Hi,
Does R provide functions for pruning of Additive Regression Tree Models?
Suppose I have grown a large set of trees, say 3000 trees, and now want to find
a 'computationally more economic' approximation of the function defined by the
trees. In the case of a single tree this is pruning. Is there anything similar
for (M)ART?
In case this matters I use the gbm package for model estimation. Intuitively,
at least in the case when the shrinkage parameter is set to a rather small
value, there there should be a way to compress the trees since consecutive
trees are correlated.
Thank you in advance for your help,
Vadim
P.S. If possible, please reply to me directly since I do not follow the r-help
list anymore due to its high traffic
[[alternative HTML version deleted]]
______________________________________________
[email protected] 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.
