def test_es_run():
def square(x):
return np.sum(x**2)
def hook(optimizer, space, function):
return
new_function = function.Function(pointer=square)
hyperparams = {
'child_ratio': 0.5
}
new_es = es.ES(hyperparams=hyperparams)
search_space = search.SearchSpace(n_agents=10, n_iterations=100,
n_variables=2, lower_bound=[0, 0],
upper_bound=[10, 10])
history = new_es.run(search_space, new_function, pre_evaluation=hook)
assert len(history.agents) > 0
assert len(history.best_agent) > 0
best_fitness = history.best_agent[-1][1]
assert best_fitness <= constants.TEST_EPSILON, 'The algorithm de failed to converge.'
def test_es_compile():
search_space = search.SearchSpace(n_agents=10,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_es = es.ES()
new_es.compile(search_space)
try:
new_es.n_children = 'a'
except:
new_es.n_children = 0
assert new_es.n_children == 0
try:
new_es.n_children = -1
except:
new_es.n_children = 0
assert new_es.n_children == 0
try:
new_es.strategy = 1
except:
new_es.strategy = np.array([1])
assert new_es.strategy == np.array([1])
def test_es_hyperparams():
hyperparams = {
'child_ratio': 0.5
}
new_es = es.ES(hyperparams=hyperparams)
assert new_es.child_ratio == 0.5
def test_es_update_strategy():
new_es = es.ES()
strategy = np.ones((4, 1))
new_strategy = new_es._update_strategy(strategy)
assert new_strategy[0][0] > 0
def test_es_update_strategy():
search_space = search.SearchSpace(n_agents=10,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_es = es.ES()
new_es.compile(search_space)
new_es._update_strategy(0)
assert new_es.strategy[0][0] > 0
def test_es_update():
def square(x):
return np.sum(x**2)
search_space = search.SearchSpace(n_agents=10,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_es = es.ES()
new_es.compile(search_space)
new_es.update(search_space, square)
def test_es_hyperparams_setter():
new_es = es.ES()
try:
new_es.child_ratio = 'a'
except:
new_es.child_ratio = 0.5
try:
new_es.child_ratio = -1
except:
new_es.child_ratio = 0.5
assert new_es.child_ratio == 0.5
def test_es_mutate_parent():
def square(x):
return np.sum(x**2)
new_es = es.ES()
search_space = search.SearchSpace(n_agents=4, n_iterations=100,
n_variables=2, lower_bound=[0, 0],
upper_bound=[10, 10])
strategy = np.zeros(4)
agent = new_es._mutate_parent(search_space.agents[0], square, strategy[0])
assert agent.position[0][0] > 0
def test_es_mutate_parent():
def square(x):
return np.sum(x**2)
search_space = search.SearchSpace(n_agents=10,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_es = es.ES()
new_es.compile(search_space)
agent = new_es._mutate_parent(search_space.agents[0], 0, square)
assert agent.position[0][0] > 0
def test_es_build():
new_es = es.ES()
assert new_es.built == True
def test_es_params():
params = {'child_ratio': 0.5}
new_es = es.ES(params=params)
assert new_es.child_ratio == 0.5
