Batchalign and Comparemodels: Difference between pages

From Eigenvector Research Documentation Wiki
(Difference between pages)
Jump to navigation Jump to search
imported>Scott
(Created page with "===Purpose=== Convert data columns based on matching ref col to target vector. ===Synopsis=== :aligned = batchalign(data,ref_column,target,options) ===Description=== Stretch...")
 
imported>Scott
No edit summary
 
Line 1: Line 1:
===Purpose===
===Purpose===
 
Create summary table of models' performance statistics.
Convert data columns based on matching ref col to target vector.


===Synopsis===
===Synopsis===


:aligned = batchalign(data,ref_column,target,options)
:[columnkeys columnlabels tablevalues] = comparemodels(models)
:[columnkeys columnlabels tablevalues] = comparemodels(models,options)


===Description===
===Description===


Stretch or contract data columns based on a reference (target) using given 'method'. Savgol options can be used to exaggerate transitions in data prior to cow.
Generate a table which summarizes the performance of a set of input models, and also shows their relevant input parameters and options. This is useful for comparing the quality of a set of models.
The table shows:
* Parameters- parameters and important options from each model (LVs, PCs, gamma, number of points, etc).
* Statistics: (RMSEC/CV/P, # of support vectors, classification error rates as False Pos/Neg Rate average over classes. In all cases, value must be a single value for a given model (no vectors). Models which are not in the supported model list are ignored (see getallowedmodeltypes).


====Inputs====
====Inputs====
 
* '''models''' = a cell array containing one or more models.
* '''data''' = (nsample, nvar) dataset or double array containing data columns to align.
 
* '''ref_column''' =  scalar indicating the column # of data that we are going to match OR ref_vector which is a vector to use in match.
 
* '''target''' = (nsample, 1) vector or dataset to which we are trying to match the ref_column or ref_vector.


====Outputs====
====Outputs====
 
* '''columnkeys''' = 1 x ncolumn cell array containing keys identifying table columns in order.
* '''aligned''': dataset or double array (depending on input data type) containing the aligned data. If it is a dataset of different number of rows from input then the class, axisscale and labels are set according to the closest row in input dataset.
* '''columnlabels''' = 1 x ncolumn cell array containing descriptive labels for table columns in order.
* '''tablevalues''' = nmodel x ncolumn cell array containing values.


===Options===
===Options===
''options'' =  a structure array with the following fields:
* '''method''': [ 'linear' | {'cow'} | 'padwithnan' ] Alignment method.


* '''savgolwidth''': Number of points in savgol filter.
options =  a structure array with the following fields:
* '''savgolorder''': Order of savgol polynomial.
* '''display''': [ 'off' | {'on'} ] governs level of display to command window.
* '''savgolderiv''': [0,1,2...] Order of derivative to take of target and ref_column before doing alignment (default = 0).
* '''plots''': [ 'final' | {'none'} ] governs level of plotting.
[[cow|Cow]] only options.
* '''category''': [ {'all'} | 'classification' | 'regression' | 'decomposition' ]  restrict output to models of this category type, or all.
* '''cow.segments ''': number of segments.
* '''cow.slack''': max range of warping.
* '''cow.plots''': Governs plotting with COW (0 = no plots).
* '''cow.correlationpower''': correlation power.
* '''cow.forceequalsegs''': Force equal segment lengths in "xt" and "xP".
* '''cow.fixmaxcorrection''': Fix maximum correction.


===See Also===
===See Also===


[[alignmat]],[[cow]], [[matchrows]]
[[modeloptimizer]], [[modeloptimizergui]]

Revision as of 13:11, 28 May 2013

Purpose

Create summary table of models' performance statistics.

Synopsis

[columnkeys columnlabels tablevalues] = comparemodels(models)
[columnkeys columnlabels tablevalues] = comparemodels(models,options)

Description

Generate a table which summarizes the performance of a set of input models, and also shows their relevant input parameters and options. This is useful for comparing the quality of a set of models. The table shows:

  • Parameters- parameters and important options from each model (LVs, PCs, gamma, number of points, etc).
  • Statistics: (RMSEC/CV/P, # of support vectors, classification error rates as False Pos/Neg Rate average over classes. In all cases, value must be a single value for a given model (no vectors). Models which are not in the supported model list are ignored (see getallowedmodeltypes).

Inputs

  • models = a cell array containing one or more models.

Outputs

  • columnkeys = 1 x ncolumn cell array containing keys identifying table columns in order.
  • columnlabels = 1 x ncolumn cell array containing descriptive labels for table columns in order.
  • tablevalues = nmodel x ncolumn cell array containing values.

Options

options = a structure array with the following fields:

  • display: [ 'off' | {'on'} ] governs level of display to command window.
  • plots: [ 'final' | {'none'} ] governs level of plotting.
  • category: [ {'all'} | 'classification' | 'regression' | 'decomposition' ] restrict output to models of this category type, or all.

See Also

modeloptimizer, modeloptimizergui

Retrieved from "https://www.wiki.eigenvector.com/index.php?title=Batchalign&oldid=8754"