Plotloads: Difference between revisions

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
imported>Jeremy
(Importing text file)
 
imported>Jeremy
 
(9 intermediate revisions by 4 users not shown)
Line 1: Line 1:
===Purpose===
===Purpose===
Extract and display loadings information from model.
 
Extract and display loadings information from a model structure.
 
===Synopsis===
===Synopsis===
:a = plotloads(modl,''options'')
:a = plotloads(modl,''options'')
:a = plotloads(loads,''labels,classes'')
:a = plotloads(loads,''labels,classes'')
:options = plotloads('options')
 
===Description===
===Description===
Given a standard model structure, relevant loading information (e.g. labels) is collected and passed to PLOTGUI for plotting. The input is the model containing loadings to plot modl. (e.g. see MODELSTRUCT). Optional input ''options'' is discussed below.
Input loads is a ''N'' by ''K'' loadings matrix (class "double"). Optional input ''labels'' is a character or cell array with ''N'' rows containing sample labels, and optional input ''classes'' is a vector with ''N'' integer elements of class identifiers.


If no output is requested then PLOTLOADS initiates an interactive plotting utility to make loadings plots. If an output is requested, no plots are made, and the output a is a dataset object containing the loadings and labels, etc.
Given a standard model structure, relevant loading information (e.g. labels) is collected and passed to [[plotgui]] for plotting.
 
If no output is requested, then '''plotloads''' initiates an interactive plotting utility to make loadings plots. If an output is requested, no plots are made.
 
====Input====
 
* '''modl''' = a standard model structure, containing loadings to plot; '''plotloads''' will extract the loadings, labels and other information for plotting.
* '''loads''' = a ''N'' by ''K'' loadings matrix (class "double")
 
====Optional Input====
 
* '''labels''' = a character or cell array with ''N'' rows containing sample labels
* '''classes''' = a vector with ''N'' integer elements of class identifiers.
 
====Output====
 
* '''a''' = a dataset object containing the loadings, labels and other information required for subsequent plotting
 
===Options===
===Options===
* ''options'' =  a structure array with the following fields:
 
* display: [ {'on'} | 'off' ], governs level of display,
'''options''' =  a structure array with the following fields:
* plots: ['none' | 'final' | {'auto'} |], governs plotting behavior, 'auto' makes plots if no output is requested {default}, and
 
* figure: [],governs where plots are made, when figure = [] plots are made in a new figure window {default}, this can also be a valid figure number (i.e. figure handle).
* '''plots''': ['none' | 'final' | {'auto'} |], governs plotting behavior
The default options can be retreived using: options = plotloads('options');.
** 'auto' makes plots if no output is requested {default}
* '''figure''': [],governs where plots are made, when figure = [] plots are made in a new figure window {default}, this can also be a valid figure number (i.e. figure handle)
* '''mode''': [2] specifies which mode of loadings to plot. Default is 2 (columns of the original data). This only has                  significance for multi-way data.
* '''block''': [1] specifies which block to plot loadings for. 1 = x-block, 2 = y-block. If specified block does not exist, an error will be thrown.
* '''undopre''': [ {'no'} | 'yes' ] undo preprocessing on loadings (to the extent possible). Corrects loadings for scaling and some other preprocessing effects. Note that this produces loadings which are not the same as what the data will be projected onto but unpreprocessed loadings may be more interpretable.
 
 
The default options can be retrieved using: options = plotloads('options');.
 
===See Also===
===See Also===
[[analysis]], [[modelstruct]], [[pca]], [[pcr]], [[plotgui]], [[plotscores]], [[pls]]
 
[[analysis]], [[mcr]], [[modelstruct]], [[modelviewer]], [[mpca]], [[pca]], [[pcr]], [[ploteigen]], [[plotgui]], [[ploteigen]], [[plotscores]], [[pls]], [[sratio]], [[vip]]

Latest revision as of 12:02, 27 April 2015

Purpose

Extract and display loadings information from a model structure.

Synopsis

a = plotloads(modl,options)
a = plotloads(loads,labels,classes)

Description

Given a standard model structure, relevant loading information (e.g. labels) is collected and passed to plotgui for plotting.

If no output is requested, then plotloads initiates an interactive plotting utility to make loadings plots. If an output is requested, no plots are made.

Input

  • modl = a standard model structure, containing loadings to plot; plotloads will extract the loadings, labels and other information for plotting.
  • loads = a N by K loadings matrix (class "double")

Optional Input

  • labels = a character or cell array with N rows containing sample labels
  • classes = a vector with N integer elements of class identifiers.

Output

  • a = a dataset object containing the loadings, labels and other information required for subsequent plotting

Options

options = a structure array with the following fields:

  • plots: ['none' | 'final' | {'auto'} |], governs plotting behavior
    • 'auto' makes plots if no output is requested {default}
  • figure: [],governs where plots are made, when figure = [] plots are made in a new figure window {default}, this can also be a valid figure number (i.e. figure handle)
  • mode: [2] specifies which mode of loadings to plot. Default is 2 (columns of the original data). This only has significance for multi-way data.
  • block: [1] specifies which block to plot loadings for. 1 = x-block, 2 = y-block. If specified block does not exist, an error will be thrown.
  • undopre: [ {'no'} | 'yes' ] undo preprocessing on loadings (to the extent possible). Corrects loadings for scaling and some other preprocessing effects. Note that this produces loadings which are not the same as what the data will be projected onto but unpreprocessed loadings may be more interpretable.


The default options can be retrieved using: options = plotloads('options');.

See Also

analysis, mcr, modelstruct, modelviewer, mpca, pca, pcr, ploteigen, plotgui, ploteigen, plotscores, pls, sratio, vip