Error plotting SVM classification results for the spam dataset

Question

I am having problem with plotting results of SVM classification for the spam dataset from kernlab package..

Code:

require(kernlab)

data(spam)

index <- sample(1:dim(spam)[1])
spamtrain <- spam[index[1:floor(dim(spam)[1]/2)], ]
spamtest <- spam[index[((ceiling(dim(spam)[1]/2)) + 1):dim(spam)[1]], ]

filter <- ksvm(type~.,data=spamtrain,kernel="rbfdot",kpar=list(sigma=0.05),C=5,cross=3)

plot(filter, data=spamtrain)

Error:

Error in `[.data.frame`(expand.grid(lis), , labels(terms(x))) : 
  undefined columns selected

Yes, it should be `plot(filter, data=spamtrain)` - I made a typo in the question. Thanks for pointing this out. — Anna Stępień, May 19 '13 at 11:25

Sarkom · Answer 1 · 2014-06-13T18:40:07.713

1

From what I can tell, plot.ksvm (which is called when you use plot on a ksvm object) can only plot objects with two features, i.e., the data matrix can only have two columns. Unfortunately, this is not mentioned anywhere in the documentation, and I just spent a couple of hours trying to understand what was going wrong. I don't know if this is planned to be fixed in the future.

edited Jun 13 '14 at 18:40

answered Jun 13 '14 at 18:34

Sarkom

823
10
12

score 0 · Answer 2 · answered May 19 '13 at 13:00

0

You should use plot from package kernlab so that it is supporting format of ksvm objects. kernlab::plot(object, data=NULL, grid = 50, slice = list())

More info in doc on kernab plot here: http://cran.r-project.org/web/packages/kernlab/kernlab.pdf

answered May 19 '13 at 13:00

kiriloff

25,609
37
148
229

Error plotting SVM classification results for the spam dataset

2 Answers2