def polynomial():
"""Dummy polynomial model and data
"""
input_data = np.array([0, 0.2, 0.4, 0.6, 0.8, 1.0])
y = 7 * input_data + 2
formula = gammy.Scalar((0, 1e-6)) * x + gammy.Scalar((0, 1e-6))
return (input_data, y, formula)
def dummy_data():
np.random.seed(42)
input_data = 10 * np.random.rand(30)
y = (5 * input_data + 2.0 * input_data**2 + 7 +
10 * np.random.randn(len(input_data)))
formula = (gammy.Scalar(prior=(0, 1e-6)) * x +
gammy.Scalar(prior=(0, 1e-6)) * x**2 +
gammy.Scalar(prior=(0, 1e-6)))
return (input_data, y, formula)
def multinomial():
return (
gammy.Scalar((0, 1)) * x[:, 0] * x[:, 1] + gammy.Scalar((0, 1)),
np.vstack(
(
np.arange(0, 1.5, 0.5),
-np.arange(0, 1.5, 0.5)
)
).T
)
def polynomial():
"""Dummy polynomial model and data
"""
np.random.seed(42)
input_data = 10 * np.random.rand(30)
y = (5 * input_data + 2.0 * input_data**2 + 7 +
10 * np.random.randn(len(input_data)))
formula = (gammy.Scalar(prior=(0, 1e-6)) * x +
gammy.Scalar(prior=(0, 1e-6)) * x**2 +
gammy.Scalar(prior=(0, 1e-6)))
return (input_data, y, formula)
def test_polynomial(fit_model):
np.random.seed(42)
input_data = 10 * np.random.rand(30)
y = (5 * input_data + 2.0 * input_data**2 + 7 +
10 * np.random.randn(len(input_data)))
formula = (gammy.Scalar(prior=(0, 1e-6)) * x +
gammy.Scalar(prior=(0, 1e-6)) * x**2 +
gammy.Scalar(prior=(0, 1e-6)))
model = fit_model(formula, gammy.numpy.Delta(0.01342))(input_data, y)
assert_almost_equal(model.predict(input_data[[0, 3, 6, 9, 12]]),
np.array(
[52.5711, 104.2863, 14.4829, 136.9858, 180.0926]),
decimal=3)
return
def test_polynomial():
np.random.seed(42)
input_data = 10 * np.random.rand(30)
y = (5 * input_data + 2.0 * input_data**2 + 7 +
10 * np.random.randn(len(input_data)))
formula = (gammy.Scalar(prior=(0, 1e-6)) * x +
gammy.Scalar(prior=(0, 1e-6)) * x**2 +
gammy.Scalar(prior=(0, 1e-6)))
model = gammy.BayesianGAM(formula).fit(input_data, y)
assert_allclose(
model.predict(input_data[[0, 3, 6, 9, 12]]),
np.array([
52.57112684, 104.28633909, 14.48274839, 136.98584437, 180.09263955
]))
return
def test_kron():
np.random.seed(42)
n = 30
input_data = np.vstack((
6 * np.random.rand(n) - 3,
6 * np.random.rand(n) - 3,
)).T
def peaks(x, y):
"""The function in Mathworks logo
"""
return (3 * (1 - x)**2 * np.exp(-(x**2) - (y + 1)**2) - 10 *
(x / 5 - x**3 - y**5) * np.exp(-x**2 - y**2) -
1 / 3 * np.exp(-(x + 1)**2 - y**2))
y = peaks(input_data[:, 0], input_data[:,
1]) + 4 + 0.3 * np.random.randn(n)
formula = gammy.Kron(
gammy.ExpSquared1d(
np.arange(-3, 3, 0.1), corrlen=0.5, sigma=4.0, energy=0.99)(x[:,
0]),
gammy.ExpSquared1d(
np.arange(-3, 3, 0.1), corrlen=0.5, sigma=4.0, energy=0.99)(
x[:, 1])) + gammy.Scalar(prior=(0, 1e-6))
model = gammy.BayesianGAM(formula).fit(input_data, y)
assert_allclose(model.predict(input_data[[1, 5, 12, 19], :]),
np.array([3.78593199, 3.56403323, 3.73910833, 3.55960657]))
return
def test_kron(fit_model):
np.random.seed(42)
n = 30
input_data = np.vstack((
6 * np.random.rand(n) - 3,
6 * np.random.rand(n) - 3,
)).T
def peaks(x, y):
"""The function in Mathworks logo
"""
return (3 * (1 - x)**2 * np.exp(-(x**2) - (y + 1)**2) - 10 *
(x / 5 - x**3 - y**5) * np.exp(-x**2 - y**2) -
1 / 3 * np.exp(-(x + 1)**2 - y**2))
y = peaks(input_data[:, 0], input_data[:,
1]) + 4 + 0.3 * np.random.randn(n)
formula = gammy.Kron(
gammy.ExpSquared1d(
np.arange(-3, 3, 0.1), corrlen=0.5, sigma=4.0, energy=0.99)(x[:,
0]),
gammy.ExpSquared1d(
np.arange(-3, 3, 0.1), corrlen=0.5, sigma=4.0, energy=0.99)(
x[:, 1])) + gammy.Scalar(prior=(0, 1e-6))
model = fit_model(formula, gammy.numpy.Delta(24.2289))(input_data, y)
assert_almost_equal(model.predict(input_data[[1, 5, 12, 19], :]),
np.array([3.7859, 3.5640, 3.7391, 3.5596]),
decimal=3)
return
def gp():
"""Dummy GP model and data
"""
np.random.seed(42)
n = 50
input_data = np.vstack(
(
2 * np.pi * np.random.rand(n),
np.random.rand(n),
)
).T
y = (
np.abs(np.cos(input_data[:, 0])) * input_data[:, 1] +
1 +
0.1 * np.random.randn(n)
)
formula = gammy.ExpSineSquared1d(
np.arange(0, 2 * np.pi, 0.1),
corrlen=1.0,
sigma=1.0,
period=2 * np.pi,
energy=0.99
)(x[:, 0]) * x[:, 1] + gammy.Scalar(prior=(0, 1e-6))
return (input_data, y, formula)
def test_gp():
np.random.seed(42)
n = 50
input_data = np.vstack((
2 * np.pi * np.random.rand(n),
np.random.rand(n),
)).T
y = (np.abs(np.cos(input_data[:, 0])) * input_data[:, 1] + 1 +
0.1 * np.random.randn(n))
formula = gammy.ExpSineSquared1d(
np.arange(0, 2 * np.pi, 0.1),
corrlen=1.0,
sigma=1.0,
period=2 * np.pi,
energy=0.99)(x[:, 0]) * x[:, 1] + gammy.Scalar(prior=(0, 1e-6))
model = gammy.BayesianGAM(formula).fit(input_data, y)
assert_allclose(
model.predict(input_data[[1, 42, 11, 26, 31]]),
np.array([1.78918855, 1.87107355, 1.30149328, 1.29221874, 1.31596102]))
return
def test_gp(fit_model):
np.random.seed(42)
n = 50
input_data = np.vstack((
2 * np.pi * np.random.rand(n),
np.random.rand(n),
)).T
y = (np.abs(np.cos(input_data[:, 0])) * input_data[:, 1] + 1 +
0.1 * np.random.randn(n))
formula = gammy.ExpSineSquared1d(
np.arange(0, 2 * np.pi, 0.1),
corrlen=1.0,
sigma=1.0,
period=2 * np.pi,
energy=0.99)(x[:, 0]) * x[:, 1] + gammy.Scalar(prior=(0, 1e-6))
model = fit_model(formula, gammy.numpy.Delta(100))(input_data, y)
assert_almost_equal(model.predict(input_data[[1, 42, 11, 26, 31]]),
np.array([1.7891, 1.8710, 1.3014, 1.2922, 1.3159]),
decimal=3)
return
def square():
return (
gammy.Scalar() * x ** 2,
np.array([0., 1., 2.])
)
def line():
return (
gammy.Scalar((1, 2)) * x,
np.array([0., 1., 2.])
)
def scalar():
return (
gammy.Scalar((0, 1)),
np.array([0., 1., 2.])
)
