search

CharlesHEaster / NeuralNet

My attempt to create a framework to create and train neural networks through fitness evolution.
0 stars 0 forks source link

Learning rate and evolve rate variables #4

Closed CharlesHEaster closed 3 years ago

CharlesHEaster commented 4 years ago

Make them have a default value (lord knows that trial constructor doesn't need any more arguments) but make them changeable through a command. Talking here of the amount the weights change with morph, and the number of them which change.

CharlesHEaster commented 3 years ago

Store it as trial instance variables. Pass through parameters in network.morph() and node.morph()

CharlesHEaster commented 3 years ago

done and dusted