Neural Net output layer coding

Anyone?

Hi,

if your training data contains a binary label then the output of the Neural Net will be a binominal attribute. The underlying net cannot be changed, but why do you want to do it?

Best regards,
Marius

Thanks for your reply.

I have three classes in my data and as far as i know it can be correctly calculated with only two neurons (binary):

Class 1
N1: 1
N2: 0

Class 2:
N1: 0
N2: 1

Class 3:
N1: 1
N2: 1

Is there any way i can manage to do that?