Hands-On Neuroevolution with Python

The eval_individual function

This function evaluates the fitness of each individual genome and has the following definition:

def eval_individual(genome, substrate, rt_environment, params):    # Create ANN from provided CPPN genome and substrate    net = NEAT.NeuralNetwork()    genome.BuildESHyperNEATPhenotype(net, substrate, params)    fitness, dist, total_count, false_detetctions = \       rt_environment.evaluate_net(net, max_fitness=MAX_FITNESS)    return fitness, dist, total_count, false_detetctions

It takes the CPPN genome to be evaluated, the substrate configuration, the test environment, and the ES-HyperNEAT hyperparameters as parameters. Using the provided parameters, we create the neural network configuration of the detector ANN and evaluate it against ...