featureSelectorInit
USE:
[BIG,x,Adrem] = featureSelectorInit(X,y,nbx, nby,nf,nb,initsel,cvsets,k,alpha,gama,flag)
DESCRIPTION:
This
function is almost similar with the function featureSelector
(see help for details) with the exception that it takes supplementary argument initsel, which is a vector whose elements are integers
indicating a initial selection of features to start
with in the feature selection process.
EXAMPLES:
CLASSIFICATION
//load the
iris data set. There are 3 classes and 50 samples in each
> irisdata
// use as
relevance measure both AR and ADC and use (l,r) search with r<>0 by starting with a initial
selection of features 2 and 4.
> [BIG,x,Adrem] = featureSelectorInit (X,y,10,3,2,1,[1 2],3,1,0.1,0.9,1)
REGRESSION
//generate a
5 features matrix X of random numbers. y is generated
in such a way that the importance of features of X in predicting y decreases
from the 1st to 3rd while the remaining 2 are completely
unimportant.
> X=randn(200,5)
> y=(2*X(:,1).^2-X(:,3))./X(:,2)
// Searches
best features using as features relevance measure the ADC and forward selection
only, as nb is set to 0, but feature 3 is taken as
initial selection
> [BIG,x,Adrem] = featureSelectorInit (X,y,5,5,1,0,[3],3,5,0,1,1)
COMMENTS:
Created by Laurentiu Adi
Tarca. Revised 10.07.2003
This help
was created on 12.03.2004