Compressmodel: Difference between revisions

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
imported>Neal
No edit summary
imported>Jeremy
No edit summary
 
Line 5: Line 5:
===Synopsis===
===Synopsis===


:[cmodel,msg] = compressmodel(model)
:[cmodel,msg] = compressmodel(model,''include'')


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


COMPRESSMODEL will remove any references in a model to excluded variables. This permits the application of the model to new data in which unused variables have been hard-excluded (i.e. previously removed or not collected). Input (model) is the model to compress.
COMPRESSMODEL will remove any references in a model to excluded variables. This permits the application of the model to new data in which unused variables have been hard-excluded (i.e. previously removed or not collected).


Outputs are (cmodel) the compressed model and (msg) which contains any warning messages reported during compression. Although compression will work on most models, some preprocessing methods and some model types may not compress correctly. In these cases, a warning will be given and reported in the output (msg).
====Inputs====
* '''model''' = model to compress
====Optional Inputs====
* '''include''' =  Index range of variables to include. If not provided, compressmodel will keep only those variables which are explictly marked as included in the original data.
 
====Outputs====
* '''cmodel''' = the compressed model
* '''msg''' = any warning messages reported during compression.
 
Although compression will work on most models, some preprocessing methods and some model types may not compress correctly. In these cases, a warning will be given and reported in the output (msg).


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


[[pca]], [[pcr]], [[pls]], [[plsda]]
[[pca]], [[pcr]], [[pls]], [[plsda]]

Latest revision as of 12:12, 22 February 2013

Purpose

Remove references to unused variables from a model.

Synopsis

[cmodel,msg] = compressmodel(model,include)

Description

COMPRESSMODEL will remove any references in a model to excluded variables. This permits the application of the model to new data in which unused variables have been hard-excluded (i.e. previously removed or not collected).

Inputs

  • model = model to compress

Optional Inputs

  • include = Index range of variables to include. If not provided, compressmodel will keep only those variables which are explictly marked as included in the original data.

Outputs

  • cmodel = the compressed model
  • msg = any warning messages reported during compression.

Although compression will work on most models, some preprocessing methods and some model types may not compress correctly. In these cases, a warning will be given and reported in the output (msg).

See Also

pca, pcr, pls, plsda