NAME

normalize_vector_2 - Normalizes the elements of a vector.

SYNOPSIS

#include "m/m_vec_norm.h"

Example compile flags (system dependent):
  -DLINUX_X86_64 -DLINUX_X86_64_OPTERON  -DGNU_COMPILER 
   -I/home/kobus/include
   -L/home/kobus/misc/load/linux_x86_64_opteron -L/usr/lib/x86_64-linux-gnu
  -lKJB                               -lfftw3  -lgsl -lgslcblas -ljpeg  -lSVM -lstdc++                    -lpthread -lSLATEC -lg2c    -lacml -lacml_mv -lblas -lg2c      -lncursesw 


int normalize_vector_2
(
	Vector **target_vpp,
	const Vector *vp,
	Norm_method normalization_method,
	double *scale_factor_ptr
);

PARAMETERS

Vector **target_vpp

Double pointer to output Vector.

const Vector *vp

Pointer to input Vector.

Norm_method normalization_method

Method to use.

Possible values for a

Norm_method

are as follows:

DONT_NORMALIZE ,
NORMALIZE_BY_MAGNITUDE ,
NORMALIZE_BY_MEAN ,
NORMALIZE_BY_SUM ,
NORMALIZE_BY_MAX_ABS_VALUE ,
NORMALIZATION_METHOD_ERROR

double *scale_factor_ptr

Optional return of scaling

DESCRIPTION

This routine normalizes an input vector and places the result into the specified target vector. The normalization method must be supplied as an argument. "target_vpp" is a double pointer to the Vector object to receive the normalized output. "vp" points to the source Vector object to normalize. "normalization_method" indicates which normalization to perform. Currently, the following values are supported: DONT_NORMALIZE - Do nothing. "target_vpp" is untouched. NORMALIZE_BY_MAGNITUDE - Vector magnitude normalized to 1. NORMALIZE_BY_MEAN - Vector mean value normailized to 1. NORMALIZE_BY_SUM - Sum of elements normalized to 1. NORMALIZE_BY_MAX_ABS_VALUE - Maximum absolute value is normalized to 1. The argument scale_factor_ptr can be used to return the scale factor used to normalize the vector. Set it to NULL if you are not interested. If "vp" is NULL, then "*target_vpp" becomes NULL also (and its contents are freed if it has any), and NO_ERROR is returned. All normalization methods divide each element of the vector by the specified quantity. Errors will be triggered if the normalization quantity is close to zero.

RETURNS

NO_ERROR on success, ERROR on failure.

ow_normalize_vector_2

DISCLAIMER

This software is not adequatedly tested. It is recomended that results are checked independantly where appropriate.

AUTHOR

Kobus Barnard

DOCUMENTER

Lindsay Martin and Kobus Barnard