normaliz.m File Reference

Detailed Description

Normalization.

Direction	Type	type parameter	Description
row-wise	Max	v	"max" normalization across each row (if present, will be first task). idxs_fea affects only this option
col-wise	Mean-centering	c	centers variables
row-wise	Vector	n	Normalized to Euclidean norm (aka "Vector Normalization")
row-wise	Area	a	Normalizes to total area (makes area to be unity)
row-wise	Amide I	1	per-row normalization to Amide I peak
row-wise	Amide II	2	per-row normalization starting at wavenumber 1585 to seach max (Amide II)
col-wise	Standardization	s	centers and forces all variable variances to 1 (so-called "standardization")
col-wise	0-1 range	r	forces each variable range to [0, 1]

References [1] Pirouette (Infometrix Inc.) Help Documentation. [2] Bruker Optik GmbH, OPUS 5 Reference Manual. Ettlingen: Bruker, 2004. [3] J. G. Kelly et al., “Biospectroscopy to metabolically profile biomolecular structure: a multistage approach linking computational analysis with biomarkers,” J. Proteome Res., vol. 10, no. 4, pp. 1437-1448, Apr. 2011.

Definition in file normaliz.m.

Go to the source code of this file.

Functions
function	normaliz (in X, in x, in types, in idxs_fea)

Function Documentation

function normaliz	(	in	X,
		in	x,
		in	types,
		in	idxs_fea
	)

Parameters

X	data matrix containing rows as observations, columns as features.
x	x-axis values of the columns of X. unused in most cases, essential for '1' or '2' normalization.
types	see table
idxs_fea	optional idxs_fea for max normalization (types = 'v'). It is the FULL RANGE, NOT limits

Returns

Normalized X