From chuck@psipsy.uct.ac.za Mon Jun 27 20:55:59 EDT 1994 Article: 17479 of comp.ai.neural-nets Xref: glinda.oz.cs.cmu.edu comp.ai.neural-nets:17479 Path: honeydew.srv.cs.cmu.edu!nntp.club.cc.cmu.edu!newsfeed.pitt.edu!gatech!swrinde!ihnp4.ucsd.edu!munnari.oz.au!hippo.ru.ac.za!ucthpx!uctvax.uct.ac.za!chuck.cem.uct.ac.za!chuck Newsgroups: comp.ai.neural-nets Subject: SLUG goes commercial Message-ID: From: chuck@psipsy.uct.ac.za (Charles Marais) Date: Sat, 25 Jun 1994 08:20:49 GMT Organization: University of Cape Town Nntp-Posting-Host: chuck.cem.uct.ac.za X-Newsreader: Trumpet for Windows [Version 1.0 Rev A] Lines: 63 ****************************************************************** ANNOUNCEMENT SLUG4 GOES COMMERCIAL ****************************************************************** Firstly, thanks to all those who have used the last public domain version, SLUG3, as found in NEURAL22.ZIP all over the place(more than 30 sites, at my last archie count). SLUG4 is now a commercial product. It includes a 52 page manual, the entire object oriented neural networks toolkit used to build SLUG4, plus all the Borland Pascal source code. It costs $59, plus $10 shipping. SLUG4 will automatically scale your data, implements epoch based training, allows you to choose transfer functions and randomizes your data on the fly during training. The toolkit provides a flexible and very easy to use paradigm for the construction of ANY neural net, and includes backprop net objects. It also contains : A neuron object A croupier object (for random 'dealing' of data) A dynamic matrix object (used for weights) A dynamic vector object (used for neuron 'slabs' etc) Code for reading and constructing Brainmaker (California Scientific) backprop nets. You can order from Southern Scientific using your Mastercard or VISA card, via FAX (27-21-788-6613). Remember to specify the expiry date. Charles Marais Dr Charles Marais Department of Chemistry University of Cape Town Private Bag Rondebosch chuck@uctvax.uct.ac.za South Africa 7700 chuck@psipsy.uct.ac.za ------------------------------------------------------------------