iver56
commented
7 years ago Until novelty search is ready for python, you can simply implement novelty search yourself. All you have to do is write the fitness function so it rewards novelty, as described here
... average distance to the k-nearest neighbors of an individual in behavior space ... http://eplex.cs.ucf.edu/noveltysearch/userspage/#howtoimplement
I'm not sure if it helps, but here's how I implemented it in one of my projects: https://github.com/iver56/cross-adaptive-audio/blob/master/fitness.py#L222
The MultiNEAT download website page says Novely Search is coming to Python soon. Is novelty search ready for python? It would be really cool if you could provide some example on how to use it.