Hi, I am using R's "kernlab" package. I am doing classification using ksvm(). My train data is 2455obs 5397variables and no NA data in there. And my code is: train$tag <- as.factor(train$tag) k<-ksvm(tag~.,train,kernel="laplacedot",C=100,cross=10) But it returns the following error: Error in votematrix[i, ret < 0] <- votematrix[i, ret < 0] + 1 : NAs are not allowed in subscripted assignments In addition: Warning messages: 1: In sqrt(-round(2 * x[lowerl:upperl, ] %*% t(y) - dotbb - dota[lowerl:upperl] %*% : NaNs produced 2: ...(the same as 1) ... But when I choose another kernel, such as rbfdot, vanilladot, polydot and tanhdot, it returns the ksvm model to me. So can you tell me where is wrong? Thank you very much for your time and attention. Sincerely, Rincy [[alternative HTML version deleted]]