class TestVine(TestCase):
def setUp(self):
data = pd.DataFrame({
'column1': np.array([
2641.16233666, 921.14476418, -651.32239137, 1223.63536668,
3233.37342355, 1373.22400821, 1959.28188858, 1076.99295365,
2029.25100261, 1835.52188141, 1170.03850556, 739.42628394,
1866.65810627, 3703.49786503, 1719.45232017, 258.90206528,
219.42363944, 609.90212377, 1618.44207239, 2323.2775272,
3251.78732274, 1430.63989981, -180.57028875, -592.84497457,
]),
'column2': np.array([
180.2425623, 192.35609972, 150.24830291, 156.62123653,
173.80311908, 191.0922843, 163.22252158, 190.73280428,
158.52982435, 163.0101334, 205.24904026, 175.42916046,
208.31821984, 178.98351969, 160.50981075, 163.19294974,
173.30395132, 215.18996298, 164.71141696, 178.84973821,
182.99902513, 217.5796917, 201.56983421, 174.92272693
]),
'column3': np.array([
-1.42432446, -0.14759864, 0.66476302, -0.04061445, 0.64305762,
1.79615407, 0.70450457, -0.05886671, -0.36794788, 1.39331262,
0.39792831, 0.0676313, -0.96761759, 0.67286132, -0.55013279,
-0.53118328, 1.23969655, -0.35985016, -0.03568531, 0.91456357,
0.49077378, -0.27428204, 0.45857406, 2.29614033
])
})
self.rvine = VineCopula(TreeTypes.REGULAR)
self.rvine.fit(data)
self.cvine = VineCopula(TreeTypes.CENTER)
self.cvine.fit(data)
self.dvine = VineCopula(TreeTypes.DIRECT)
self.dvine.fit(data)
def test_get_likelihood(self):
uni_matrix = np.array([[0.1, 0.2, 0.3, 0.4]])
# FIX ME: there is some randomness in rvine, will do another test
rvalue = self.rvine.get_likelihood(uni_matrix)
expected = -0.2859820599667698
assert abs(rvalue - expected) < 10E-3
cvalue = self.cvine.get_likelihood(uni_matrix)
expected = -0.27565584158521045
assert abs(cvalue - expected) < 10E-3
dvalue = self.dvine.get_likelihood(uni_matrix)
expected = -0.27565584158521045
assert abs(dvalue - expected) < 10E-3
def test_to_dict(self):
""" """
# Setup
instance = VineCopula('regular')
instance.fitted = True
instance.n_sample = 100
instance.n_var = 10
instance.depth = 3
instance.truncated = 3
tree = Tree('regular')
instance.trees = [tree]
uni = KDEUnivariate()
instance.unis = [uni]
tau_mat = np.array([
[0, 1],
[1, 0]
])
instance.tau_mat = tau_mat
u_matrix = np.array([
[0, 1],
[1, 0]
])
instance.u_matrix = u_matrix
expected_result = {
'type': 'copulas.multivariate.vine.VineCopula',
'fitted': True,
'vine_type': 'regular',
'n_sample': 100,
'n_var': 10,
'depth': 3,
'truncated': 3,
'trees': [
{
'type': 'copulas.multivariate.tree.RegularTree',
'tree_type': 'regular',
'fitted': False
}
],
'tau_mat': [
[0, 1],
[1, 0]
],
'u_matrix': [
[0, 1],
[1, 0]
],
'unis': [
{
'type': 'copulas.univariate.kde.KDEUnivariate',
'fitted': False
}
]
}
# Run
result = instance.to_dict()
# Check
assert result == expected_result
def test_from_dict(self):
# Setup
vine_dict = {
'type': 'copulas.multivariate.vine.VineCopula',
'vine_type': 'regular',
'fitted': True,
'n_sample': 100,
'n_var': 10,
'depth': 3,
'truncated': 3,
'trees': [
{
'type': 'copulas.multivariate.tree.RegularTree',
'tree_type': 'regular',
'fitted': False
}
],
'tau_mat': [
[0, 1],
[1, 0]
],
'u_matrix': [
[0, 1],
[1, 0]
],
'unis': [
{
'type': 'copulas.univariate.kde.KDEUnivariate',
'fitted': False
}
]
}
# Run
instance = Multivariate.from_dict(vine_dict)
# Check
assert instance.vine_type == 'regular'
assert instance.n_sample == 100
assert instance.n_var == 10
assert instance.depth == 3
assert instance.truncated == 3
assert len(instance.trees) == 1
assert instance.trees[0].to_dict() == Tree('regular').to_dict()
assert (instance.tau_mat == np.array([
[0, 1],
[1, 0]
])).all()
assert (instance.u_matrix == np.array([
[0, 1],
[1, 0]
])).all()
def test_serialization_unfitted_model(self):
# Setup
instance = VineCopula('regular')
# Run
result = VineCopula.from_dict(instance.to_dict())
# Check
assert result.to_dict() == instance.to_dict()
def test_serialization_fit_model(self):
# Setup
instance = VineCopula('regular')
X = pd.DataFrame(data=[
[1, 0, 0],
[0, 1, 0],
[0, 0, 1]
])
instance.fit(X)
# Run
result = VineCopula.from_dict(instance.to_dict())
# Check
compare_nested_dicts(result.to_dict(), instance.to_dict())
def test_sample(self):
"""After being fit, a vine can sample new data."""
# Setup
vine = VineCopula(TreeTypes.REGULAR)
X = pd.DataFrame([
[1, 0, 0, 0],
[0, 1, 0, 0],
[0, 0, 1, 0],
[0, 0, 0, 1]
])
vine.fit(X)
# Run
result = vine.sample()
# Check
assert len(result) == vine.n_var
class TestVine(TestCase):
def setUp(self):
data = pd.DataFrame({
'column1':
np.array([
2641.16233666,
921.14476418,
-651.32239137,
1223.63536668,
3233.37342355,
1373.22400821,
1959.28188858,
1076.99295365,
2029.25100261,
1835.52188141,
1170.03850556,
739.42628394,
1866.65810627,
3703.49786503,
1719.45232017,
258.90206528,
219.42363944,
609.90212377,
1618.44207239,
2323.2775272,
3251.78732274,
1430.63989981,
-180.57028875,
-592.84497457,
]),
'column2':
np.array([
180.2425623, 192.35609972, 150.24830291, 156.62123653,
173.80311908, 191.0922843, 163.22252158, 190.73280428,
158.52982435, 163.0101334, 205.24904026, 175.42916046,
208.31821984, 178.98351969, 160.50981075, 163.19294974,
173.30395132, 215.18996298, 164.71141696, 178.84973821,
182.99902513, 217.5796917, 201.56983421, 174.92272693
]),
'column3':
np.array([
-1.42432446, -0.14759864, 0.66476302, -0.04061445, 0.64305762,
1.79615407, 0.70450457, -0.05886671, -0.36794788, 1.39331262,
0.39792831, 0.0676313, -0.96761759, 0.67286132, -0.55013279,
-0.53118328, 1.23969655, -0.35985016, -0.03568531, 0.91456357,
0.49077378, -0.27428204, 0.45857406, 2.29614033
])
})
self.rvine = VineCopula(TreeTypes.REGULAR)
self.rvine.fit(data)
self.cvine = VineCopula(TreeTypes.CENTER)
self.cvine.fit(data)
self.dvine = VineCopula(TreeTypes.DIRECT)
self.dvine.fit(data)
def test_get_likelihood(self):
uni_matrix = np.array([[0.1, 0.2, 0.3, 0.4]])
# FIX ME: there is some randomness in rvine, will do another test
rvalue = self.rvine.get_likelihood(uni_matrix)
expected = -0.26888124854583245
assert abs(rvalue - expected) < 10E-3
cvalue = self.cvine.get_likelihood(uni_matrix)
expected = -0.27565584158521045
assert abs(cvalue - expected) < 10E-3
dvalue = self.dvine.get_likelihood(uni_matrix)
expected = -0.27565584158521045
assert abs(dvalue - expected) < 10E-3
def test_serialization_unfitted_model(self):
"""An unfitted vine can be serialized and deserialized and kept unchanged."""
# Setup
instance = VineCopula('regular')
# Run
result = VineCopula.from_dict(instance.to_dict())
# Check
assert result.to_dict() == instance.to_dict()
def test_serialization_fit_model(self):
"""A fitted vine can be serialized and deserialized and kept unchanged."""
# Setup
instance = VineCopula('regular')
X = pd.DataFrame(data=[[1, 0, 0], [0, 1, 0], [0, 0, 1]])
instance.fit(X)
# Run
result = VineCopula.from_dict(instance.to_dict())
# Check
compare_nested_dicts(result.to_dict(), instance.to_dict())
@patch('copulas.multivariate.vine.np.random.randint', autospec=True)
@patch('copulas.multivariate.vine.np.random.uniform', autospec=True)
@pytest.mark.skipif(sys.version_info > (3, 8),
reason="Fails on py38. To be reviewed.")
def test_sample_row(self, uniform_mock, randint_mock):
"""After being fit, a vine can sample new data."""
# Setup
instance = VineCopula(TreeTypes.REGULAR)
X = pd.DataFrame(
[[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]],
columns=list('ABCD'))
instance.fit(X)
uniform_mock.return_value = np.array([0.1, 0.25, 0.5, 0.75])
randint_mock.return_value = 1
expected_result = np.array(
[-0.3196499, -0.16358588, 0.418420, 1.5688347])
# Run
result = instance._sample_row()
# Check
compare_nested_iterables(result, expected_result)
uniform_mock.assert_called_once_with(0, 1, 4)
randint_mock.assert_called_once_with(0, 4)
@patch('copulas.multivariate.vine.VineCopula._sample_row', autospec=True)
def test_sample(self, sample_mock):
"""After being fit, a vine can sample new data."""
# Setup
vine = VineCopula(TreeTypes.REGULAR)
X = pd.DataFrame(
[[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]],
columns=list('ABCD'))
vine.fit(X)
expected_result = pd.DataFrame([
{
'A': 1,
'B': 2,
'C': 3,
'D': 4
},
{
'A': 1,
'B': 2,
'C': 3,
'D': 4
},
{
'A': 1,
'B': 2,
'C': 3,
'D': 4
},
{
'A': 1,
'B': 2,
'C': 3,
'D': 4
},
{
'A': 1,
'B': 2,
'C': 3,
'D': 4
},
])
sample_mock.return_value = np.array([1, 2, 3, 4])
# Run
result = vine.sample(5)
# Check
compare_nested_iterables(result, expected_result)
assert sample_mock.call_count == 5
def test_sample_random_state(self):
"""When random_state is set, the generated samples are always the same."""
# Setup
vine = VineCopula(TreeTypes.REGULAR, random_seed=0)
X = pd.DataFrame([[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, 0],
[0, 0, 0, 1]])
vine.fit(X)
expected_result = pd.DataFrame(
[[0.101933, 0.527734, 0.080266, 0.078328]], columns=range(4))
# Run
result = vine.sample(1)
# Check
compare_nested_iterables(result, expected_result)
class TestVine(TestCase):
def setUp(self):
data = pd.DataFrame({
'column1':
np.array([
2641.16233666,
921.14476418,
-651.32239137,
1223.63536668,
3233.37342355,
1373.22400821,
1959.28188858,
1076.99295365,
2029.25100261,
1835.52188141,
1170.03850556,
739.42628394,
1866.65810627,
3703.49786503,
1719.45232017,
258.90206528,
219.42363944,
609.90212377,
1618.44207239,
2323.2775272,
3251.78732274,
1430.63989981,
-180.57028875,
-592.84497457,
]),
'column2':
np.array([
180.2425623, 192.35609972, 150.24830291, 156.62123653,
173.80311908, 191.0922843, 163.22252158, 190.73280428,
158.52982435, 163.0101334, 205.24904026, 175.42916046,
208.31821984, 178.98351969, 160.50981075, 163.19294974,
173.30395132, 215.18996298, 164.71141696, 178.84973821,
182.99902513, 217.5796917, 201.56983421, 174.92272693
]),
'column3':
np.array([
-1.42432446, -0.14759864, 0.66476302, -0.04061445, 0.64305762,
1.79615407, 0.70450457, -0.05886671, -0.36794788, 1.39331262,
0.39792831, 0.0676313, -0.96761759, 0.67286132, -0.55013279,
-0.53118328, 1.23969655, -0.35985016, -0.03568531, 0.91456357,
0.49077378, -0.27428204, 0.45857406, 2.29614033
])
})
self.rvine = VineCopula(TreeTypes.REGULAR)
self.rvine.fit(data)
self.cvine = VineCopula(TreeTypes.CENTER)
self.cvine.fit(data)
self.dvine = VineCopula(TreeTypes.DIRECT)
self.dvine.fit(data)
def test_get_likelihood(self):
uni_matrix = np.array([[0.1, 0.2, 0.3, 0.4]])
# FIX ME: there is some randomness in rvine, will do another test
rvalue = self.rvine.get_likelihood(uni_matrix)
expected = -0.26888124854583245
assert abs(rvalue - expected) < 10E-3
cvalue = self.cvine.get_likelihood(uni_matrix)
expected = -0.27565584158521045
assert abs(cvalue - expected) < 10E-3
dvalue = self.dvine.get_likelihood(uni_matrix)
expected = -0.27565584158521045
assert abs(dvalue - expected) < 10E-3
def test_to_dict(self):
"""to_dict returns the internal parameters to replicate one instance."""
# Setup
instance = VineCopula('regular')
instance.fitted = True
instance.n_sample = 100
instance.n_var = 10
instance.depth = 3
instance.truncated = 3
tree = Tree('regular')
instance.trees = [tree]
uni = KDEUnivariate()
instance.unis = [uni]
tau_mat = np.array([[0, 1], [1, 0]])
instance.tau_mat = tau_mat
u_matrix = np.array([[0, 1], [1, 0]])
instance.u_matrix = u_matrix
expected_result = {
'type':
'copulas.multivariate.vine.VineCopula',
'fitted':
True,
'vine_type':
'regular',
'n_sample':
100,
'n_var':
10,
'depth':
3,
'truncated':
3,
'trees': [{
'type': 'copulas.multivariate.tree.RegularTree',
'tree_type': 'regular',
'fitted': False
}],
'tau_mat': [[0, 1], [1, 0]],
'u_matrix': [[0, 1], [1, 0]],
'unis': [{
'type': 'copulas.univariate.kde.KDEUnivariate',
'fitted': False,
'constant_value': None
}]
}
# Run
result = instance.to_dict()
# Check
assert result == expected_result
def test_from_dict(self):
"""from_dict creates a new instance from its parameters."""
# Setup
vine_dict = {
'type':
'copulas.multivariate.vine.VineCopula',
'vine_type':
'regular',
'fitted':
True,
'n_sample':
100,
'n_var':
10,
'depth':
3,
'truncated':
3,
'trees': [{
'type': 'copulas.multivariate.tree.RegularTree',
'tree_type': 'regular',
'fitted': False
}],
'tau_mat': [[0, 1], [1, 0]],
'u_matrix': [[0, 1], [1, 0]],
'unis': [{
'type': 'copulas.univariate.kde.KDEUnivariate',
'fitted': False,
'constant_value': None
}]
}
# Run
instance = Multivariate.from_dict(vine_dict)
# Check
assert instance.vine_type == 'regular'
assert instance.n_sample == 100
assert instance.n_var == 10
assert instance.depth == 3
assert instance.truncated == 3
assert len(instance.trees) == 1
assert instance.trees[0].to_dict() == Tree('regular').to_dict()
assert (instance.tau_mat == np.array([[0, 1], [1, 0]])).all()
assert (instance.u_matrix == np.array([[0, 1], [1, 0]])).all()
def test_serialization_unfitted_model(self):
"""An unfitted vine can be serialized and deserialized and kept unchanged."""
# Setup
instance = VineCopula('regular')
# Run
result = VineCopula.from_dict(instance.to_dict())
# Check
assert result.to_dict() == instance.to_dict()
def test_serialization_fit_model(self):
"""A fitted vine can be serialized and deserialized and kept unchanged."""
# Setup
instance = VineCopula('regular')
X = pd.DataFrame(data=[[1, 0, 0], [0, 1, 0], [0, 0, 1]])
instance.fit(X)
# Run
result = VineCopula.from_dict(instance.to_dict())
# Check
compare_nested_dicts(result.to_dict(), instance.to_dict())
@patch('copulas.multivariate.vine.np.random.randint', autospec=True)
@patch('copulas.multivariate.vine.np.random.uniform', autospec=True)
def test_sample_row(self, uniform_mock, randint_mock):
"""After being fit, a vine can sample new data."""
# Setup
instance = VineCopula(TreeTypes.REGULAR)
X = pd.DataFrame(
[[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]],
columns=list('ABCD'))
instance.fit(X)
uniform_mock.return_value = np.array([0.1, 0.25, 0.5, 0.75])
randint_mock.return_value = 1
expected_result = np.array(
[-1.63155227, -0.16358589, -1.63155227, -1.62583869])
# Run
result = instance._sample_row()
# Check
compare_nested_iterables(result, expected_result)
uniform_mock.assert_called_once_with(0, 1, 4)
randint_mock.assert_called_once_with(0, 4)
@patch('copulas.multivariate.vine.VineCopula._sample_row', autospec=True)
def test_sample(self, sample_mock):
"""After being fit, a vine can sample new data."""
# Setup
vine = VineCopula(TreeTypes.REGULAR)
X = pd.DataFrame(
[[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]],
columns=list('ABCD'))
vine.fit(X)
expected_result = pd.DataFrame([
{
'A': 1,
'B': 2,
'C': 3,
'D': 4
},
{
'A': 1,
'B': 2,
'C': 3,
'D': 4
},
{
'A': 1,
'B': 2,
'C': 3,
'D': 4
},
{
'A': 1,
'B': 2,
'C': 3,
'D': 4
},
{
'A': 1,
'B': 2,
'C': 3,
'D': 4
},
])
sample_mock.return_value = np.array([1, 2, 3, 4])
# Run
result = vine.sample(5)
# Check
compare_nested_iterables(result, expected_result)
assert sample_mock.call_count == 5
def test_sample_random_state(self):
"""When random_state is set, the generated samples are always the same."""
# Setup
vine = VineCopula(TreeTypes.REGULAR, random_seed=0)
X = pd.DataFrame([[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, 0],
[0, 0, 0, 1]])
vine.fit(X)
expected_result = pd.DataFrame([[
-1.6315522689646478, 0.527734420510573, -1.6315522689646478,
-1.6315522689646478
]],
columns=range(4))
# Run
result = vine.sample(1)
# Check
assert result.equals(expected_result)
class TestVine(TestCase):
def setUp(self):
data = pd.DataFrame({
'column1':
np.array([
2641.16233666,
921.14476418,
-651.32239137,
1223.63536668,
3233.37342355,
1373.22400821,
1959.28188858,
1076.99295365,
2029.25100261,
1835.52188141,
1170.03850556,
739.42628394,
1866.65810627,
3703.49786503,
1719.45232017,
258.90206528,
219.42363944,
609.90212377,
1618.44207239,
2323.2775272,
3251.78732274,
1430.63989981,
-180.57028875,
-592.84497457,
]),
'column2':
np.array([
180.2425623, 192.35609972, 150.24830291, 156.62123653,
173.80311908, 191.0922843, 163.22252158, 190.73280428,
158.52982435, 163.0101334, 205.24904026, 175.42916046,
208.31821984, 178.98351969, 160.50981075, 163.19294974,
173.30395132, 215.18996298, 164.71141696, 178.84973821,
182.99902513, 217.5796917, 201.56983421, 174.92272693
]),
'column3':
np.array([
-1.42432446, -0.14759864, 0.66476302, -0.04061445, 0.64305762,
1.79615407, 0.70450457, -0.05886671, -0.36794788, 1.39331262,
0.39792831, 0.0676313, -0.96761759, 0.67286132, -0.55013279,
-0.53118328, 1.23969655, -0.35985016, -0.03568531, 0.91456357,
0.49077378, -0.27428204, 0.45857406, 2.29614033
])
})
self.rvine = VineCopula('regular')
self.rvine.fit(data)
self.cvine = VineCopula('center')
self.cvine.fit(data)
self.dvine = VineCopula('direct')
self.dvine.fit(data)
def test_get_likelihood(self):
uni_matrix = np.array([[0.1, 0.2, 0.3, 0.4]])
# FIX ME: there is some randomness in rvine, will do another test
rvalue = self.rvine.get_likelihood(uni_matrix)
expected = -0.2859820599667698
assert abs(rvalue - expected) < 10E-3
cvalue = self.cvine.get_likelihood(uni_matrix)
expected = -0.27565584158521045
assert abs(cvalue - expected) < 10E-3
dvalue = self.dvine.get_likelihood(uni_matrix)
expected = -0.27565584158521045
assert abs(dvalue - expected) < 10E-3
@expectedFailure
def test_sample(self):
sample_r = self.rvine.sample()
sample_c = self.cvine.sample()
sample_d = self.cvine.sample()
assert len(sample_r) == 4
assert len(sample_c) == 4
assert len(sample_d) == 4
