def test_rbf_obj_and_obj_jac_match():
"""obj and obj_jac functions should return the same obj value."""
attrs = random.randint(1, 10)
outs = random.randint(1, 10)
model = rbf.RBF(attrs, random.randint(1, 10), outs)
dataset = datasets.get_random_regression(10, attrs, outs)
# Don't use exactly the same parameters, to ensure obj functions are actually
# using the given parameters
parameters = random.uniform(-1.0, 1.0) * model._weight_matrix.ravel()
assert helpers.approx_equal(
model._get_obj(parameters, dataset[0], dataset[1]),
model._get_obj_jac(parameters, dataset[0], dataset[1])[0])
def _check_jacobian(make_model_func):
attrs = random.randint(1, 10)
outs = random.randint(1, 10)
model = make_model_func(attrs, random.randint(1, 10), outs)
inp_matrix, tar_matrix = datasets.get_random_regression(10, attrs, outs)
# Test jacobian of error function
f = lambda xk: mlp._mlp_obj(model, inp_matrix, tar_matrix, xk)
df = lambda xk: mlp._mlp_obj_jac(model, inp_matrix, tar_matrix, xk)[1]
helpers.check_gradient(f,
df,
inputs=mlp._flatten(model._weight_matrices),
f_shape='scalar')
def _check_obj_and_obj_jac_match(make_model_func, classification=False):
"""obj and obj_jac functions should return the same obj value."""
attrs = random.randint(1, 10)
outs = random.randint(1, 10)
model = make_model_func(attrs, random.randint(1, 10), outs)
if classification:
dataset = datasets.get_random_classification(10, attrs, outs)
else:
dataset = datasets.get_random_regression(10, attrs, outs)
# Don't use exactly the same parameters, to ensure obj functions are actually
# using the given parameters
parameters = random.uniform(-1.0, 1.0) * mlp._flatten(
model._weight_matrices)
assert helpers.approx_equal(
mlp._mlp_obj(model, dataset[0], dataset[1], parameters),
mlp._mlp_obj_jac(model, dataset[0], dataset[1], parameters)[0])
def test_split_dataset():
input_matrix, target_matrix = datasets.get_random_regression(
random.randint(100, 150), random.randint(2, 5), random.randint(1, 3))
num_sets = random.randint(2, 5)
sets = validation._split_dataset(input_matrix, target_matrix, num_sets)
# The right number of sets is created
assert len(sets) == num_sets
for i in range(num_sets):
for j in range(i+1, num_sets):
# Check that each set is about equal in size
assert len(sets[i]) >= len(sets[j])-5 and len(sets[i]) <= len(sets[j])+5
# Check that each set has unique patterns
patterns = zip(*sets[i])
other_patterns = zip(*sets[j])
for pattern in patterns:
for other_pattern in other_patterns:
assert not ((pattern[0] == other_pattern[0]).all()
and (pattern[1] == other_pattern[1]).all())
def test_get_random_regression_dataset():
num_points = random.randint(1, 100)
input_size = random.randint(1, 100)
num_targets = random.randint(2, 10)
input_matrix, target_matrix = datasets.get_random_regression(
num_points, input_size, num_targets)
assert len(input_matrix) == num_points
assert len(target_matrix) == num_points
for inp_vec, tar_vec in zip(input_matrix, target_matrix):
assert len(inp_vec) == input_size
assert len(tar_vec) == num_targets
# Check values in range
for val in inp_vec:
assert -1 <= val <= 1
# Target is a random vector
for val in tar_vec:
assert -1 <= val <= 1