def test_cmaes_get_best_params_best():
opt = CMAESOptimizer()
opt.init(10)
params = np.empty(10)
opt.get_next_parameters(params)
opt.set_evaluation_feedback(np.array([0.0]))
best_params = opt.get_best_parameters(method="best")
assert_array_almost_equal(params, best_params)
def test_cmaes_respects_bounds():
opt = CMAESOptimizer(bounds=[[-5, 4], [10, 20]], variance=10000.0,
random_state=0)
opt.init(2)
params = np.empty(2)
opt.get_next_parameters(params)
assert_true(np.all(params >= np.array([-5, 10])))
assert_true(np.all(params <= np.array([4, 20])))
def test_cmaes_stop_fitness_variance():
opt = CMAESOptimizer(n_samples_per_update=5)
opt.init(2)
params = np.empty(2)
it = 0
while not opt.is_behavior_learning_done():
opt.get_next_parameters(params)
opt.set_evaluation_feedback([0.0])
it += 1
assert_equal(it, 6)
def test_cmaes_stop_conditioning():
def objective(x):
return -1e10 * x[1] ** 2
opt = CMAESOptimizer(random_state=0)
opt.init(2)
params = np.empty(2)
it = 0
while not opt.is_behavior_learning_done():
opt.get_next_parameters(params)
opt.set_evaluation_feedback(objective(params))
it += 1
assert_less(it, 600)
def test_acmaes():
x = np.zeros(n_dims)
opt = CMAESOptimizer(x, active=True, random_state=0, log_to_stdout=False)
opt.init(n_dims)
r = eval_loop(x, opt, n_dims)
assert_greater(r.max(), -1e-5)
n_generations = 20
n_samples_per_update = 20
n_params = 2
for objective_name in ["Sphere", "SchaffersF7"]:
objective = FUNCTIONS[objective_name](0, n_params)
initial_params = 4.0 * np.ones(n_params)
cmaes = CMAESOptimizer(
initial_params=initial_params, variance=0.1, active=True,
n_samples_per_update=n_samples_per_update,
bounds=np.array([[-5, 5], [-5, 5]]), random_state=0)
cmaes.init(n_params)
n_rows = 4
plt.figure(figsize=(n_generations * 3 / n_rows, 3 * n_rows))
path = []
for it in range(n_generations):
plt.subplot(n_rows, int(n_generations / n_rows, it + 1))
plot_objective()
last_mean = cmaes.mean.copy()
path.append(last_mean)
last_cov = cmaes.var * cmaes.cov
X = np.empty((n_samples_per_update, n_params))
F = np.empty((n_samples_per_update, 1))
for i in range(n_samples_per_update):
cmaes.get_next_parameters(X[i])
def test_cmaes_no_initial_params():
opt = CMAESOptimizer()
opt.init(10)
params = np.empty(10)
opt.get_next_parameters(params)
def test_cmaes_diagonal_cov():
opt = CMAESOptimizer(covariance=np.zeros(10))
opt.init(10)
params = np.empty(10)
opt.get_next_parameters(params)
