def genetic_algorithm_binary(fitness_fn,
chr_size,
pop_size=100,
total_generations=20,
cx_rate=0.7,
ind_mt_rate=1.0,
op_mt_rate=0.01,
elitism=False):
# initial population
pop = pg.make_integer_population(pop_size, chr_size)
pop = pg.evaluate_population(pop, fitness_fn)
pg.evolution_progress(0, pop)
# evolutionary loop
for i in range(1, total_generations):
pop = pg.select_population(pop, pg.tournament_selection)
offsprings = pg.apply_crossover(pop, cx_rate, pg.uniform_crossover)
offsprings = pg.apply_mutation(offsprings,
ind_mt_rate,
pg.flip_mutation,
gene_rate=op_mt_rate)
if elitism:
pop = pg.elite_strategy(
pg.generational_replacement(pop, offsprings),
pg.best_individual(pop))
else:
pop = pg.generational_replacement(pop, offsprings)
pop = pg.evaluate_population(pop, fitness_fn)
pg.evolution_progress(i, pop)
return pop
def test_worst_individual():
np.random.seed(42)
pop = pg.make_integer_population(10, 5)
pop = pg.evaluate_population(pop, lambda x: 1. / (1. + pg.onemax(x)))
worst = pg.worst_individual(pop)
assert np.array_equal(worst.genotype, np.array([0, 1, 0, 0, 0]))
assert worst.fitness.value == 0.5
def test_best_individual():
np.random.seed(42)
pop = pg.make_integer_population(10, 5)
pop = pg.evaluate_population(pop, lambda x: 1. / (1. + pg.onemax(x)))
best = pg.best_individual(pop)
assert np.array_equal(best.genotype, np.array([1, 1, 1, 1, 1]))
assert best.fitness.value == 0.16666666666666666
def test_m_plus_lambda():
np.random.seed(42)
sphere_model_size = 30
fitness_fn = lambda x: pg.sphere_model(x[:sphere_model_size])
size = 10
pop1 = pg.make_uniform_population(size, sphere_model_size)
pop1 = pg.evaluate_population(pop1, fitness_fn)
pop1_clone = pop1.clone()
pop2 = pg.make_uniform_population(size, sphere_model_size)
pop2 = pg.evaluate_population(pop2, fitness_fn)
pop1 = pg.mu_plus_lambda_replacement(pop1, pop2)
f_p1c = np.array(
[pop1_clone.individuals[i].fitness.value for i in range(size)])
f_p2 = np.array([pop2.individuals[i].fitness.value for i in range(size)])
f_p1 = np.array([pop1.individuals[i].fitness.value for i in range(size)])
res = np.array([
7.76332642, 8.08389311, 8.42498289, 8.77341594, 8.9523664, 9.45857558,
9.49688766, 9.52769953, 9.7264757, 9.83516821
])
assert np.allclose(f_p1, res) is True
assert np.allclose(f_p1, f_p1c) is False
assert f_p1.any() == f_p2.any()
def test_m_comma_lambda():
np.random.seed(42)
sphere_model_size = 30
fitness_fn = lambda x: pg.sphere_model(x[:sphere_model_size])
size = 10
pop1 = pg.make_uniform_population(size, sphere_model_size)
pop1 = pg.evaluate_population(pop1, fitness_fn)
pop1_clone = pop1.clone()
pop2 = pg.make_uniform_population(size, sphere_model_size)
pop2 = pg.evaluate_population(pop2, fitness_fn)
pop1 = pg.mu_comma_lambda_replacement(pop1, pop2)
f_p1c = np.array(
[pop1_clone.individuals[i].fitness.value for i in range(size)])
f_p2 = np.array([pop2.individuals[i].fitness.value for i in range(size)])
f_p1 = np.array([pop1.individuals[i].fitness.value for i in range(size)])
res = np.array([
7.7633264242708808, 8.4249828886076532, 9.4585755842303119,
9.8351682100669695, 10.109666113362849, 10.120464613773503,
11.148338435537882, 12.62638917659477, 12.68750163403401,
13.486951247705489
])
assert np.allclose(f_p1, res) is True
assert np.allclose(f_p1, f_p1c) is False
assert f_p1.any() == f_p2.any()
def test_elite_strategy():
np.random.seed(42)
size = 10
ind_size = 20
pop = pg.make_integer_population(size, ind_size)
fitness_function = lambda x: 1. / pg.onemax(x)
pop = pg.make_integer_population(size, ind_size)
pop = pg.evaluate_population(pop, fitness_function)
best = pg.best_individual(pop)
pop = pg.elite_strategy(pop, best)
found = False
for ind in pop.individuals:
if ind.equal(best) is True:
found = True
break
assert found is True
def test_select_population():
np.random.seed(42)
pop = pg.make_integer_population(10, 5)
pop = pg.evaluate_population(pop, lambda x: 1. / (1. + pg.onemax(x)))
diff = 0
for i in range(pop.size):
if not np.array_equal(pop.individuals[i].genotype,
pop.individuals[i].genotype):
diff += 1
assert diff == 0
original_pop = pop.clone()
pop = pg.select_population(pop, pg.tournament_selection)
for i in range(pop.size):
if not np.array_equal(pop.individuals[i].genotype,
original_pop.individuals[i].genotype):
diff += 1
assert diff != 0