I am trying to rfImpute into a training data set that is only 34526 by 43. I am using the latest R and RStudio (I don't think the RStudio version matters) and there seems to be an error for random forests of long vectors not being supported but I haven't even hit that limit. One answer for the random forest was that you should just split up the random forest model from 1:600000 and 600000-1200000. And I'm not even close to that many variables. I reduced the number of variables in the model to see if that would work but even with 9 variables, it would still shoot out that long vectors are not supported in C. Below is the code that I am running. Is there any way around this? I ran the rfImpute on another data set that is 14895 by 40 which worked.
cpt_prim_cups_imputed <- rfImpute(Censor ~ ., data=cpt_prim_cups_train_test)