marf.Storage
Class FeatureSet

java.lang.Object
  marf.Storage.TrainingSample
      marf.Storage.Cluster
          marf.Storage.FeatureSet

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, ITrainingSample

public class FeatureSet
extends Cluster

FeatureSet -- Encapsulates subject's feature vectors. Additionally, can compute mean and median vectors off the present feature vectors.

$Id: FeatureSet.java,v 1.18 2008/02/17 02:26:53 mokhov Exp $

Since:

0.3.0.1

Version:

$Revision: 1.18 $

Author:

Serguei Mokhov

See Also:

Serialized Form

Field Summary

protected java.util.Vector

oFeatureVectors A Vector of TrainingSamples.

Fields inherited from class marf.Storage.TrainingSample

adDataVector, iSubjectID, oFilenames

Constructor Summary

FeatureSet()
Construct a training set object.

Method Summary

boolean	addFeatureVector(double[] padFeatureVector, java.lang.String pstrFilename, int piSubjectID) Adds new feature vector to the mean and recomputes the mean.
java.lang.Object	clone() Implements Cloneable interface for the FeatureSet object.
void	dumpCSV() Dump the current training set to disk.
java.util.Vector	getFeatureVectors() Retrieves training samples.
static java.lang.String	getMARFSourceCodeRevision() Returns source code revision information.
int	getMeanCount() Retrieves current mean count.
double[]	getMeanVector() Simply retrieves the data vector.
double[]	getMedianVector() Simply retrieves the data vector.
void	restoreCSV() Retrieve the current training set from disk.
int	size() Sizes of the feature vectors set.
java.lang.String	toString() Provides string representation of the training set data in addition to that of the parent Cluster.

Methods inherited from class marf.Storage.Cluster

addFilename, dumpCSV, existsFilename, restoreCSV, setMeanVector

Methods inherited from class marf.Storage.TrainingSample

getDataVector, getSubjectID, setDataVector, setFeatureVector, setFilename, setSubjectID

Methods inherited from class java.lang.Object

equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

oFeatureVectors

protected java.util.Vector oFeatureVectors

A Vector of TrainingSamples.

Constructor Detail

FeatureSet

public FeatureSet()

Construct a training set object.

Method Detail

getFeatureVectors

public java.util.Vector getFeatureVectors()

Retrieves training samples.

Returns:

vector of training samples.

size

public int size()

Sizes of the feature vectors set.

Specified by:

size in interface ITrainingSample

Overrides:

size in class TrainingSample

Returns:

number of training samples in the set

See Also:

ITrainingSample.size()

restoreCSV

public void restoreCSV()
                throws StorageException

Retrieve the current training set from disk. TODO: implement.

Throws:

StorageException

NotImplementedException

Since:

0.3.0.5

dumpCSV

public void dumpCSV()
             throws StorageException

Dump the current training set to disk. TODO: implement.

Throws:

StorageException

NotImplementedException

Since:

0.3.0.5

getMeanVector

public double[] getMeanVector()

Description copied from class: TrainingSample

Simply retrieves the data vector. Internally calls getDataVector().

Specified by:

getMeanVector in interface ITrainingSample

Overrides:

getMeanVector in class TrainingSample

Returns:

array of doubles representing the mean for that cluster

See Also:

TrainingSample.getDataVector()

getMedianVector

public double[] getMedianVector()

Description copied from class: TrainingSample

Simply retrieves the data vector. Internally calls getDataVector().

Specified by:

getMedianVector in interface ITrainingSample

Overrides:

getMedianVector in class TrainingSample

Returns:

array of doubles representing the mean for that cluster

See Also:

TrainingSample.getDataVector()

addFeatureVector

public boolean addFeatureVector(double[] padFeatureVector,
                                java.lang.String pstrFilename,
                                int piSubjectID)

Description copied from class: Cluster

Adds new feature vector to the mean and recomputes the mean.

Specified by:

addFeatureVector in interface ITrainingSample

Overrides:

addFeatureVector in class Cluster

Parameters:

padFeatureVector - vector to add

pstrFilename - filename to add

piSubjectID - for which subject that vector is

Returns:

true if the vector was added; false otherwise

See Also:

ITrainingSample.setFeatureVector(double[], String, int)

getMeanCount

public int getMeanCount()

Description copied from class: Cluster

Retrieves current mean count.

Specified by:

getMeanCount in interface ITrainingSample

Overrides:

getMeanCount in class Cluster

Returns:

mean count

See Also:

ITrainingSample.getMeanCount()

clone

public java.lang.Object clone()

Implements Cloneable interface for the FeatureSet object. Performs a "deep" copy of this object including all the feature vectors.

Overrides:

clone in class Cluster

Since:

0.3.0.5

See Also:

Object.clone()

toString

public java.lang.String toString()

Provides string representation of the training set data in addition to that of the parent Cluster.

Overrides:

toString in class Cluster

Since:

0.3.0.6

See Also:

Cluster.toString()

getMARFSourceCodeRevision

public static java.lang.String getMARFSourceCodeRevision()

Returns source code revision information.

Returns:

revision string

Since:

0.3.0.2