Python neat.Checkpointer() Examples

def test_parallel():
    """Test parallel run using ParallelEvaluator (subprocesses)."""
    # Load configuration.
    local_dir = os.path.dirname(__file__)
    config_path = os.path.join(local_dir, 'test_configuration')
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_path)

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(VERBOSE))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(1, 5))

    # Run for up to 19 generations.
    pe = neat.ParallelEvaluator(1 + multiprocessing.cpu_count(), eval_dummy_genome_nn)
    p.run(pe.evaluate, 19)

    stats.save() 

def evaluation():
    p = neat.Checkpointer.restore_checkpoint('neat-checkpoint-%i' % CHECKPOINT)
    winner = p.run(eval_genomes, 1)     # find the winner in restored population

    # show winner net
    node_names = {-1: 'In0', -2: 'In1', -3: 'In3', -4: 'In4', 0: 'act1', 1: 'act2'}
    visualize.draw_net(p.config, winner, True, node_names=node_names)

    net = neat.nn.FeedForwardNetwork.create(winner, p.config)
    while True:
        s = env.reset()
        while True:
            env.render()
            a = np.argmax(net.activate(s))
            s, r, done, _ = env.step(a)
            if done: break 

def run():
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation, CONFIG)
    pop = neat.Population(config)

    # recode history
    stats = neat.StatisticsReporter()
    pop.add_reporter(stats)
    pop.add_reporter(neat.StdOutReporter(True))
    pop.add_reporter(neat.Checkpointer(5))

    pop.run(eval_genomes, 10)       # train 10 generations

    # visualize training
    visualize.plot_stats(stats, ylog=False, view=True)
    visualize.plot_species(stats, view=True) 

def main():
    local_dir = os.path.dirname(__file__)
    config = Config(neat.DefaultGenome, neat.DefaultReproduction,
                    neat.DefaultSpeciesSet, neat.DefaultStagnation,
                    os.path.join(local_dir, 'train_config.txt'))
    config.save_best = True
    config.checkpoint_time_interval = 3

    pop = population.Population(config)
    stats = neat.StatisticsReporter()
    pop.add_reporter(stats)
    pop.add_reporter(neat.StdOutReporter(True))
    pop.add_reporter(neat.StatisticsReporter())
    pop.add_reporter(neat.Checkpointer(2))
    winner = pop.run(eval_fitness, 100)
    with open('winner.pkl', 'wb') as f:
        pickle.dump(winner, f) 

def _run(self, config_file, n):
        config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                             neat.DefaultSpeciesSet, neat.DefaultStagnation,
                             config_file)
        p = neat.Population(config)
        p.add_reporter(neat.StdOutReporter(True))
        p.add_reporter(neat.Checkpointer(5))
        stats = neat.StatisticsReporter()
        p.add_reporter(stats)
        print("loaded checkpoint...")
        winner = p.run(self._eval_genomes, n)
        win = p.best_genome
        pickle.dump(winner, open('winner.pkl', 'wb'))
        pickle.dump(win, open('real_winner.pkl', 'wb'))

        visualize.draw_net(config, winner, True)
        visualize.plot_stats(stats, ylog=False, view=True)
        visualize.plot_species(stats, view=True) 

def test_threaded_evaluation():
    """Tests a neat evolution using neat.threaded.ThreadedEvaluator"""
    # Load configuration.
    local_dir = os.path.dirname(__file__)
    config_path = os.path.join(local_dir, 'test_configuration')
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_path)

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(True))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(1, 5))

    # Run for up to 19 generations.
    pe = neat.ThreadedEvaluator(4, eval_dummy_genome_nn)
    p.run(pe.evaluate, 19)

    stats.save() 

def test_run_nn_recurrent():
    """Basic test of nn.recurrent function."""
    # Load configuration.
    local_dir = os.path.dirname(__file__)
    config_path = os.path.join(local_dir, 'test_configuration')
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_path)
    config.feed_forward = False

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(VERBOSE))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(1, 5))

    # Run for up to 19 generations.
    p.run(eval_dummy_genomes_nn_recurrent, 19)

    stats.save() 

def _run(self, config_file, n):
        config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                             neat.DefaultSpeciesSet, neat.DefaultStagnation,
                             config_file)
        # p = neat.Population(config)
        p = neat.Checkpointer.restore_checkpoint(self.file_name)
        p.add_reporter(neat.StdOutReporter(True))
        p.add_reporter(neat.Checkpointer(5))
        stats = neat.StatisticsReporter()
        p.add_reporter(stats)
        print("loaded checkpoint...")
        winner = p.run(self._eval_genomes, n)
        win = p.best_genome
        pickle.dump(winner, open('winner.pkl', 'wb'))
        pickle.dump(win, open('real_winner.pkl', 'wb'))

        visualize.draw_net(config, winner, True)
        visualize.plot_stats(stats, ylog=False, view=True)
        visualize.plot_species(stats, view=True) 

def run(config_file):
    # Load configuration.
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_file)

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(True))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(5))

    # Run for up to 300 generations.
    winner = p.run(eval_genomes, 300)

    # Display the winning genome.
    print('\nBest genome:\n{!s}'.format(winner))

    # Show output of the most fit genome against training data.
    print('\nOutput:')
    winner_net = neat.nn.FeedForwardNetwork.create(winner, config)
    for xi, xo in zip(xor_inputs, xor_outputs):
        output = winner_net.activate(xi)
        print("input {!r}, expected output {!r}, got {!r}".format(xi, xo, output))

    node_names = {-1:'A', -2: 'B', 0:'A XOR B'}
    visualize.draw_net(config, winner, True, node_names=node_names)
    visualize.plot_stats(stats, ylog=False, view=True)
    visualize.plot_species(stats, view=True)

    p = neat.Checkpointer.restore_checkpoint('neat-checkpoint-4')
    p.run(eval_genomes, 10) 

def run(config_file):
    # Load configuration.
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_file)

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(True))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(50))

    # Run for up to 300 generations.
    winner = p.run(eval_genomes, 300)

    # Display the winning genome.
    print('\nBest genome:\n{!s}'.format(winner))

    # Show output of the most fit genome against training data.
    print('\nOutput:')
    winner_net = neat.nn.FeedForwardNetwork.create(winner, config)
    for xi, xo in zip(xor_inputs, xor_outputs):
        output = winner_net.activate(xi)
        print("input {!r}, expected output {!r}, got {!r}".format(xi, xo, output))

    node_names = {-1:'A', -2: 'B', 0:'A XOR B'}
    visualize.draw_net(config, winner, True, node_names=node_names)
    visualize.plot_stats(stats, ylog=False, view=True)
    visualize.plot_species(stats, view=True)

    p = neat.Checkpointer.restore_checkpoint('neat-checkpoint-49')
    p.run(eval_genomes, 10) 

def run_primary(addr, authkey, generations):
    """Starts a DistributedEvaluator in primary mode."""
    # Load configuration.
    local_dir = os.path.dirname(__file__)
    config_path = os.path.join(local_dir, 'test_configuration')
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_path)

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(True))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(max(1, int(generations / 3)), 5))

    # Run for the specified number of generations.
    de = neat.DistributedEvaluator(
        addr,
        authkey=authkey,
        eval_function=eval_dummy_genome_nn,
        mode=MODE_PRIMARY,
        secondary_chunksize=15,
    )
    print("Starting DistributedEvaluator")
    sys.stdout.flush()
    de.start()
    print("Running evaluate")
    sys.stdout.flush()
    p.run(de.evaluate, generations)
    print("Evaluated")
    sys.stdout.flush()
    de.stop(wait=5)
    print("Did de.stop")
    sys.stdout.flush()

    stats.save() 

def test_run_ctrnn():
    """Basic test of continuous-time recurrent neural network (ctrnn)."""
    # Load configuration.
    local_dir = os.path.dirname(__file__)
    config_path = os.path.join(local_dir, 'test_configuration')
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_path)
    config.feed_forward = False

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(VERBOSE))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(1, 5))

    # Run for up to 19 generations.
    p.run(eval_dummy_genomes_ctrnn, 19)

    stats.save()

    unique_genomes = stats.best_unique_genomes(5)
    assert 1 <= len(unique_genomes) <= 5, "Unique genomes: {!r}".format(unique_genomes)
    genomes = stats.best_genomes(5)
    assert 1 <= len(genomes) <= 5, "Genomes: {!r}".format(genomes)
    stats.best_genome()

    p.remove_reporter(stats) 

def test_serial5():
    """Test more configuration variations for simple serial run."""
    # Load configuration.
    local_dir = os.path.dirname(__file__)
    config_path = os.path.join(local_dir, 'test_configuration5')
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_path)

    if VERBOSE:
        print("config.genome_config.__dict__: {!r}".format(
            config.genome_config.__dict__))

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(VERBOSE))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(15, 1))

    # Run for up to 45 generations.
    p.run(eval_dummy_genomes_nn, 45)

    stats.save()
    # stats.save_genome_fitness(with_cross_validation=True)

    stats.get_fitness_stdev()
    stats.best_unique_genomes(5)
    stats.best_genomes(5)
    stats.best_genome()

    p.remove_reporter(stats) 

def test_serial3():
    """Test more configuration variations for simple serial run."""
    # Load configuration.
    local_dir = os.path.dirname(__file__)
    config_path = os.path.join(local_dir, 'test_configuration3')
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_path)

    if VERBOSE:
        print("config.genome_config.__dict__: {!r}".format(
            config.genome_config.__dict__))

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(VERBOSE))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(15, 1))

    # Run for up to 45 generations.
    p.run(eval_dummy_genomes_nn, 45)

    stats.save()
    # stats.save_genome_fitness(with_cross_validation=True)

    stats.get_fitness_stdev()
    stats.best_unique_genomes(5)
    stats.best_genomes(5)
    stats.best_genome()

    p.remove_reporter(stats) 

def test_serial_random():
    """Test basic (dummy fitness function) non-parallel run w/random activation, aggregation init."""
    # Load configuration.
    local_dir = os.path.dirname(__file__)
    config_path = os.path.join(local_dir, 'test_configuration2')
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_path)

    if VERBOSE:
        print("config.genome_config.__dict__: {!r}".format(
            config.genome_config.__dict__))

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(VERBOSE))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(15, 1))

    # Run for up to 45 generations.
    p.run(eval_dummy_genomes_nn, 45)

    stats.save()
    # stats.save_genome_fitness(with_cross_validation=True)

    stats.get_fitness_stdev()
    stats.best_unique_genomes(5)
    stats.best_genomes(5)
    stats.best_genome()

    p.remove_reporter(stats) 

def test_serial():
    """Test basic (dummy fitness function) non-parallel run."""
    # Load configuration.
    local_dir = os.path.dirname(__file__)
    config_path = os.path.join(local_dir, 'test_configuration')
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_path)

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(True))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(1, 5))

    # Run for up to 19 generations.
    p.run(eval_dummy_genomes_nn, 19)

    stats.save()
    # stats.save_genome_fitness(with_cross_validation=True)

    assert len(stats.get_fitness_stdev())
    stats.best_unique_genomes(5)
    stats.best_genomes(5)
    stats.best_genome()

    p.remove_reporter(stats) 

def run(config_file):
    # Load configuration.
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_file)

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(True))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(5))

    # Run for up to 300 generations.
    winner = p.run(eval_genomes, 300)

    # Display the winning genome.
    print('\nBest genome:\n{!s}'.format(winner))

    # Show output of the most fit genome against training data.
    print('\nOutput:')
    winner_net = neat.nn.FeedForwardNetwork.create(winner, config)
    for xi, xo in zip(xor_inputs, xor_outputs):
        output = winner_net.activate(xi)
        print("input {!r}, expected output {!r}, got {!r}".format(xi, xo, output))

    node_names = {-1:'A', -2: 'B', 0:'A XOR B'}
    visualize.draw_net(config, winner, True, node_names=node_names)
    visualize.plot_stats(stats, ylog=False, view=True)
    visualize.plot_species(stats, view=True)

    p = neat.Checkpointer.restore_checkpoint('neat-checkpoint-4')
    p.run(eval_genomes, 10) 

def run_experiment(config_file):
    """
    The function to run XOR experiment against hyper-parameters 
    defined in the provided configuration file.
    The winner genome will be rendered as a graph as well as the
    important statistics of neuroevolution process execution.
    Arguments:
        config_file: the path to the file with experiment 
                    configuration
    """
    # Load configuration.
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_file)

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(True))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(5, filename_prefix='out/neat-checkpoint-'))

    # Run for up to 300 generations.
    best_genome = p.run(eval_genomes, 300)

    # Display the best genome among generations.
    print('\nBest genome:\n{!s}'.format(best_genome))

    # Show output of the most fit genome against training data.
    print('\nOutput:')
    net = neat.nn.FeedForwardNetwork.create(best_genome, config)
    for xi, xo in zip(xor_inputs, xor_outputs):
        output = net.activate(xi)
        print("input {!r}, expected output {!r}, got {!r}".format(xi, xo, output))

    # Check if the best genome is an adequate XOR solver
    best_genome_fitness = eval_fitness(net)
    if best_genome_fitness > config.fitness_threshold:
        print("\n\nSUCCESS: The XOR problem solver found!!!")
    else:
        print("\n\nFAILURE: Failed to find XOR problem solver!!!")

    # Visualize the experiment results
    node_names = {-1:'A', -2: 'B', 0:'A XOR B'}
    visualize.draw_net(config, best_genome, True, node_names=node_names, directory=out_dir)
    visualize.plot_stats(stats, ylog=False, view=True, filename=os.path.join(out_dir, 'avg_fitness.svg'))
    visualize.plot_species(stats, view=True, filename=os.path.join(out_dir, 'speciation.svg')) 

def run_experiment(config_file, n_generations=100):
    """
    The function to run the experiment against hyper-parameters 
    defined in the provided configuration file.
    The winner genome will be rendered as a graph as well as the
    important statistics of neuroevolution process execution.
    Arguments:
        config_file: the path to the file with experiment 
                    configuration
    """
    # Load configuration.
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_file)

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(True))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(5, filename_prefix='out/spb-neat-checkpoint-'))

    # Run for up to N generations.
    best_genome = p.run(eval_genomes, n=n_generations)

    # Display the best genome among generations.
    print('\nBest genome:\n{!s}'.format(best_genome))

    # Check if the best genome is a winning Single-Pole balancing controller 
    net = neat.nn.FeedForwardNetwork.create(best_genome, config)
    print("\n\nEvaluating the best genome in random runs")
    success_runs = evaluate_best_net(net, config, additional_num_runs)
    print("Runs successful/expected: %d/%d" % (success_runs, additional_num_runs))
    if success_runs == additional_num_runs:
        print("SUCCESS: The stable Single-Pole balancing controller found!!!")
    else:
        print("FAILURE: Failed to find the stable Single-Pole balancing controller!!!")

    # Visualize the experiment results
    node_names = {-1:'x', -2:'dot_x', -3:'θ', -4:'dot_θ', 0:'action'}
    visualize.draw_net(config, best_genome, True, node_names=node_names, directory=out_dir, fmt='svg')
    visualize.plot_stats(stats, ylog=False, view=True, filename=os.path.join(out_dir, 'avg_fitness.svg'))
    visualize.plot_species(stats, view=True, filename=os.path.join(out_dir, 'speciation.svg')) 

def run_experiment(config_file, n_generations=100, silent=False):
    """
    The function to run the experiment against hyper-parameters 
    defined in the provided configuration file.
    The winner genome will be rendered as a graph as well as the
    important statistics of neuroevolution process execution.
    Arguments:
        config_file: the path to the file with experiment 
                    configuration
    Returns:
        True if experiment finished with successful solver found. 
    """
    # set random seed
    seed = 1559231616#int(time.time())#
    random.seed(seed)

    # Load configuration.
    config = neat.Config(neat.DefaultGenome, neat.DefaultReproduction,
                         neat.DefaultSpeciesSet, neat.DefaultStagnation,
                         config_file)

    # Create the population, which is the top-level object for a NEAT run.
    p = neat.Population(config)

    # Add a stdout reporter to show progress in the terminal.
    p.add_reporter(neat.StdOutReporter(True))
    stats = neat.StatisticsReporter()
    p.add_reporter(stats)
    p.add_reporter(neat.Checkpointer(5, filename_prefix='out/tpbm-neat-checkpoint-'))

    # Run for up to N generations.
    best_genome = p.run(eval_genomes, n=n_generations)

    # Display the best genome among generations.
    print('\nBest genome:\n{!s}'.format(best_genome))

    # Check if the best genome is a winning Double-Pole-Markov balancing controller 
    net = neat.nn.FeedForwardNetwork.create(best_genome, config)
    print("\n\nEvaluating the best genome in random runs")
    success_runs = evaluate_best_net(net, config, additional_num_runs)
    print("Runs successful/expected: %d/%d" % (success_runs, additional_num_runs))
    if success_runs == additional_num_runs:
        print("SUCCESS: The stable Double-Pole-Markov balancing controller found!!!")
    else:
        print("FAILURE: Failed to find the stable Double-Pole-Markov balancing controller!!!")

    print("Random seed:", seed)

    # Visualize the experiment results
    if not silent or success_runs == additional_num_runs:
        node_names = {-1:'x', -2:'dot_x', -3:'θ_1', -4:'dot_θ_1', -5:'θ_2', -6:'dot_θ_2', 0:'action'}
        visualize.draw_net(config, best_genome, True, node_names=node_names, directory=out_dir, fmt='svg')
        visualize.plot_stats(stats, ylog=False, view=True, filename=os.path.join(out_dir, 'avg_fitness.svg'))
        visualize.plot_species(stats, view=True, filename=os.path.join(out_dir, 'speciation.svg'))
    
    return success_runs == additional_num_runs