whistleClassifier.training

Class TrainingSelector

java.lang.Object

whistleClassifier.training.TrainingSelector

Direct Known Subclasses:

SequentialTrainingSelector

public abstract class TrainingSelector
extends java.lang.Object

Class containing functions to randomly select suitable groups of training and test data from a TrainingDataCollection

Author:

Doug Gillespie

See Also:

TrainingDataCollection, WhistleFragmenter

Constructor Summary

Constructors

Constructor and Description
TrainingSelector()

Method Summary

Methods

Modifier and Type	Method and Description
int	getNumSpecies()
abstract int	getNumTestSets()
double[][]	getParameterArray(int testTrainSet) Get a parameter array for feeding straight into the classifier for either the training set or the test set.
int	getTotalFragments(int iSpecies, double minFreq, double maxFreq, int minContourLength)
TrainingDataCollection	getTrainingDataCollection() get the training data collection
WhistleFragmenter	getWhistleFragmenter() Get the whistle fragmenter
void	setTrainingDataCollection(TrainingDataCollection trainingDataCollection) Sets the training data collection
void	setWhistleFragmenter(WhistleFragmenter whistleFragmenter) Set the whistle fragmenter to use.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

TrainingSelector

public TrainingSelector()

Method Detail

setTrainingDataCollection

public void setTrainingDataCollection(TrainingDataCollection trainingDataCollection)

Sets the training data collection

Parameters:

trainingDataCollection - training data

getTrainingDataCollection

public TrainingDataCollection getTrainingDataCollection()

get the training data collection

Returns:

training data collection

setWhistleFragmenter

public void setWhistleFragmenter(WhistleFragmenter whistleFragmenter)

Set the whistle fragmenter to use.

Parameters:

whistleFragmenter - whistle fragmenter

getWhistleFragmenter

public WhistleFragmenter getWhistleFragmenter()

Get the whistle fragmenter

Returns:

fragmenter

getNumSpecies

public int getNumSpecies()

Returns:

getNumTestSets

public abstract int getNumTestSets()

getTotalFragments

public int getTotalFragments(int iSpecies,
                    double minFreq,
                    double maxFreq,
                    int minContourLength)

getParameterArray

public double[][] getParameterArray(int testTrainSet)

Get a parameter array for feeding straight into the classifier for either the training set or the test set.

Parameters:

testTrainSet - index o set: 0 = training; 1 - n = test.