Hello,
I'm trying to use the party package function varimp() to get
conditional variable importance measures, as I'm aware that some of my
variables are correlated. However I keep getting error messages (such
as the example below). I get similar errors with three separate
datasets that I'm using. At a guess it might be something to do with
the very large number of variables (e.g. 23 variables, 250 or so data
points) but I was wondering if anyone had any other ideas. It works
fine for regular variable importance calculation.
Code:
biomass.cf<-cforest(Total.biomass ~ .,
data=biomass, control=cforest_unbiased(ntree=2500, mtry=8))
biomass.cf.vi<-varimp(biomass.cf, conditional=TRUE)
Error:
Error in if (node[[5]][[1]] == variableID) cp <- node[[5]][[3]] :
argument is of length zero
In addition: Warning messages:
1: In matrix(as.logical(cl), nrow = nlevels(x)) :
data length [2] is not a sub-multiple or multiple of the number of rows [17]
2: In matrix(as.logical(cl), nrow = nlevels(x)) :
data length [2] is not a sub-multiple or multiple of the number of rows [17]
3: In matrix(as.logical(cl), nrow = nlevels(x)) :
data length [2] is not a sub-multiple or multiple of the number of rows [17]
4: In matrix(as.logical(cl), nrow = nlevels(x)) :
data length [2] is not a sub-multiple or multiple of the number of rows [17]
Many thanks,
Meghann Mears, PhD student
University of Sheffield
Department of Animal & Plant Sciences
