You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to SharpNEAT
Hello everyone I am having issues with finding the right activation function for the substrate. I centered the data around 0 with -.5 to +.5 being the bounds. I am interested in the auxArgs optional argument which can be used potentially for variations of an activation function for example Linear.
My question is to use auxArgs I first have to make this equal true right?
public bool AcceptsAuxArgs { get { return false; } }
Then here I can define auxArgs: public double Calculate(double x, double[] auxArgs) is this correct? Also is there a limit to how many auxArgs I can define?
Last question I see that other versions of HyperNeat even ES -HyperNeat-agent simulator is setup so that I can put an activation function for hidden layers and one for outputs which I think might be different then the one used for the substrate. Am I correct in this assumption? Basically is the sunbstrate activation function located in lets say walkerbox experiment class that is for the substrate so would adding one tot he output be possible in this version of HyperNeat?