[R] library kernlab, ksvm(x,y,scaled=T)

[strinz at freenet.de]

Hello,

I encountered the following problem with the parameter scaled in ksvm()
from package kernlab:
[Package kernlab version 0.9-5]

library(kernlab)

> 	svp   			=ksvm(x=mydata,y=y,scaled=T)  
Using automatic sigma estimation (sigest) for RBF or laplace kernel 

> 	svp   			=ksvm(x=mydata,y=y,scaled=F)  
Using automatic sigma estimation (sigest) for RBF or laplace kernel 
Error in drop((scal^2) * crossprod(fitted(ret) - y)/m) : 
	object "scal" not found

> class(mydata)
[1] "matrix"

> range(mydata)
[1]   0 100

what is the problem here ?

best regards
Björn




"Jetzt Handykosten senken mit klarmobil - 14 Ct./Min.! Hier klicken"
http://produkte.shopping.freenet.de/handy_voip_isdn/klarmobil/index.html?pid=730025