[mlpack-git] [mlpack/mlpack] NeuralEvolution - implemented gene, genome (#686)

Marcus Edel notifications at github.com
Wed Jul 13 10:20:11 EDT 2016


> +                           genome2.aLinkGenes[idx].Weight() * linkEnabledInGenome2);
> +        ++coincident;
> +      }
> +    }
> +
> +    deltaW = deltaW / coincident;
> +    return deltaW;
> +  }
> +
> +  // Whether two genome belong to same species or not.
> +  bool IsSameSpecies(Genome& genome1, Genome& genome2) {
> +    double deltaD = Disjoint(genome1, genome2);
> +    double deltaW = WeightDiff(genome1, genome2);
> +    double delta = aCoeffDisjoint * deltaD + aCoeffWeightDiff * deltaW;
> +
> +    if (delta < aCompatThreshold) {

Where does this ``aCompatThreshold = 1.0``  parameter come from? Depending on the calculation of the disjoint, excess, and weight difference this could be different.

---
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/mlpack/mlpack/pull/686/files/ff45785b8669f82beb38ef5392f48bcdf5c83f6a#r70633729
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack-git/attachments/20160713/6047140b/attachment.html>


More information about the mlpack-git mailing list