The order doesn't really matter. If you are activating the NN every time step, the output might be 0 for the first 1-3 steps only, but that shouldn't be apetar_chervenski
My implementation represents the neurons as callable functions, and I do a sorting on the basis of activation sequence, so in effect, RubyNEAT is also a typelordalveric
Thank you both for your input. Right now I'm just trying to get the XOR experiment working. One of the problems I'm trying to overcome is the execution orderbbjprizes
The intent is to share ideas, questions, and comments regarding the implementation, theory, and possible variations and future directions of NEAT.
- Artificial Intelligence
- Aug 24, 2003
- This is a restricted group.
- Attachments are permitted.
- Members cannot hide email address.
- Listed in Yahoo Groups directory.
- Membership requires approval.
- Messages are not moderated.
- All members can post messages.