I want to create a new column that will label a certain amount of columns 1 and the remaining number of columns 0. If I want 20% of my columns to have this option I can get close by taking:
test$Rand_Num <- sample(1:5, nrow(test), replace=TRUE) test$Output <- ifelse(test$Rand_Num==1,1,0)
However I would like to be able to say if I had 1000 columns then 200 are randomly 1 and the rest are labeled as 0 that I can quickly change to 30%, etc.. for different scenarios.
Thanks!
If you want to randomly select columns such that 20% (or some other percentage) of columns are randomly selected, you could return a vector of selected columns (for input data.frame df) with
p <- 0.2 # change me!
nselect <- round(p*ncol(df), 0)
whichcolumns <- sample(1:ncol(df), nselect)
To turn this into a vector of zeroes and ones, you could do something like
whichcolumns_01 <- rep(0, ncol(df))
whichcolumns_01[whichcolumns] <- 1
Admittedly, this has some duct tape in it, but it should work.
