Newsgroups: comp.ai.neural-nets,comp.lang.smalltalk
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornellcs!travelers.mail.cornell.edu!news.tc.cornell.edu!news.cac.psu.edu!news.pop.psu.edu!hudson.lm.com!newsfeed.pitt.edu!gatech!hookup!noc.tor.hookup.net!locon!mtnlake.com!kulinek
From: kulinek@mtnlake.com (Eugene Kulinek)
Subject: Re: Looking for PD neural net Smalltalk code
X-Newsreader: Trumpet for Windows Version 1.0 Rev A
Sender: news@mtnlake.com (News admin)
Nntp-Posting-Host: 204.225.173.103
Organization: Moutain Lake Software Corporation
Lines: 30
Message-ID: <kulinek.11.00169385@mtnlake.com>
References:  <D96InB.JAF@festival.ed.ac.uk>
Date: Tue, 30 May 1995 02:34:25 GMT
Xref: glinda.oz.cs.cmu.edu comp.ai.neural-nets:24390 comp.lang.smalltalk:24086

In article <D96InB.JAF@festival.ed.ac.uk> olly@festival.ed.ac.uk (O Morgan) writes:
>From: olly@festival.ed.ac.uk (O Morgan)
>Subject: Looking for PD neural net Smalltalk code
>Date: Fri, 26 May 1995 09:35:33 GMT

>Can anyone point me to any Neural net code written in Smalltalk? 
>I'm looking for any classification algorithms, starting with Backprop or
>more efficient, radial basis functions, and Self Organising Map.

>I'm using Digitalk Smalltalk V, version 2.0, but I'd be interested to
>see code for any platform.

>Thanks for any help

>----------------------------------------------------------------------------
> Olly Morgan @ Scottish Agricultural College,  Edinburgh EH9 3JG, Scotland
>               Tel: (+44 131) 535 4128         E.Mail: O.Morgan@ed.ac.uk
>----------------------------------------------------------------------------

I wrote a ST code for BP net. It is similar to C++ version described in Blum's 
book. I wrote it in IBM ST but I do not think there is much difference with 
Digitalk since I had no UI. 

My main concern was speed - slow. Digitalk ST is faster, however. Also, I had
some Transcript messages and these were slowing down even more.

I can briefly describe the design if you like. This was my main interest in 
this excercise. 

Eugene
