de.lmu.ifi.dbs.elki.math.linearalgebra
Class CovarianceMatrix

java.lang.Object
  de.lmu.ifi.dbs.elki.math.linearalgebra.CovarianceMatrix

public class CovarianceMatrix
extends Object

Class for computing covariance matrixes using stable mean and variance computations. This class encapsulates the mathematical aspects of computing this matrix. See DatabaseUtil for easier to use APIs. For use in algorithms, it is more appropriate to use StandardCovarianceMatrixBuilder since this class can be overriden with a stabilized covariance matrix builder!

Field Summary

(package private) double[][]	elements The covariance matrix
(package private) double[]	mean The means
(package private) double[]	nmea Temporary storage, to avoid reallocations
protected double	wsum The current weight

Constructor Summary

CovarianceMatrix(int dim)
Constructor.

Method Summary

Matrix	destroyToNaiveMatrix() Obtain the covariance matrix according to the population statistics: n degrees of freedom.
Matrix	destroyToSampleMatrix() Obtain the covariance matrix according to the sample statistics: (n-1) degrees of freedom.
Vector	getMeanVector() Get the mean as vector.
<F extends NumberVector<? extends F,?>> F	getMeanVector(Relation<? extends F> relation) Get the mean as vector.
static CovarianceMatrix	make(Matrix mat) Static Constructor.
static CovarianceMatrix	make(Relation<? extends NumberVector<?,?>> relation) Static Constructor from a full relation.
static CovarianceMatrix	make(Relation<? extends NumberVector<?,?>> relation, Iterable<DBID> ids) Static Constructor from a full relation.
Matrix	makeNaiveMatrix() Obtain the covariance matrix according to the population statistics: n degrees of freedom.
Matrix	makeSampleMatrix() Obtain the covariance matrix according to the sample statistics: (n-1) degrees of freedom.
void	put(double[] val) Add a single value with weight 1.0
void	put(double[] val, double weight) Add data with a given weight.
void	put(NumberVector<?,?> val) Add a single value with weight 1.0
void	put(NumberVector<?,?> val, double weight) Add data with a given weight.
void	put(Vector val) Add a single value with weight 1.0
void	put(Vector val, double weight) Add data with a given weight.

Methods inherited from class java.lang.Object

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

Field Detail

mean

double[] mean

The means

elements

double[][] elements

The covariance matrix

nmea

double[] nmea

Temporary storage, to avoid reallocations

wsum

protected double wsum

The current weight

Constructor Detail

CovarianceMatrix

public CovarianceMatrix(int dim)

Constructor.

Parameters:

dim - Dimensionality

Method Detail

put

public void put(double[] val)

Add a single value with weight 1.0

Parameters:

val - Value

put

public void put(double[] val,
                double weight)

Add data with a given weight.

Parameters:

val - data

weight - weight

put

public final void put(Vector val)

Add a single value with weight 1.0

Parameters:

val - Value

put

public final void put(Vector val,
                      double weight)

Add data with a given weight.

Parameters:

val - data

weight - weight

put

public void put(NumberVector<?,?> val)

Add a single value with weight 1.0

Parameters:

val - Value

put

public void put(NumberVector<?,?> val,
                double weight)

Add data with a given weight.

Parameters:

val - data

weight - weight

getMeanVector

public Vector getMeanVector()

Get the mean as vector.

Returns:

Mean vector

getMeanVector

public <F extends NumberVector<? extends F,?>> F getMeanVector(Relation<? extends F> relation)

Get the mean as vector.

Returns:

Mean vector

makeSampleMatrix

public Matrix makeSampleMatrix()

Obtain the covariance matrix according to the sample statistics: (n-1) degrees of freedom. This method duplicates the matrix contents, so it does allow further updates. Use destroyToSampleMatrix() if you do not need further updates.

Returns:

New matrix

makeNaiveMatrix

public Matrix makeNaiveMatrix()

Obtain the covariance matrix according to the population statistics: n degrees of freedom. This method duplicates the matrix contents, so it does allow further updates. Use destroyToNaiveMatrix() if you do not need further updates.

Returns:

New matrix

destroyToSampleMatrix

public Matrix destroyToSampleMatrix()

Obtain the covariance matrix according to the sample statistics: (n-1) degrees of freedom. This method doesn't require matrix duplication, but will not allow further updates, the object should be discarded. Use makeSampleMatrix() if you want to perform further updates.

Returns:

New matrix

destroyToNaiveMatrix

public Matrix destroyToNaiveMatrix()

Obtain the covariance matrix according to the population statistics: n degrees of freedom. This method doesn't require matrix duplication, but will not allow further updates, the object should be discarded. Use makeNaiveMatrix() if you want to perform further updates.

Returns:

New matrix

make

public static CovarianceMatrix make(Matrix mat)

Static Constructor.

Parameters:

mat - Matrix to use the columns of

make

public static CovarianceMatrix make(Relation<? extends NumberVector<?,?>> relation)

Static Constructor from a full relation.

Parameters:

relation - Relation to use.

make

public static CovarianceMatrix make(Relation<? extends NumberVector<?,?>> relation,
                                    Iterable<DBID> ids)

Static Constructor from a full relation.

Parameters:

relation - Relation to use.

ids - IDs to add