[R] predict.rpart and large datasets

Chris Wright chris at rapid.msu.montana.edu
Wed Nov 6 18:09:34 CET 2002

Hi folks,

I'm getting the following error message when using predict.rpart on a
large dataset (say 100,000 records) within Splus:

Terminating S Session: Signal: bad address signal

Has anyone run into this error, discovered a work-around, etc.?

More generally, I'm trying to use predict.rpart to generate predictions
from very large, geographic datasets (we're talking all of Yellowstone
National Park, on the order of 20 million pixels).  Has anyone come up
with a reasonable method to process big datasets using predict.rpart?
I've been experimenting with a C program that sequentially calls
predict.rpart to process smaller subsets of a large dataset using the
call_S() Splus function, but it's really slow.  I would appreciate any


Chris Wright

p.s. I would like to subscribe to the r-help list 

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

More information about the R-help mailing list