edu.ucla.sspace.matrix

Class Normalize

java.lang.Object

edu.ucla.sspace.matrix.Normalize

public class Normalize
extends Object

A collection of normalization techniques available on a Matrix. Each method provided will alter the values of a given matrix in place with the values generated by the specified normalization technique.

Author:

Keith Stevens

Method Summary

Methods

Modifier and Type	Method and Description
static void	byColumn(Matrix m) Normalize the values of each column in the Matrix to be normalized by the length of each column.
static void	byCorrelation(Matrix m, boolean saveNegatives) Normalize the values of the Matrix by using the Pearson correlation.
static void	byLength(Matrix m) Normalize the values of each row in the Matrix to be normalized by the l2 norm of each row.
static void	byMagnitude(Matrix m) Normalize the all the values values in the Matrix to be normalized by the l2 norm of the entire Matrix.
static void	byRow(Matrix m) Normalize the values of each row in Matrix to be normalized by the length of each row.

Methods inherited from class java.lang.Object

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

Method Detail

byRow

public static void byRow(Matrix m)

Normalize the values of each row in Matrix to be normalized by the length of each row.

Parameters:

m - The matrix to normalize.

byColumn

public static void byColumn(Matrix m)

Normalize the values of each column in the Matrix to be normalized by the length of each column.

Parameters:

m - The matrix to normalize.

byLength

public static void byLength(Matrix m)

Normalize the values of each row in the Matrix to be normalized by the l2 norm of each row.

Parameters:

m - The matrix to normalize.

byMagnitude

public static void byMagnitude(Matrix m)

Normalize the all the values values in the Matrix to be normalized by the l2 norm of the entire Matrix.

Parameters:

m - The matrix to normalize.

byCorrelation

public static void byCorrelation(Matrix m,
                 boolean saveNegatives)

Normalize the values of the Matrix by using the Pearson correlation. This will give values between -1 and 1. If saveNegatives is is true, negative correlations will be saved, and otherwise they are reduced to zero.

Parameters:

m - The Matrix to normalize.

saveNegatives - If true, save all values, even if they are negative.