jsat.datatransform

Class WhitenedZCA

java.lang.Object

jsat.datatransform.DataTransformBase

jsat.datatransform.WhitenedPCA

jsat.datatransform.WhitenedZCA

All Implemented Interfaces:

Serializable, Cloneable, DataTransform, InPlaceTransform, Parameterized

public class WhitenedZCA
extends WhitenedPCA
implements InPlaceTransform

An extension of WhitenedPCA, is the Whitened Zero Component Analysis. Whitened ZCA can not project to a lower dimension, as it rotates the output in the original dimension.

Author:

Edward Raff

See Also:

Serialized Form

Field Summary

Fields inherited from class jsat.datatransform.WhitenedPCA

dimensions, regularization, transform

Constructor Summary

Constructors

Constructor and Description
WhitenedZCA() Creates a new WhitenedZCA transform that uses up to 50 dimensions for the transformed space.
WhitenedZCA(DataSet dataSet) Creates a new Whitened ZCA transform from the given data.
WhitenedZCA(DataSet dataSet, double regularization) Creates a new Whitened ZCA transform from the given data.
WhitenedZCA(double regularization, int dims) Creates a new WhitenedZCA transform
WhitenedZCA(int dims) Creates a new WhitenedZCA transform

Method Summary

Modifier and Type	Method and Description
void	fit(DataSet dataSet) Fits this transform to the given dataset.
void	mutableTransform(DataPoint dp) Mutates the given data point.
boolean	mutatesNominal() By default returns false.
protected void	setUpTransform(SingularValueDecomposition svd) Creates the transform matrix to be used when converting data points.

Methods inherited from class jsat.datatransform.WhitenedPCA

clone, getDimensions, getRegularization, guessDimensions, setDimensions, setRegularization, transform

Methods inherited from class jsat.datatransform.DataTransformBase

getParameter, getParameters

Methods inherited from class java.lang.Object

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

Methods inherited from interface jsat.datatransform.DataTransform

clone, transform

Constructor Detail

WhitenedZCA

public WhitenedZCA()

Creates a new WhitenedZCA transform that uses up to 50 dimensions for the transformed space. This may not be optimal for any given dataset.

Parameters:

dims - the number of dimensions to project down to

WhitenedZCA

public WhitenedZCA(int dims)

Creates a new WhitenedZCA transform

Parameters:

dims - the number of dimensions to project down to

WhitenedZCA

public WhitenedZCA(double regularization,
                   int dims)

Creates a new WhitenedZCA transform

Parameters:

regularization - the amount of regularization to add, avoids numerical instability

dims - the number of dimensions to project down to

WhitenedZCA

public WhitenedZCA(DataSet dataSet,
                   double regularization)

Creates a new Whitened ZCA transform from the given data.

Parameters:

dataSet - the data set to whiten

regularization - the amount of regularization to add, avoids numerical instability

WhitenedZCA

public WhitenedZCA(DataSet dataSet)

Creates a new Whitened ZCA transform from the given data. The regularization parameter will be chosen as the log of the condition of the covariance.

Parameters:

dataSet - the data set to whiten

Method Detail

fit

public void fit(DataSet dataSet)

Description copied from interface: DataTransform

Fits this transform to the given dataset. Some transforms can only be learned from classification or regression datasets. If an incompatible dataset type is given, a FailedToFitException exception may be thrown.

Specified by:

fit in interface DataTransform

Overrides:

fit in class WhitenedPCA

Parameters:

dataSet - the dataset to fir this transform to

mutableTransform

public void mutableTransform(DataPoint dp)

Description copied from interface: InPlaceTransform

Mutates the given data point. This causes side effects, altering the data point to have the same value as the output of DataTransform.transform(jsat.classifiers.DataPoint).

Specified by:

mutableTransform in interface InPlaceTransform

Parameters:

dp - the data point to alter

mutatesNominal

public boolean mutatesNominal()

Description copied from interface: InPlaceTransform

By default returns false. Only returns true if this transform will mutableTransform the nominal feature values of a data point.

Specified by:

mutatesNominal in interface InPlaceTransform

Returns:

true if nominal feature values are mutated, false otherwise.

setUpTransform

protected void setUpTransform(SingularValueDecomposition svd)

Description copied from class: WhitenedPCA

Creates the transform matrix to be used when converting data points. It is called in the constructor after all values are set.

Overrides:

setUpTransform in class WhitenedPCA

Parameters:

svd - the SVD of the covariance of the source data set