Dear R community,

I’m trying to use the “party” package to view variables that are 
important in my data frame (FIM_RS4). My data has a combination of ordinal, 
nominal, categorical, and continuous data. I’ve written my code below, with 
my predictor variables being 2:36, and my response is 37 (Mussel). If anyone 
could point me in the direction of how to get rid of this error message, I 
would very much appreciate the guidance.

FIM_RS4<-read.csv("FIM_RS_4.csv", header=TRUE, stringsAsFactors=T) 

#all ordinal variables have been ordered in my complete code as well

mycontrols <- cforest_control(ntree=5000,mtry=4, minsplit=5)

cforest.rf <- cforest(as.factor(Mussel)~., data=FIM_RS4[,c(37,2:36)], 
controls=mycontrols)

# indata=FIM_RS4[,2:36]

# rf.model=cforest.rf

temp <- varimp.function(indata=FIM_RS4[,2:6], 
rf.model=cforest.rf,num.iterations=5, seedset = 2000,condition.varimp = 
T,write_csv=F)

Ø  Error in model@fit(data, ...) : error code 1 from Lapack routine 'dgesdd'

 #The reason I’m assigning a varimp.function to temp is to run iterations in 
the next piece of code.

#Same error occurs if I do not assign the iteration function varimp.function to 
temp, and simply do:

varimp(cforest.rf, conditional = TRUE)



Ø  Error in model@fit(data, ...) : error code 1 from Lapack routine 'dgesdd'



Many thanks in advance,

Rox
        [[alternative HTML version deleted]]

______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Reply via email to