Hands-On Neuroevolution with Python

The evaluate_individual_solution function implementation

This is the function that evaluates the performance of a particular maze solver against the maze environment simulator. It is implemented as follows:

1. First, we create the phenotype ANN of the maze solver and use it as a controller to guide the robot through the maze:

    n_item = archive.NoveltyItem(generation=generation,                                  genomeId=genome_id)    # run the simulation    maze_env = copy.deepcopy(robot.orig_maze_environment)    multi_net = NEAT.NeuralNetwork()    genome.BuildPhenotype(multi_net)    depth = 8    try:        genome.CalculateDepth()        depth = genome.GetDepth()    except:        pass    control_net = ANN(multi_net, depth=depth)    distance = maze.maze_simulation_evaluate(        env=maze_env, net=control_net,  time_steps=SOLVER_TIME_STEPS, ...