[R] Combine multiple random forests contained in a list

Bert Gunter gunter.berton at gene.com
Sat Jul 27 07:42:22 CEST 2013


I would say that the use of Reduce in this context is bad practice.

from ?Reduce :

"Reduce uses a binary function to successively combine the elements of
a given vector and a possibly given initial value."

combine() is obviously not a binary function. do.call() seems to be
THE appropriate idiom.

-- Bert

On Fri, Jul 26, 2013 at 9:26 PM, arun <smartpink111 at yahoo.com> wrote:
> HI,
> Using the example in ?combine
> library(randomForest)
> rf1 <- randomForest(Species ~ ., iris, ntree=50, norm.votes=FALSE)
>       rf2 <- randomForest(Species ~ ., iris, ntree=50, norm.votes=FALSE)
>       rf3 <- randomForest(Species ~ ., iris, ntree=50, norm.votes=FALSE)
>       rf.all <- combine(rf1, rf2, rf3)
> lst1<- list(rf1,rf2,rf3)
>
> rf.allL<- do.call(`combine`,lst1)
> #or
> rf.allL<- Reduce(`combine,lst1)
>  identical(rf.all,rf.allL)
> #[1] TRUE
> A.K.
>
>
> Is there a quick and easy way to pass randomForest objects contained in a list into the combine() function?
>
> As a result of calling randomForest through lapply(), I now have 10 randomForests in a list (rfors)
>
> I want to combine all 10 of them. Understandably combine(rfors)
> doesn't work as it doesn't recognise the individual forests within the
> list. I have spend quite sometime messing around with unlist(), lapply()
>  and apply() to try and extract the information in a suitable format but
>  to no avail. The only thing that works is combine(rfors[[1]],
> rfors[[2]] ...etc).
>
> This is a bit cumbersome though, not least because the number of
>  random forests I'll need to combine is likely to change. Any sleek and
> elegant solution to this someone can suggest?
>
> Thanks in advance for any help.
>
> Anna
>
> ______________________________________________
> R-help at 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.



-- 

Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374
Website:
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm



More information about the R-help mailing list