def test_weight_and_integration_1d(self):
"""
Test that the Gaussian distribution intagrates to the weight.
"""
weight = 1.0
gaussian = Gaussian(cov=2.0,
mean=4.0,
log_weight=np.log(weight),
var_names=["a"])
definite_integral, _ = integrate.quad(gaussian.potential, -100.0,
100.0)
self.assertAlmostEqual(definite_integral, weight)
# test another weight and with canonical form
weight = 2.0
gaussian = Gaussian(cov=2.0,
mean=4.0,
log_weight=np.log(weight),
var_names=["a"])
gaussian._update_canform()
gaussian.covform = False
definite_integral, _ = integrate.quad(gaussian.potential, -100.0,
100.0)
self.assertAlmostEqual(definite_integral, weight)
def test_impossible_update_canform(self):
"""
Test that the _update_canform raises a LinAlgError when the covariance matrix is not invertible.
"""
gaussian = Gaussian(cov=np.zeros([2, 2]),
mean=[0.0, 0.0],
log_weight=0.0,
var_names=["a", "b"])
with self.assertRaises(np.linalg.LinAlgError):
gaussian._update_canform()
def test_marginalise_2d_canform(self):
"""
Test that the Gaussian marginalisation function returns the correct result for a two dimensional Gaussians.
"""
gaussian = Gaussian(cov=[[7.0, 2.0], [2.0, 1.0]],
mean=[4.0, 1.0],
log_weight=0.0,
var_names=["a", "b"])
expected_result = Gaussian(cov=7.0,
mean=4.0,
log_weight=0.0,
var_names=["a"])
expected_result._update_canform()
gaussian._update_canform()
actual_result = gaussian.marginalize(vrs=["a"], keep=True)
self.assertTrue(expected_result.equals(actual_result))