weka.classifiers.m5
Class Matrix

java.lang.Object
  |
  +--weka.classifiers.m5.Matrix

public final class Matrix

extends java.lang.Object

Class for handling a matrix

Author:

Yong Wang (yongwang@cs.waikato.ac.nz)

Constructor Summary

Matrix(int nr, int nc)
Constructs a matrix

Method Summary

void	lubksb(int n, int[] indx, double[] b) LU backward substitution
int[]	ludcmp(int n, int[] indx) LU decomposition
Matrix	multiply(Matrix b, int l, int m, int n) Reurns the multiplication of two matrices
double[]	regression(Matrix y, int n, int m) Linear regression
java.lang.String	toString(int nrl, int nrh, int ncl, int nch) Converts a matrix to a string
Matrix	transpose(int n, int m) Returns the transpose of a matrix [0:n-1][0:m-1]

Methods inherited from class java.lang.Object

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

Constructor Detail

Matrix

public Matrix(int nr,
              int nc)

Constructs a matrix

Parameters:

nr - the number of the rows

nc - the number of the columns

Method Detail

toString

public final java.lang.String toString(int nrl,
                                       int nrh,
                                       int ncl,
                                       int nch)

Converts a matrix to a string

Parameters:

nrl - the smallest index of the rows

nrh - the largest index of the rows

ncl - the smallest index of the column

ncl - the largest index of the column

Returns:

the converted string

transpose

public final Matrix transpose(int n,
                              int m)

Returns the transpose of a matrix [0:n-1][0:m-1]

Parameters:

n - the number of rows

m - the number of columns

Returns:

the transposed matrix

multiply

public final Matrix multiply(Matrix b,
                             int l,
                             int m,
                             int n)

Reurns the multiplication of two matrices

Parameters:

b - the multiplication matrix

l - the number of the rows of the instance matrix

m - the number of the columns of the instance matrix, and the number of the rows of matrix b

n - the number of the columns of matrix b

Returns:

the product matrix

regression

public final double[] regression(Matrix y,
                                 int n,
                                 int m)

Linear regression

Parameters:

y - the dependent variable vector

n - the number of the observations

m - the number of the coefficients

Returns:

the coefficients

lubksb

public final void lubksb(int n,
                         int[] indx,
                         double[] b)

LU backward substitution

Parameters:

n - the number of the coefficients

indx - the index

b - the double vector, storing constant terms in the equation sets; it later stores the computed coefficients' values

ludcmp

public final int[] ludcmp(int n,
                          int[] indx)

LU decomposition

Parameters:

n - the number of coefficients

indx - the index

Returns:

the integer vector of the attributes's singularities