Purity
Contents |
Purpose
Calculation of pure variables.
Synopsis
- [purint,purspec] = purity(data,ncomp,options);
- [model] = purity(data,ncomp);
- [purint,purspec] = purity(data,ncomp,model);
- [model] = purity(data,model);
Description
PURITY calculates pure variables and resolves data into ncomp spectra of the pure components purspec and their contributions purint. For more information about the algorithm see PURITYENGINE. Data can be a matrix with the data or a dataset object.
The output arguments purity_values contains the purity values for all the variables and can be plotted as the "purity spectrum". The argument length_values contains the purity_values multiplied by the length of the variables. This results in a "length spectrum" that is easier to relate to the original data than the purity spectrum
Options
options = a structure array with the following fields:
- display: ['off'|{'on'}] display to command window.
- plot: ['off'|{'on'}] plotting of result.
- axistype: {2x1} [char]
- Mode 1: [{continuous}|'discrete'|'bar']
- Mode 2: [{continuous}|'discrete'|'bar'] defines plots. If emtpy the values of the (future) DSO field will be used in case they are not defined, the 'continuous' defaults will be used.
- select: [{[]},[1 2]] if empty, pure rows/columns will be selected from last slab, otherwise, the numbers identify from which slab(s) the pure rows/columns are selected.
- offset: [3 10] default noise correction factor for the two slabs.
- offset_row2col: 3 scalar value row2col offset, default is offset(1).
- mode: ['rows',{'cols'},'row2col'] determines if pure rows, cols are selected. row2col 2 is row-to-column solution.
- algorithm: 'purityengine' defines algorithm used.
- interactive: ['on',{'off'}, defines interactivity; 'on', 'cursor','inactivate','reactivate'] 'reactivate', 'cursor', 'inactivate', 'reactivate' are used for higher level calls for interactivity,'off' is used for demos and command mode applications.
- resolve: ['off'|{'on'}] indicates if the resolved results are required or not.
Examples
Resolving 4 components in a data set:
[purint,purspec]=purity(data,4)
Algorithm
The core algorithm is the function purityengine.