[Rcpp-devel] New package for neural networks on CRAN using Rcpp: RSNNS

Christoph Bergmeir c.bergmeir at decsai.ugr.es
Mon Nov 15 17:08:24 CET 2010


Hello Rcpp developers and other people on the list,

Romain Francois encouraged me to send this email to the list announcing a package my lab recently submitted to CRAN.

http://cran.r-project.org/web/packages/RSNNS/


Here is the package description:

The Stuttgart Neural Network Simulator (SNNS) is a library containing many standard implementations of neural networks. This package wraps the SNNS functionality to make it available from within R. Using the RSNNS low-level interface, nearly all of the functionality and flexibility of SNNS can be accessed. Furthermore, the package contains a convenient high-level interface, so that the most common neural network topologies and learning algorithms integrate seamlessly into R.


Some more remarks:

In particular, we ported SNNS to C++, and use Rcpp for the R-wrapping. As SNNS is a quite comprehensive neural network software, RSNNS has considerably more network topologies, learning functions, activation functions, etc. available as existing solutions within R (e.g. recurrent networks, som, art1, art2, rbf,...). So we hope that the package can significantly enhance and ease the use of neural networks in R.

We developed/ported a bunch of examples and tests that are all available within the package, and can serve as a basis for your own implementations. However, as the library is large and complex and we are not the people who originally developed it, we really encourage feedback and bug reports.

Regards,
Christoph


More information about the Rcpp-devel mailing list