Ncrossval: Difference between revisions
Jump to navigation
Jump to search
imported>Jeremy (Importing text file) |
imported>Jeremy (Importing text file) |
||
| Line 1: | Line 1: | ||
===Purpose=== | ===Purpose=== | ||
Cross-validation for multilinear PLS (NPLS). | Cross-validation for multilinear PLS (NPLS). | ||
===Synopsis=== | ===Synopsis=== | ||
:[press,cumpress,rmsecv,rmsec,cvpred,misclassed] = ncrossval(x,y,rm,cvi,ncomp,out,pre) | :[press,cumpress,rmsecv,rmsec,cvpred,misclassed] = ncrossval(x,y,rm,cvi,ncomp,out,pre) | ||
===Description=== | ===Description=== | ||
Performs cross-validation of NPLS. If two-way unfold-PLS is desired convert input x to two-way x. By default, the data are centered across the first mode, but no scaling is applied. This can be changed by using additional input arguments. | Performs cross-validation of NPLS. If two-way unfold-PLS is desired convert input x to two-way x. By default, the data are centered across the first mode, but no scaling is applied. This can be changed by using additional input arguments. | ||
====INPUTS==== | ====INPUTS==== | ||
* '''x''' = X-block matrix, | * '''x''' = X-block matrix, | ||
* '''y''' = Y-block matrix, and | * '''y''' = Y-block matrix, and | ||
* '''rm''' = regression method (must be 'npl') | * '''rm''' = regression method (must be 'npl') | ||
* '''cvi''' = see CROSSVAL | * '''cvi''' = see CROSSVAL | ||
* '''ncomp''' = maximum number of factors. | * '''ncomp''' = maximum number of factors. | ||
* '''out''' = see CROSSVAL | * '''out''' = see CROSSVAL | ||
* '''pre''' = see CROSSVAL | * '''pre''' = see CROSSVAL | ||
====OUTPUTS==== | ====OUTPUTS==== | ||
* '''See''' CROSSVAL | * '''See''' CROSSVAL | ||
===See Also=== | ===See Also=== | ||
[[crossval]], [[npls]] | [[crossval]], [[npls]] | ||
Revision as of 15:26, 3 September 2008
Purpose
Cross-validation for multilinear PLS (NPLS).
Synopsis
- [press,cumpress,rmsecv,rmsec,cvpred,misclassed] = ncrossval(x,y,rm,cvi,ncomp,out,pre)
Description
Performs cross-validation of NPLS. If two-way unfold-PLS is desired convert input x to two-way x. By default, the data are centered across the first mode, but no scaling is applied. This can be changed by using additional input arguments.
INPUTS
- x = X-block matrix,
- y = Y-block matrix, and
- rm = regression method (must be 'npl')
- cvi = see CROSSVAL
- ncomp = maximum number of factors.
- out = see CROSSVAL
- pre = see CROSSVAL
OUTPUTS
- See CROSSVAL