def test__get_univariate_error(self):
"""If something else is passed, rasie a TypeError."""
distribution = 123
ct = GaussianCopulaTransformer(distribution=distribution)
with pytest.raises(TypeError):
ct._get_univariate()
def test__get_univariate_class(self):
"""If a class is passed, create an instance without args."""
distribution = copulas.univariate.Univariate
ct = GaussianCopulaTransformer(distribution=distribution)
univariate = ct._get_univariate()
assert isinstance(univariate, copulas.univariate.Univariate)
def test__get_univariate_instance(self):
"""If a univariate instance is passed, make a copy."""
distribution = copulas.univariate.Univariate()
ct = GaussianCopulaTransformer(distribution=distribution)
univariate = ct._get_univariate()
assert univariate is not distribution
assert isinstance(univariate, copulas.univariate.Univariate)
assert dir(univariate) == dir(distribution)
def test_int_nan(self):
data = np.array([1, 2, 1, 2, 1, np.nan])
ct = GaussianCopulaTransformer(dtype=int)
transformed = ct.fit_transform(data)
assert isinstance(transformed, np.ndarray)
assert transformed.shape == (6, 2)
reverse = ct.reverse_transform(transformed)
np.testing.assert_array_almost_equal(reverse, data, decimal=2)
def test_int(self):
data = np.array([1, 2, 1, 2, 1])
ct = GaussianCopulaTransformer(dtype=int)
transformed = ct.fit_transform(data)
assert isinstance(transformed, np.ndarray)
assert transformed.shape == (5, )
reverse = ct.reverse_transform(transformed)
assert list(reverse) == [1, 2, 1, 2, 1]
def test__get_univariate_tuple(self):
"""If a tuple is passed, create an instance using the given args."""
distribution = (copulas.univariate.Univariate, {
'candidates': 'a_candidates_list'
})
ct = GaussianCopulaTransformer(distribution=distribution)
univariate = ct._get_univariate()
assert isinstance(univariate, copulas.univariate.Univariate)
assert univariate.candidates == 'a_candidates_list'
def test_not_null_column(self):
data = np.array([1, 2, 1, 2, np.nan, 1])
ct = GaussianCopulaTransformer(null_column=False)
transformed = ct.fit_transform(data)
assert isinstance(transformed, np.ndarray)
assert transformed.shape == (6, )
reverse = ct.reverse_transform(transformed)
np.testing.assert_array_almost_equal(reverse, data, decimal=2)
def test_stats(self):
data = np.random.normal(loc=4, scale=4, size=1000)
ct = GaussianCopulaTransformer()
transformed = ct.fit_transform(data)
assert isinstance(transformed, np.ndarray)
assert transformed.shape == (1000, )
np.testing.assert_almost_equal(transformed.mean(), 0, decimal=1)
np.testing.assert_almost_equal(transformed.std(), 1, decimal=1)
reverse = ct.reverse_transform(transformed)
np.testing.assert_array_almost_equal(reverse, data, decimal=1)
def test___init__super_attrs(self):
"""super() arguments are properly passed and set as attributes."""
ct = GaussianCopulaTransformer(dtype='int', nan='mode', null_column=False)
assert ct.dtype == 'int'
assert ct.nan == 'mode'
assert ct.null_column is False
def test___init__non_distr(self):
"""If distribution is not an str, it is store as given."""
univariate = copulas.univariate.Univariate()
ct = GaussianCopulaTransformer(distribution=univariate)
assert ct._distribution is univariate
def test___init__str_distr(self):
"""If distribution is an str, it is resolved using the _DISTRIBUTIONS dict."""
ct = GaussianCopulaTransformer(distribution='univariate')
assert ct._distribution is copulas.univariate.Univariate