Here is a reproducible example. Few new variables are added to denote correlated, low variance and all zero variables.
library(caret) iris1 <- iris iris1$
