org.openimaj.math.matrix.algorithm.ica

Class IndependentComponentAnalysis

java.lang.Object

org.openimaj.math.matrix.algorithm.ica.IndependentComponentAnalysis

Direct Known Subclasses:

SymmetricFastICA

public abstract class IndependentComponentAnalysis
extends Object

Constructor Summary

Constructors

Constructor and Description
IndependentComponentAnalysis()

Method Summary

Modifier and Type	Method and Description
void	estimateComponents(double[][] data) Estimate the independent components of the given data array.
void	estimateComponents(Jama.Matrix data) Estimate the independent components of the given data array.
protected abstract void	estimateComponentsWhitened(Jama.Matrix Z, double[] mean, Jama.Matrix X, Jama.Matrix CC) Estimate the IC's from the given decorrelated (mean-centred and whitened) data matrix, Z.
abstract Jama.Matrix	getDemixingMatrix()
abstract Jama.Matrix	getIndependentComponentMatrix()
abstract Jama.Matrix	getSignalToInterferenceMatrix()

Methods inherited from class java.lang.Object

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

Constructor Detail

IndependentComponentAnalysis

public IndependentComponentAnalysis()

Method Detail

getSignalToInterferenceMatrix

public abstract Jama.Matrix getSignalToInterferenceMatrix()

getDemixingMatrix

public abstract Jama.Matrix getDemixingMatrix()

getIndependentComponentMatrix

public abstract Jama.Matrix getIndependentComponentMatrix()

estimateComponents

public void estimateComponents(double[][] data)

Estimate the independent components of the given data array. Each row corresponds to an observation with the number of dimensions equal to the number of columns.

Parameters:

data - the data

estimateComponents

public void estimateComponents(Jama.Matrix data)

Estimate the independent components of the given data array. Each row corresponds to an observation with the number of dimensions equal to the number of columns.

Parameters:

data - the data

estimateComponentsWhitened

protected abstract void estimateComponentsWhitened(Jama.Matrix Z,
                                                   double[] mean,
                                                   Jama.Matrix X,
                                                   Jama.Matrix CC)

Estimate the IC's from the given decorrelated (mean-centred and whitened) data matrix, Z.

Parameters:

Z - the whitened data; one observation per row

mean - the mean of each dimension

X - the mean-centered data; one observation per row