def test_raises_error_invalid_mutation_probability(crossover, mutation, prob,
expected_error):
with pytest.raises(expected_error):
_ = VarAnd(crossover=crossover,
mutation=mutation,
crossover_probability=0.0,
mutation_probability=prob)
def test_var_and_mutation(single_value_population_of_4, crossover, mutation_n):
variation = VarAnd(crossover=crossover,
mutation=mutation_n,
crossover_probability=0.0,
mutation_probability=1.0)
offspring = variation(single_value_population_of_4, 4)
for off, indv in zip(offspring, single_value_population_of_4):
assert off.value == -indv.value
def test_var_and_for_correct_offspring_size(single_value_population_of_4,
num_offspring, crossover,
mutation):
variation = VarAnd(crossover=crossover,
mutation=mutation,
crossover_probability=0.5,
mutation_probability=0.5)
offspring = variation(single_value_population_of_4, num_offspring)
assert len(offspring) == num_offspring
def test_var_and_stats_are_correct_size(single_value_population_of_4,
crossover, mutation):
variation = VarAnd(crossover=crossover,
mutation=mutation,
crossover_probability=0.5,
mutation_probability=0.5)
_ = variation(single_value_population_of_4, 4)
assert len(variation.crossover_offspring) == 4
assert len(variation.mutation_offspring) == 4
def test_var_and_correct_stats_output(single_value_population_of_4, crossover,
mutation, cross_prob, mut_prob,
expected_cross, expected_mut):
variation = VarAnd(crossover=crossover,
mutation=mutation,
crossover_probability=cross_prob,
mutation_probability=mut_prob)
_ = variation(single_value_population_of_4, 4)
for cross, mut in zip(variation.crossover_offspring,
variation.mutation_offspring):
assert cross == expected_cross
assert mut == expected_mut
def test_var_and_approximate_probabilities(single_value_population_of_100,
crossover, mutation):
np.random.seed(1)
variation = VarAnd(crossover=crossover,
mutation=mutation,
crossover_probability=0.5,
mutation_probability=0.5)
_ = variation(single_value_population_of_100, 100)
num_cross = np.sum(variation.crossover_offspring)
num_mut = np.sum(variation.mutation_offspring)
num_both = np.sum(
np.logical_and(variation.crossover_offspring,
variation.mutation_offspring))
num_rep = 100 - np.sum(
np.logical_or(variation.crossover_offspring,
variation.mutation_offspring))
assert num_cross == 56
assert num_mut == 44
assert num_both == 22
assert num_rep == 22