def setup_method(self):
pop_size = 500
length = 10
init_population = get_onemax_samples(pop_size, length)
optimizer = Optimizer(sample_data=init_population)
model1 = get_onemax_model()
optimizer.add_objective("ones", model1.predict, direction="maximize")
model2 = get_linear_model(length)
optimizer.add_objective("linear_min",
model2.predict,
direction="minimize")
optimizer.add_discrete_constraint(fname="1", constraints=[0, 1, 2])
optimizer.add_onehot_groupconstraint(group=["3", "4", "5"],
lower=1,
upper=1)
optimizer.add_sumtotal_groupconstraint(group=["7", "8", "9"],
lower=2,
upper=2)
optimizer.compile()
self.optimizer = optimizer
self.strategy = self.optimizer.strategy
self.init_population = self.optimizer.spawn_population(100).values
def setup_method(self):
pop_size = 500
length = 10
init_population = get_onemax_samples(pop_size, length)
optimizer = Optimizer(sample_data=init_population)
model1 = get_onemax_model()
optimizer.add_objective("ones", model1.predict, direction="maximize")
model2 = get_linear_model(length)
optimizer.add_objective("linear_min",
model2.predict,
direction="minimize")
optimizer.add_discrete_constraint(fname="2", constraints=[0, 1, 2])
optimizer.add_discrete_constraint(fname="3", constraints=[0, 1, 2])
optimizer.add_onehot_groupconstraint(group=["4", "5"])
optimizer.add_sumtotal_groupconstraint(group=["6", "7", "8"],
lower=5,
upper=5)
self.optimizer = optimizer
def test_evaluate(self):
X = get_onemax_samples(500, 10)
fitness = self.optimizer.evaluate_population(X)
assert np.all(fitness.columns == ["ones", "linear_min"])
def setup_method(self):
length = 20
init_popualtion = get_onemax_samples(500, length)
optimizer = Optimizer(sample_data=init_popualtion)
model1 = get_onemax_model()
optimizer.add_objective("ones", model1.predict, direction="maximize")
model2 = get_linear_model(length)
optimizer.add_objective("linear_min",
model2.predict,
direction="minimize")
model3 = get_linear_model(length)
optimizer.add_objective("linear_max",
model3.predict,
direction="maximize")
optimizer.add_discrete_constraint(fname="2", constraints=[0, 1, 2])
optimizer.add_discrete_constraint(fname="3", constraints=[0, 1, 2])
optimizer.add_valuerange_constraint(fname="2", lower=0, upper=2)
optimizer.add_valuerange_constraint(fname="3", lower=0, upper=2)
optimizer.add_onehot_groupconstraint(group=["4", "5"])
optimizer.add_onehot_groupconstraint(group=["6", "7"])
optimizer.add_sumtotal_groupconstraint(group=["8", "9", "10"],
lower=1,
upper=1)
optimizer.add_sumtotal_groupconstraint(group=["11", "12"],
lower=0,
upper=1)
def user_func(X):
print("test func is called!!")
print(X.columns)
return X
optimizer.add_user_constraint(user_func)
self.optimizer = optimizer
self.tempdir = os.path.join(os.path.expanduser("~"), "temp")
if not os.path.exists(self.tempdir):
os.makedirs(self.tempdir)
def setup_method(self):
init_population = get_onemax_samples(500, 10)
self.optimizer = Optimizer(sample_data=init_population)
model1 = get_onemax_model()
self.optimizer.add_objective("ones",
model1.predict,
direction="maximize")
model2 = get_linear_model(10)
self.optimizer.add_objective("linear_min",
model2.predict,
direction="minimize")
def setup_method(self):
pop_size = 500
length = 20
init_population = get_onemax_samples(pop_size, length)
optimizer = Optimizer(sample_data=init_population)
model1 = get_onemax_model()
optimizer.add_objective("ones", model1.predict, direction="maximize")
model2 = get_linear_model(length)
optimizer.add_objective("linear_min",
model2.predict,
direction="minimize")
optimizer.add_discrete_constraint(fname="2", constraints=[0, 1, 2])
optimizer.add_discrete_constraint(fname="3", constraints=[0, 1, 2])
optimizer.add_onehot_groupconstraint(group=["4", "5", "6"], n=1)
optimizer.add_sumtotal_groupconstraint(group=["7", "8", "9"],
lower=2,
upper=2)
optimizer.add_onehot_groupconstraint(group=["10", "11", "12", "13"],
n=3)
optimizer.add_onehot_groupconstraint(group=["14", "15"],
lower=3,
upper=3)
optimizer.compile()
self.optimizer = optimizer
population = self.optimizer.spawner.spawn(30)
population = self.optimizer.strategy.constraint(population.values)
self.population = pd.DataFrame(
population, columns=self.optimizer.config.feature_names)
def setup_method(self):
pop_size = 500
length = 10
init_population = get_onemax_samples(pop_size, length)
optimizer = Optimizer(sample_data=init_population)
model1 = get_onemax_model()
optimizer.add_objective("ones", model1.predict, direction="maximize")
model2 = get_linear_model(length)
optimizer.add_objective("linear_min",
model2.predict,
direction="minimize")
optimizer.add_discrete_constraint(fname="2", constraints=[0, 1, 2])
optimizer.add_discrete_constraint(fname="3", constraints=[0, 1, 2])
optimizer.add_onehot_groupconstraint(group=["4", "5", "6"])
optimizer.add_sumtotal_groupconstraint(group=["7", "8", "9"],
lower=2,
upper=2)
optimizer.compile()
self.optimizer = optimizer
population = self.optimizer.spawner.spawn(30)
population = self.optimizer.strategy.constraint(population.values)
self.population = pd.DataFrame(
population, columns=self.optimizer.config.feature_names)
self.export_dir = os.path.join(os.path.expanduser("~"), "optga")
if not os.path.exists(self.export_dir):
os.makedirs(self.export_dir)
def setup_method(self):
init_population = get_onemax_samples(500, 10)
self.optimizer = Optimizer(sample_data=init_population)
