This function is an essential part of the population fitness evaluation, and it is invoked from the eval_genomes function, discussed earlier, to evaluate the maze-solving performance of each genome in the population.
The evaluation of the individual genome as a maze-solving agent through the maze navigation simulation is as follows:
def eval_individual(genome_id, genome, genomes, n_items_map, config): n_item = archive.NoveltyItem( generation=trial_sim.population.generation, genomeId=genome_id) n_items_map[genome_id] = n_item maze_env = copy.deepcopy(trial_sim.orig_maze_environment) control_net = neat.nn.FeedForwardNetwork.create(genome, config) goal_fitness = maze.maze_simulation_evaluate( env=maze_env, ...
