Package weka.core.matrix

Class MatrixHelper

java.lang.Object

weka.core.matrix.MatrixHelper

Direct Known Subclasses:

MatrixHelper

public class MatrixHelper
extends Object

Some matrix operations.

Author:

FracPete (fracpete at waikato dot ac dot nz)

Constructor Summary

Constructors

Constructor	Description
MatrixHelper()

Method Summary

Modifier and Type	Method	Description
static weka.core.matrix.Matrix	columnAsVector(weka.core.matrix.Matrix m, int columnIndex)	returns the given column as a vector (actually a n x 1 matrix)
static boolean	equal(weka.core.matrix.Matrix m1, weka.core.matrix.Matrix m2)	Compares the two matrices.
static boolean	equal(weka.core.matrix.Matrix m1, weka.core.matrix.Matrix m2, double epsilon)	Compares the two matrices.
static weka.core.matrix.Matrix	getAll(weka.core.Instances instances)	returns the data as matrix
static weka.core.matrix.Matrix	getDominantEigenVector(weka.core.matrix.Matrix m)	determines the dominant eigenvector for the given matrix and returns it
static weka.core.matrix.Matrix	getVector(weka.core.matrix.Matrix m, int columnIndex)	returns the (column) vector of the matrix at the specified index
static weka.core.matrix.Matrix	getX(weka.core.Instance instance)	returns the data minus the class column as matrix
static weka.core.matrix.Matrix	getX(weka.core.Instances instances)	returns the data minus the class column as matrix
static weka.core.matrix.Matrix	getY(weka.core.Instance instance)	returns the data class column as matrix
static weka.core.matrix.Matrix	getY(weka.core.Instances instances)	returns the data class column as matrix
static weka.core.matrix.Matrix	getY(weka.core.Instances instances, int[] cols)	returns the data class columns as matrix
static void	normalizeVector(weka.core.matrix.Matrix v)	normalizes the given vector (inplace)
static void	setVector(weka.core.matrix.Matrix v, weka.core.matrix.Matrix m, int columnIndex)	stores the data from the (column) vector in the matrix at the specified index
static weka.core.Instances	toInstances(weka.core.Instances header, weka.core.matrix.Matrix x, weka.core.matrix.Matrix y)	returns the X and Y matrix again as Instances object, based on the given header (must have a class attribute set).

Methods inherited from class java.lang.Object

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

Constructor Detail

MatrixHelper

public MatrixHelper()

Method Detail

getAll

public static weka.core.matrix.Matrix getAll(weka.core.Instances instances)

returns the data as matrix

Parameters:

instances - the data to work on

Returns:

the data

getX

public static weka.core.matrix.Matrix getX(weka.core.Instances instances)

returns the data minus the class column as matrix

Parameters:

instances - the data to work on

Returns:

the data without class attribute

getX

public static weka.core.matrix.Matrix getX(weka.core.Instance instance)

returns the data minus the class column as matrix

Parameters:

instance - the instance to work on

Returns:

the data without the class attribute

getY

public static weka.core.matrix.Matrix getY(weka.core.Instances instances)

returns the data class column as matrix

Parameters:

instances - the data to work on

Returns:

the class attribute

getY

public static weka.core.matrix.Matrix getY(weka.core.Instances instances,
                                           int[] cols)

returns the data class columns as matrix

Parameters:

instances - the data to work on

cols - the class columns

Returns:

the class attribute

getY

public static weka.core.matrix.Matrix getY(weka.core.Instance instance)

returns the data class column as matrix

Parameters:

instance - the instance to work on

Returns:

the class attribute

toInstances

public static weka.core.Instances toInstances(weka.core.Instances header,
                                              weka.core.matrix.Matrix x,
                                              weka.core.matrix.Matrix y)

returns the X and Y matrix again as Instances object, based on the given header (must have a class attribute set).

Parameters:

header - the format of the instance object

x - the X matrix (data)

y - the Y matrix (class)

Returns:

the assembled data

columnAsVector

public static weka.core.matrix.Matrix columnAsVector(weka.core.matrix.Matrix m,
                                                     int columnIndex)

returns the given column as a vector (actually a n x 1 matrix)

Parameters:

m - the matrix to work on

columnIndex - the column to return

Returns:

the column as n x 1 matrix

setVector

public static void setVector(weka.core.matrix.Matrix v,
                             weka.core.matrix.Matrix m,
                             int columnIndex)

stores the data from the (column) vector in the matrix at the specified index

Parameters:

v - the vector to store in the matrix

m - the receiving matrix

columnIndex - the column to store the values in

getVector

public static weka.core.matrix.Matrix getVector(weka.core.matrix.Matrix m,
                                                int columnIndex)

returns the (column) vector of the matrix at the specified index

Parameters:

m - the matrix to work on

columnIndex - the column to get the values from

Returns:

the column vector

getDominantEigenVector

public static weka.core.matrix.Matrix getDominantEigenVector(weka.core.matrix.Matrix m)

determines the dominant eigenvector for the given matrix and returns it

Parameters:

m - the matrix to determine the dominant eigenvector for

Returns:

the dominant eigenvector

normalizeVector

public static void normalizeVector(weka.core.matrix.Matrix v)

normalizes the given vector (inplace)

Parameters:

v - the vector to normalize

equal

public static boolean equal(weka.core.matrix.Matrix m1,
                            weka.core.matrix.Matrix m2)

Compares the two matrices.

Parameters:

m1 - the first matrix

m2 - the second matrix

Returns:

true if the same dimension and values, otherwise false

equal

public static boolean equal(weka.core.matrix.Matrix m1,
                            weka.core.matrix.Matrix m2,
                            double epsilon)

Compares the two matrices.

Parameters:

m1 - the first matrix

m2 - the second matrix

epsilon - the minimal accepted difference between the cells

Returns:

true if the same dimension and values, otherwise false