R help - Aug 2001 - rpart 3.1.0 bug?

I just updated rpart to the latest version (3.1.0).  There are a number of 
changes between this and previous versions, and some of the code I've been 
using with earlier versions (e.g. 3.0.2) no longer work.

Here is a simple illustration of a problem I'm having with xpred.rpart.

iris.test.rpart<-rpart(iris$Species~., data=iris[,1:4], 
parms=list(prior=c(0.5,0.25, 0.25)))
+ )
 > xpred.rpart(iris.test.rpart)
Error in as.double.default(fit$parms) : (list) object cannot be coerced to 
vector type 14
 >
The problem seems to be with the parms parameter in the call to rpart.  If 
I leave parms out of the fit statement, xpred.rpart works fine.

I can't deduce from the help files or the readme what has changed that 
causes this error (I also can't exactly figure out what the error is).  The 
example use of xpred.rpart works fine, but doesn't have a parms argument in 
the rpart function call.

Any help appreciated.


Dr. Marc R. Feldesman
email:  feldesmanm at pdx.edu
email:  feldesman at attglobal.net
fax:    503-725-3905

"Don't know where I'm going.
Don't like where I've been.
There may be no exit.
But hell, I'm going in."  Jimmy Buffett

Powered by Superchoerus - the 700 MHz Coppermine Box

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Extra right paren got copied into the first message by mistake.  The 
original did not have the extra and the code below does not work.

Also, if you try to use xpred.rpart with "fit2" from the rpart help
page,
it gives the same error as below (i.e.
 >xfit <-xpred.rpart(fit2) )
 > Error in as.double.default(fit$parms) : (list) object cannot be coerced 
to vector type 14



 >iris.test.rpart<-rpart(iris$Species~., data=iris[,1:4], 
parms=list(prior=c(0.5,0.25, 0.25)))
 > xpred.rpart(iris.test.rpart)
 > Error in as.double.default(fit$parms) : (list) object cannot be coerced 
to vector type 14




At 01:59 PM 8/12/01, Marc R. Feldesman wrote:
 >I just updated rpart to the latest version (3.1.0).  There are a number of
 >changes between this and previous versions, and some of the code I've
been
 >using with earlier versions (e.g. 3.0.2) no longer work.
 >
 >Here is a simple illustration of a problem I'm having with xpred.rpart.
 >
 >iris.test.rpart<-rpart(iris$Species~., data=iris[,1:4],
 >parms=list(prior=c(0.5,0.25, 0.25)))
 >+ )
 > > xpred.rpart(iris.test.rpart)
 >Error in as.double.default(fit$parms) : (list) object cannot be coerced to
 >vector type 14
 > >
 >The problem seems to be with the parms parameter in the call to rpart.  If
 >I leave parms out of the fit statement, xpred.rpart works fine.
 >
 >I can't deduce from the help files or the readme what has changed that
 >causes this error (I also can't exactly figure out what the error is). 
The
 >example use of xpred.rpart works fine, but doesn't have a parms
argument in
 >the rpart function call.
 >
 >Any help appreciated.
 >
 >
 >Dr. Marc R. Feldesman
 >email:  feldesmanm at pdx.edu
 >email:  feldesman at attglobal.net
 >fax:    503-725-3905
 >
 >"Don't know where I'm going.
 >Don't like where I've been.
 >There may be no exit.
 >But hell, I'm going in."  Jimmy Buffett
 >
 >Powered by Superchoerus - the 700 MHz Coppermine Box
 >
 >-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
 >r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
 >Send "info", "help", or "[un]subscribe"
 >(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
 >_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
 >


Dr. Marc R. Feldesman
email:  feldesmanm at pdx.edu
email:  feldesman at attglobal.net
fax:    503-725-3905

"Don't know where I'm going.
Don't like where I've been.
There may be no exit.
But hell, I'm going in."  Jimmy Buffett

Powered by Superchoerus - the 700 MHz Coppermine Box

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

There seems to be a simple, if inelegant, fix for the xpred.rpart problem.

In the file xpred.rpart, I found that changing the following line fixed the 
bug:


costs <- fit$call$costs
     if (is.null(costs))
         costs <- rep(1, nvar)
         parms<-as.double(fit$parms)

Change last line to:
     parms <- as.vector(as.double(unlist(fit$parms)))

This works with all my test examples.  I have no idea how this ramifies to 
other areas of the code that I don't use.  YMMV.


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._