This is an announcement for a relatively new Tcl project: tcl-fann

Tcl-fann enables artificial neural networks processing in Tcl. It is 
using the FANN (Fast Artificial Neural Networks) library underneath.

Tcl-fann supports:
  * Fully, sparse and shortcut type connected multi-layer networks.
  * Backpropagation training (RPROP, Quickprop, Batch, Incremental)
  * Evolving topology training which dynamically builds and trains the 
ANN (Cascade)
  * Versatile (possible to adjust many parameters and features on-the-fly)
  * Several different activation functions implemented (including 
stepwise linear functions for that extra bit of speed)
  * Saving and loading of entire ANNs
  * Cross-platform (linux/unix & MS Windows (not tested))

Dependencies:
  * Tcl 8.5 or greater.
  * FANN Library (http://leenissen.dk/fann/) Version 2.2.0 or greater.


Platform: Linux/Unix & MS Windows
Home page: http://sourceforge.net/projects/tcl-fann/
Man page: http://tcl-fann.sourceforge.net/
Author: Alexandros Stergiakis