[R] sampsize in Random Forests

[Federico]

Dear Naiara and Andy,

My strategy in cases with unbalanced data is:

tmp <- as.vector(table(factors));
num_clases <- length(tmp);
min_size <- tmp[order(tmp,decreasing=FALSE)[1]];
vector_for_sampsize <- rep(min_size,num_clases);

Then:
randomForest(..., y=factors, sampsize=vector_for_sampsize)

I hope this helps,

Federico



On 10 mar, 17:00, "Liaw, Andy" <andy_l... at merck.com> wrote:
> Are you sure there are 100 sites in your data?  Here's an example:
>
> R> library(randomForest)randomForest4.5-23
> Type rfNews() to see new features/changes/bug fixes.
> R> f <- factor(sample(1:4, nrow(iris), replace=TRUE))
> R> rf1 <-randomForest(iris[1:4], iris[[5]], strata=f,sampsize=rep(5,
> nlevels(f)))
> R> rf1
>
> Call:
>  randomForest(x = iris[1:4], y = iris[[5]], strata = f,sampsize=
> rep(5,      nlevels(f)))
>                Type of random forest: classification
>                      Number of trees: 500
> No. of variables tried at each split: 2
>
>         OOB estimate of  error rate: 4.67%
> Confusion matrix:
>            setosa versicolor virginica class.error
> setosa         50          0         0        0.00
> versicolor      0         47         3        0.06
> virginica       0          4        46        0.08
>
>
>
> > -----Original Message-----
> > From: r-help-boun... at r-project.org
> > [mailto:r-help-boun... at r-project.org] On Behalf Of Naiara Pinto
> > Sent: Sunday, March 09, 2008 5:19 PM
> > To: r-h... at r-project.org
> > Subject: [R]sampsizein Random Forests
>
> > Hi all,
>
> > I have a dataset where each point is assigned to a class A, B, C, or
> > D. Each point is also assigned to a study site. Each study site is
> > coded with a number ranging between 1-100. This information is stored
> > in the vector studySites.
>
> > I want to run randomForests using stratified sampling, so I
> > chose the option
> > strata = factor(studySites)
>
> > But I am not sure how to control the number of samples taken from each
> > study site. I tried to use 10 points from each study site:
> > mySampSize = rep(10, 100)
>
> > So my function call looks like:
> > RF =randomForest(myClass~., data=myData, mtry=5, importance=TRUE,
> > strata = factor(studySites),sampsize=mySampSize)
>
> > ButrandomForestgives me the following error:
> > Error inrandomForest.default(m, y, ...) :
> >sampsizecan not be larger than class frequency
>
> > Does anybody have any idea why this happens?
>
> > Thank you very much,
>
> > Naiara.
>
> > ______________________________________________
> > R-h... 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.
>
> ------------------------------------------------------------------------------
> Notice:  This e-mail message, together with any attachme...{{dropped:15}}
>
> ______________________________________________
> R-h... at r-project.org mailing listhttps://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guidehttp://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.