Plot string kernel svm/decision boundary using kernlab

Question

I am using kernlab to learn the basics of kernel SVMs, and am running into some difficulty plotting the results for the model applied to the reuters dataset.

Jean-Philippe Vert has written an excellent tutorial which I am following (http://members.cbio.mines-paristech.fr/~jvert/svn/tutorials/practical/stringkernels/stringkernels.R), but I would like to modify it to include a plot of the actual decision boundary, not just the error. My code looks like this:

library(kernlab)
data(reuters)

y <- rlabels
x <- reuters

kmax <- 20
errspectrum <- numeric(kmax)
errboundrange <- numeric(kmax)

for (k in seq(kmax)) {
  cat('.')
  sk <- stringdot(type="spectrum", length=k, normalized=TRUE)
  svp <- ksvm(x,y,kernel=sk,scale=c(),cross=5)
  plot(svp, data = x, main = k)
  errspectrum[k] <- cross(svp)
  sk <- stringdot(type="boundrange", length=k, normalized=TRUE)
  svp <- ksvm(x,y,kernel=sk,scale=c(),cross=5)
  errboundrange[k] <- cross(svp)    
}

They key line is plot(svp, data = x, main = k)- I would like to plot something like this: https://github.com/topepo/caret/issues/337 but I receive the following error:

Error in sub[, 2] : subscript out of bounds

I'm sure this is something silly on my part but I couldn't find a solution. Any tips would be much appreciated. Thanks!