def test_surrogate_optimize_with_all_selection_and_infill_methods(
infill, surrogate_select_method):
# Define original problem
try:
problem = benchmarks.zdt3()
# Sample
randomSample = sampling.rand(problem, 15)
# Define surrogate ensemble
# Define Optimizer
optimizer = NSGA2(pop_size=20,
n_offsprings=10,
sampling=get_sampling("real_random"),
crossover=get_crossover("real_sbx", prob=0.9,
eta=15),
mutation=get_mutation("real_pm", eta=20),
eliminate_duplicates=True)
# Define termination criteria
termination = get_termination("n_gen", 10)
# Define infill criteria
infill_method = infill
# Define surrogate selection
surrogate_selection_function = surrogate_select_method
# Optimize
samples = copy.deepcopy(randomSample)
surrogate_ensemble = [
LinearRegression(),
KNeighborsRegressor(),
DecisionTreeRegressor(),
]
res = surrogate_optimization.optimize(problem,
optimizer,
termination,
surrogate_ensemble,
samples,
infill_method,
surrogate_selection_function,
n_infill=2,
max_samples=20)
except Exception as e:
raise pytest.fail(
"infill: {0} /n and surrogate_selection: {1}/n raise a error: {2}".
format(infill, surrogate_select_method, e))
def test_rand_samples_should_respect_upper_bound(test_problem):
upper_bound = test_problem.xu
samples = sampling.rand(test_problem, 1000)
X_samples = samples['X']
if (type(X_samples[0]) is np.ndarray):
sample_bigger_than_upper_bound = False
for sample in X_samples:
sample_bigger_than_upper_bound = any(sample > upper_bound)
if (sample_bigger_than_upper_bound):
break
assert (not sample_bigger_than_upper_bound)
else:
max_sample = max(X_samples)
assert (max_sample < upper_bound)
def test_rand_samples_should_respect_lower_bound(test_problem):
lower_bound = test_problem.xl
samples = sampling.rand(test_problem, 1000)
X_samples = samples['X']
if (type(X_samples[0]) is np.ndarray):
sample_smaller_than_lower_bound = False
for sample in X_samples:
sample_smaller_than_lower_bound = any(sample < lower_bound)
if (sample_smaller_than_lower_bound):
break
assert (not sample_smaller_than_lower_bound)
else:
min_sample = min(X_samples)
assert (min_sample > lower_bound)
def test_rand_should_return_50_samples(test_problem):
assert (len(sampling.rand(test_problem, 50)['X']) == 50)
def test_rand_sampling_should_return_dict_with_X_F_G(test_problem,
variable_name):
assert (variable_name in sampling.rand(test_problem))
def test_rand_sampling_should_return_a_dict(test_problem):
assert (type(sampling.rand(test_problem)) is dict)
from pymoo.factory import get_sampling, get_crossover, get_mutation
from pymoo.factory import get_termination
from pymoo.util.plotting import plot
import time
import copy
if __name__ == "__main__":
# Define original problem
# problem = benchmarks.mw1()
# Define original problem
problem = benchmarks.zdt3()
# Sample
randomSample = sampling.rand(problem, 70)
# Define surrogate ensemble
# Define Optimizer
optimizer = NSGA2(pop_size=250,
n_offsprings=100,
sampling=get_sampling("real_random"),
crossover=get_crossover("real_sbx", prob=0.9, eta=15),
mutation=get_mutation("real_pm", eta=20),
eliminate_duplicates=True)
# Define termination criteria
termination = get_termination("n_gen", 100)