bool	_bStopped
	A flag indicating whether the optimization should be continued.
GoodnessMeasure &	_measure
	The goodness measure.
double	_dGoodnessThreshold
	The goodness threshold.