My small personal implementation of NEAT (NeuroEvolution of Augmenting Topologies).
Work in progress.
The only thing needed by neat.py
is a fitness function.
This fitness function recieves a list of individuals (i.e. a population) and it is expected to alter the fitness of those individuals.
Each individual is nothing but a dictionary (hashmap) and
includes a fitness
key.
Check the xor.py
file for an example. You can test it out by
simply running python3 xor.py
.