R help - Jun 2011 - ipred: possible to use svm in errorest?

Dear all,


is it possible to run the support vector machine command "svm" from
the
package e1071 within the errorest function from the library ipred?
It works fine for lda and rda but I get an error message (see below)

Thank you for your help.
Best wsihes, Daneil


# Classification

data(iris)

# force predict to return class labels only
mypredict.lda <- function(object, newdata)
predict(object, newdata = newdata)$class
> # 10-fold cv of LDA for Iris data
> errorest(Species ~ ., data=iris, model=svm, estimator = "cv",
predict=
> mypredict.lda)
Error in predict(object, newdata = newdata)$class :
    $ operator is invalid for atomic vectors

 

--

On 21.06.2011 13:56, Daniel Stahl wrote:
> Dear all,
>
>
> is it possible to run the support vector machine command "svm"
from the
> package e1071 within the errorest function from the library ipred?
> It works fine for lda and rda but I get an error message (see below)
>
> Thank you for your help.
> Best wsihes, Daneil
>
>
> # Classification
>
> data(iris)
>
> # force predict to return class labels only
> mypredict.lda<- function(object, newdata)
> predict(object, newdata = newdata)$class
>
>> # 10-fold cv of LDA for Iris data
>> errorest(Species ~ ., data=iris, model=svm, estimator = "cv",
predict>> mypredict.lda)
> Error in predict(object, newdata = newdata)$class :
>      $ operator is invalid for atomic vectors

The point of specifying a predict function is to tell errorest how to 
get the classes from predict(). Since predict.svm already returns the 
classes you can omit it:
errorest(Species ~ ., data=iris, model=svm, estimator = "cv")

Uwe Ligges

>
>
> --
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.