Hi, I am working on a project to parallelize the look-ahead for board evaluation. I am looking to switch from pubeval to an evaluator that is zero-sum in order to make things easier. Is there an easy way to extract the neural net code for evaluating race and contact positions from gnubg? If you could point me where to look and things to consider I would greatly appreciate it. Also, if you know of an evaluator from another project that is easier to grab a hold of, I would be willing to use that as well.
Thanks, Eddie _______________________________________________ Bug-gnubg mailing list [email protected] http://lists.gnu.org/mailman/listinfo/bug-gnubg
