def test_encode(self):
param_1 = parametrization.DirectParam(np.array([1, 2, 1, 4]),
bounds = [0, 1])
param_2 = parametrization.DirectParam(np.array([2, 2, 1]),
bounds = [0.9, 2.1])
param_3 = QuadraticParam(np.array([1, 2, 1, 1, 5]), 1.5)
param = composite_parametrization.CompositeParam(
[param_1, param_2, param_3])
np.testing.assert_array_almost_equal(
param.encode(),
np.array([1, 2, 1, 4, 2, 2, 1, 1, 2, 1, 1, 5]), 7)
def test_calculate_gradient(self):
param_1 = parametrization.DirectParam(np.array([1, 2, 1, 4]),
bounds = [0, 1])
param_2 = parametrization.DirectParam(np.array([2, 2, 1]),
bounds = [0.9, 2.1])
param_3 = QuadraticParam(np.array([1, 2, 1, 1, 5]), 1.5)
param = composite_parametrization.CompositeParam(
[param_1, param_2, param_3])
vec = np.array([1, 2, 1, 2, 4, 1, 1, 3, 3, 4, 5, 1])
np.testing.assert_array_almost_equal(
param.calculate_gradient() @ vec,
np.array([1, 2, 1, 2, 4, 1, 1, 9, 18, 12, 15]), 7)
def test_deserialize(self):
param_1 = parametrization.DirectParam(np.array([1, 2, 1, 4]),
bounds=[0, 1])
param_2 = parametrization.DirectParam(np.array([2, 3.5, 1]),
bounds=[0.9, 2.1])
param_3 = QuadraticParam(np.array([1, 2]), 1.5)
param = composite_parametrization.CompositeParam(
[param_1, param_2, param_3])
param.deserialize({"param_0": {"vector": np.array([1, 2, 2, 3])},
"param_1": {"vector": np.array([1, 2, 5])},
"param_2": {"vector": np.array([1, 5])}})
np.testing.assert_array_almost_equal(
param_1.encode(), np.array([1, 1, 1, 1]), 7)
np.testing.assert_array_almost_equal(
param_2.encode(), np.array([1, 2, 2.1]), 7)
np.testing.assert_array_almost_equal(
param_3.encode(), np.array([1.5, 7.5]), 7)
def test_bounds(self):
param_1 = parametrization.DirectParam(np.array([1, 2, 1, 4]),
bounds=[0, 1])
param_2 = parametrization.DirectParam(np.array([2, 3.5, 1]),
bounds=[0.9, 2.1])
param_3 = QuadraticParam(np.array([1, 2, 1, 1, 5]), 1.5)
param = composite_parametrization.CompositeParam(
[param_1, param_2, param_3])
lower_bounds, upper_bounds = param.get_bounds()
np.testing.assert_array_equal(
lower_bounds,
np.array([0, 0, 0, 0, 0.9, 0.9, 0.9,
None, None, None, None, None]))
np.testing.assert_array_equal(
upper_bounds,
np.array([1, 1, 1, 1, 2.1, 2.1, 2.1,
None, None, None, None, None]))
def test_serialize(self):
param_1 = parametrization.DirectParam(np.array([1, 2, 1, 4]),
bounds=[0, 1])
param_2 = parametrization.DirectParam(np.array([2, 3.5, 1]),
bounds=[0.9, 2.1])
param_3 = QuadraticParam(np.array([1, 2]), 1.5)
param = composite_parametrization.CompositeParam(
[param_1, param_2, param_3])
data = param.serialize()
np.testing.assert_array_almost_equal(
data["param_0"]["vector"],
np.array([1, 2, 1, 4]), 7)
np.testing.assert_array_almost_equal(
data["param_1"]["vector"],
np.array([2, 3.5, 1]), 7)
np.testing.assert_array_almost_equal(
data["param_2"]["vector"],
np.array([1.5, 3]), 7)
def test_from_vector(self):
param_1 = parametrization.DirectParam(np.array([1, 2, 1, 4]),
bounds=[0, 1])
param_2 = parametrization.DirectParam(np.array([2, 3.5, 1]),
bounds=[0.9, 2.1])
param_3 = QuadraticParam(np.array([1, 2]), 1.5)
param = composite_parametrization.CompositeParam(
[param_1, param_2, param_3])
param.from_vector(np.array([1, 1, 2, 1, 4, 3, 1, 2, 2]))
np.testing.assert_array_almost_equal(
param_1.to_vector(),
np.array([1, 1, 1, 1]), 7)
np.testing.assert_array_almost_equal(
param_2.to_vector(),
np.array([2.1, 2.1, 1]), 7)
np.testing.assert_array_almost_equal(
param_3.to_vector(),
np.array([4.5, 4.5]), 7)
