[R] randomForests predict problem

[Yves Brostaux]

Well, thank you for your answer, but this is not doing the right thing, 
that is predicting the Class value for the test set Soybean[test,]. It 
gives instead prediction for data used for forest computation (ignoring all 
data with NA's) ; 'data' argument is simply ignored as the right name for 
this argument is 'newdata', which still gives the same error when named.

 > length(sb.rf.pred)
[1] 445
 > dim(Soybean[test,])
[1] 150  36
 > dim(Soybean[-test,])
[1] 533  36
 > sb.rf.pred <- predict(sb.rf, newdata=st)
Error in matrix(t1$countts, nr = nclass, nc = ntest) :
         No data to replace in matrix(...)

At 13:13 02/04/03, you wrote:

> > Hello everybody,
> >
> > I'm testing the randomForest package in order to do some simulations and I
> > get some trouble with the prediction of new values. The random forest
> > computation is fine but each time I try to predict values with the newly
> > created object, I get an error message. I thought I was because NA values
> > in the dataframe, but I cleaned them and still got the same error. What am
> > I doing wrong ?
> >
> >  > library(mlbench)
> >  > library(randomForest)
> >  > data(Soybean)
> >  > test <- sample(1:683, 150, replace=F)
> >  > sb.rf <- randomForest(Class~., data=Soybean[-test,])
> >  > sb.rf.pred <- predict(sb.rf, Soybean[test,])
> > Error in matrix(t1$countts, nr = nclass, nc = ntest) :
> >          No data to replace in matrix(...)
>
>
>try
>
>R> test <- sample(1:683, 150, replace=FALSE)
>R>
>R> st <- Soybean[test,]
>R>
>R> sb.rf <- randomForest(Class~., data=Soybean, subset=-test)
>R> sb.rf.pred <- predict(sb.rf, data=st)
>R>
>R> sb.rf.pred[1:10]
>  [1] diaporthe-stem-canker diaporthe-stem-canker diaporthe-stem-canker
>  [4] diaporthe-stem-canker diaporthe-stem-canker diaporthe-stem-canker
>  [7] diaporthe-stem-canker charcoal-rot          charcoal-rot
>[10] charcoal-rot
>19 Levels: 2-4-d-injury alternarialeaf-spot anthracnose ...
>rhizoctonia-root-rot
>
>
>Torsten