Closed lastnod closed 6 years ago
My feeling is that this is a higher-level feature that doesn't belong in our core libraries. Of course anyone is free to build an external genetic algorithm library on top of TF.js primitives, no problem! (Note none of Python TensorFlow, TF Layers, or Keras provide any explicit support for genetic algorithms afaik).
I'll close this as "won't fix" in a day or two if there are no other opinions.
As an FYI we're hoping to tackle this with ml5! See: https://github.com/ml5js/ml5-library/issues/113
Excellent! See also #252.
no problem guy great job by the way. Just a feature request ! :)
Could you implement a crossover function in order to perform reinforcement learning ? (genetic algorithm for me :) ) and obviously with crossover we need a simple mutate function to tune weight, maybe with possibility to put your own tuning function like gaussian distribution.
i hope it's not already implemented !