Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!news.alpha.net!uwm.edu!cs.utexas.edu!howland.reston.ans.net!ix.netcom.com!netcom.com!komodo
From: komodo@netcom.com (Tom Johnson)
Subject: Separate or scalable inputs?
Message-ID: <komodoD6EoJM.1Go@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
X-Newsreader: TIN [version 1.2 PL1]
Date: Sun, 2 Apr 1995 11:42:58 GMT
Lines: 12
Sender: komodo@netcom5.netcom.com

I am working a project that will require over 100 inputs. As this is 
quite large, I was looking for ways to reduce the number of inputs.

Here is my question. If I use 3 distinct binary inputs to represent
animal, vegetable or mineral (I will be actually be using about 50 inputs
with 10 maximum turned on at any one time) can I get just as effective 
results if i use 1 input with 0 to .33 representing animal, .34 to .67 
representing vegetable, etc.?

Any thoughts would be a great time saver and highly appreciated.

Tom
