[R] svm(e1071) and scaling of weights

I expected, that I will get the same prediction, if I multiply the weights for all classes with a constant factor, but I got different results. Please look for the following code. > library(e1071) > data(Glass, package = "mlbench") > index <- 1:nrow(Glass) > testindex <- sample(index, trunc(lengt

[R] SVM e1071

Dear R-users, Does anyone know how to get margins' information using SVM under package e1071? I have a two class classification problem and I'd like to have, for each input example, the distance of this example to the margin just like it's possible to obtain using C-based SVM-light, for instan