Hello,
I shall appreciate any help on the problem as mentioned below. I am using Ada
package, which has a predict function that works like predict.rpart. When I run
a piece of code (e.g. pred <- predict(gdis, train[,-1], type = "prob") ) in R,
and type "pred" at the R prompt, I get the probability table such as the one
given below [,1] [,2][1,] 0.2011456 0.7988544[2,] 0.2011456
0.7988544[3,] 0.2011456 0.7988544[4,] 0.2011456 0.7988544[5,] 0.2011456
0.7988544However, when I try to do the same thing (code given below) from Excel
through Rinterface, I am getting "#Rerror". Call Rinterface.RRun("pred <-
predict(gdis, train[,-1], type = 'prob')")Call RInterface.GetArray("pred",
Range("C20", "C24")) I went through the mailing list archive and tried the
following alternatives: 1. Call Rinterface.GetArray("pred[,1]", Range("C20",
"C24"))(Results in #Rerror) 2. Dim cmdString As String cmdString =
"predict(gdis, train[,-1], type = 'prob')" Call
Rinterface.RRun("myOutput<-capture.output(print(try({" & cmdString & "},
silent=TRUE)))") (Geting the following output[1] "Error in predict.rpart(f,
newdata = dat, type = \"prob\") : \n\tobject \"train\" not
found\n"attr(,"class")[1] "try-error" ) 3. Call
Rinterface.GetArray("as.matrix(pred)", Range("C20", "C24"))(Results in
#Rerror)The Ada manual says that the return value of the predict is a matrix of
probability values. I tried even the following, which too resulted in "#Rerror"
4. Call Rinterface.GetArray("pred[1,1]", Range("C20", "C24")) Looking forward
to any useful hint, Sincerely,Sakib
_________________________________________________________________
Download the latest version of Windows Live Messenger NOW!
http://get.live.com/en-ie/messenger/overview
_______________________________________________
Rcom-l mailing list
Rcom-l@mailman.csd.univie.ac.at
http://mailman.csd.univie.ac.at/mailman/listinfo/rcom-l
More information (including a Wiki) at http://rcom.univie.ac.at