i tried NN modeling soft called Thinks PRO

i tought NN network following math rules:
x*y
x+y

i used this type of network

input layer:
. 2 nodes
. no input processing
hidden layer:
. 2 nodes
. back propogation(BPN) learning
. sigma Pi input function
. linear transfer
output layer:
. 2 nodes
. back propogation(BPN) learning
. dot product input function
. linear transfer

i gave it 7-10 examples. and it works perfect.

[ This Message was edited by: nonama on 2002-09-10 22:35 ]

Cool. Maybe sometime you might want to teach it to look at the source code for a program and determine what language it is written in? That's probably going to be quite easy...

its not a program, its neural network model....