def test_mcmc_with_other_prior(self):
test_rel = relationship.Relationship(utils.straight_line, TEST_X,
TEST_Y)
test_rel.max_likelihood('mini')
def other_prior():
"""
Another potential prior.
"""
priors = []
for i, variable in enumerate(test_rel.variables):
loc = variable.n
scale = 1
priors.append(norm(loc=loc, scale=scale))
return priors
actual_results = sampling.mcmc(test_rel,
prior_function=other_prior,
n_burn=10,
n_samples=10)
assert_equal(
isinstance(actual_results["distributions"][0], Distribution), True)
assert_equal(
isinstance(actual_results["distributions"][1], Distribution), True)
assert_equal(actual_results["distributions"][0].size, 500)
assert_equal(actual_results["distributions"][1].size, 500)
def test_ln_likelihood(self):
test_rel = relationship.Relationship(utils.straight_line, TEST_X,
TEST_Y)
expected_lnl = -3
actual_lnl = optimize.ln_likelihood([1., 0.], test_rel.function,
test_rel.x, test_rel.y)
assert_almost_equal(actual_lnl, expected_lnl, decimal=0)
def test_nested_sampling(self):
test_rel = relationship.Relationship(utils.straight_line,
TEST_X,
TEST_Y,
bounds=((0, 10), (-1, 1)))
actual_results = sampling.nested_sampling(test_rel, maxiter=100)
assert_equal(isinstance(actual_results, dict), True)
assert_equal(isinstance(actual_results["logz"][-1], float), True)
assert_equal(isinstance(actual_results["logzerr"][-1], float), True)
def test_negative_lnl(self):
test_rel = relationship.Relationship(utils.straight_line, TEST_X,
TEST_Y)
expected_negtive_lnl = 3
actual_negative_lnl = optimize.negative_lnl([1., 0.],
test_rel.function,
test_rel.x, test_rel.y)
assert_almost_equal(actual_negative_lnl,
expected_negtive_lnl,
decimal=0)
def test_mcmc_with_variable_median_zero(self):
test_rel = relationship.Relationship(utils.straight_line, TEST_X,
TEST_Y)
test_rel.variables[0] = Distribution(np.zeros((7)))
actual_results = sampling.mcmc(test_rel, n_burn=10, n_samples=10)
assert_equal(
isinstance(actual_results["distributions"][0], Distribution), True)
assert_equal(
isinstance(actual_results["distributions"][1], Distribution), True)
assert_equal(actual_results["distributions"][0].size, 500)
assert_equal(actual_results["distributions"][1].size, 500)
def test_mcmc(self):
test_rel = relationship.Relationship(utils.straight_line, TEST_X,
TEST_Y)
test_rel.max_likelihood('mini')
actual_results = sampling.mcmc(test_rel, n_burn=10, n_samples=10)
assert_equal(
isinstance(actual_results["distributions"][0], Distribution), True)
assert_equal(
isinstance(actual_results["distributions"][1], Distribution), True)
assert_equal(actual_results["distributions"][0].size, 500)
assert_equal(actual_results["distributions"][1].size, 500)
def test_nested_sampling_b_with_other_prior(self):
test_rel = relationship.Relationship(utils.straight_line,
TEST_X,
TEST_Y,
bounds=((0, 10), (-1, 1)))
test_rel.max_likelihood('mini')
def other_prior():
"""
Another potential prior.
"""
priors = []
for i, variable in enumerate(test_rel.variables):
loc = variable.n
scale = 1
priors.append(norm(loc=loc, scale=scale))
return priors
actual_results = sampling.nested_sampling(test_rel,
prior_function=other_prior,
maxiter=100)
assert_equal(isinstance(actual_results, dict), True)
assert_equal(isinstance(actual_results["logz"][-1], float), True)
assert_equal(isinstance(actual_results["logzerr"][-1], float), True)
def test_max_lnlikelihood(self):
test_rel = relationship.Relationship(utils.straight_line, TEST_X,
TEST_Y)
actual_best_variables = optimize.max_ln_likelihood(test_rel, 'mini')
assert_almost_equal(actual_best_variables, np.array([1, 0]), decimal=0)