\name{ClfFs}
\alias{ClfFs}
\title{Classifier with Feature Selection}
\description{Apply a sequence of feature selection provide
  in the constructor parameters before applying the classifier it-self.
}

\usage{
}

\arguments{
  \item{}{}
  \item{}{}
\value{

} 
\seealso{

}
\examples{
lml()
mlData=MlData(file='small.csv',grpCol='class',subjCol='subject')

## 1. Filters
fsSvm=ClfFs(fs1=FsFilterTest(test="t.test",N=2), clf=ClfSvm())
oSvm=fsSvm$learn(mlData)$predict(mlData)

fsLda=ClfFs(fs1=FsFilterTest(test="t.test",N=2), clf=ClfLda())
oLda=fsLda$learn(mlData)$predict(mlData)

## 2. Filters is used to rank features and plot results
## along the number of features with cross validation
## => See CvResults

## 3. Wrappers
lml()
mlData=MlData(file='small.csv',grpCol='class',subjCol='subject')
wrapLda=ClfFs(fs1=FsWrapperSFS(objFunc=ManovaFstat()), clf=ClfLda())
wrapLda$learn(mlData)
wrapLda$predict(mlData)

## 4. Filter + wrapper
lml()
mlData=MlData(file='medium_scramble.csv',grpCol='class',subjCol='subject')
filtWrapLda=ClfFs(
 fs1=FsFilterTest(test="t.test",FDR=.2),
 fs2=FsWrapperSFS(objFunc=ManovaFstat()),
 clf=ClfLda())
wrapLda$learn(mlData)$predict(mlData)
}