R devel - Nov 2005 - Classification Trees and basic Random Forest pkg using tree structures in C

Hello R-devel:

I have written a package, called "woods", that does classification
trees
(R function CT), and currently, only the most basic functionality of
Random Forest, e.g. bagged trees with choices about sample size, with/without
replacement, size of (random) subset of covariates drawn when nodes are 
split.  My reason for writing this is twofold.  First, I wanted to base
this development entirely in C (as others have done), but using data structures
such as a node, pointer to node (for trees), and pointer to pointer of node (for
forests) implemented in C. The algorithm which
does bagging isn't any faster (its 30% slower) than one by Leo Breiman/Adele
Cutler/Andy Liaw/ Matt Weiner. The CT function runs about equally as fast
as Professor Brian Ripley's.

The only interesting feature is that the tree structure has been implemented in
C. Its a neater way to carry stuff around and I am
guessing would make future implementation easier. 

Because of its inherent redundancy from the users standpoint, it isn't
something to send to CRAN. However, I was wondering whether anyone is
interested in a copy?

Grant Izmirlian
NCI

Izmirlian, Grant (NIH/NCI) wrote:
<snipped>
> The only interesting feature is that the tree structure has been
> implemented in C. Its a neater way to carry stuff around and I am 
> guessing would make future implementation easier.
> 
> Because of its inherent redundancy from the users standpoint, it
> isn't something to send to CRAN. However, I was wondering whether
> anyone is interested in a copy?
Hi,

Hmm, why didn't you just post a URL? Incidentally I am actually very
interested in seeing your code. I am working on a project where
the data set is extremely large, but the permuntation of the states of
the data is extremely small. Each piece of data consists of only 4 
states, so stuffing it as an R object (which takes up 32-byte? on
32-bit machines) or even an char vector is quite wasteful; so I
have written a "strange" data.frame where internally it uses only
2-bit for storage. (it is still work-in-process but I have got to
the point of being able to get and set each 2-bit cell now).

Hin-Tak Leung

Hello Hin-Tak:

Thanks for your interest. This is just a short not to tell you and others that
the URL idea is a good one. This will take a few days at our organization.
When its available I will post again to this thread. In the meantime, I will
will send copies directly to those interested. So far, you and one other person.

Regards,

Grant