home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai.neural-nets
- Path: sparky!uunet!wupost!tulane!rouge!darwin.sura.net!newshost.LaTech.edu!quasar!jmains
- From: jmains@engr.latech.edu (John P. Mains)
- Subject: Neural Net PROBLEM for you Experts
- Message-ID: <1993Jan4.025010.3113@engr.LaTech.edu>
- Sender: usenet@engr.LaTech.edu
- Nntp-Posting-Host: quasar
- Reply-To: jmains@engr.latech.edu
- Organization: Louisiana Tech University
- Distribution: usa
- Date: Mon, 4 Jan 1993 02:50:10 GMT
- Lines: 14
-
- I understand only the basics of neural nets and unfortunately don't have the
- time to do an in depth study of what I need. Therefore I give you this problem (please email me)
-
- I have 50-60 inputs
- I need a neural net capable of recognizing about 40,000 to 50,000 patterns.
- I would prefer real time processing to accuracy, or as close as possible.
-
- What I need to know is the kind of net I need, and as much specifics about it you can give me.
- If you know of a better way to do this other than nets, input would be appreciated.
- -------------
- Thanx
- JP
- jmains@engr.latech.edu
-
-