def test_from_dict(self):
"""from_dict generates a new instance from its parameters."""
# Setup
copula = GaussianMultivariate()
copula.fit(self.data)
copula_dict = copula.to_dict()
# Run
new_copula = GaussianMultivariate.from_dict(copula_dict)
# Asserts
assert isinstance(new_copula, GaussianMultivariate)
assert new_copula.columns == ['column1', 'column2', 'column3']
assert len(new_copula.univariates) == 3
for new_univariate, old_univariate in zip(copula.univariates, new_copula.univariates):
assert new_univariate.to_dict() == old_univariate.to_dict()
def test_from_dict(self):
""" """
# Setup
covariance = [[
1.006711409395973, -0.11010327176239865, 0.8776048563471857,
0.823443255069628
],
[
-0.11010327176239865, 1.006711409395972,
-0.4233383520816991, -0.3589370029669186
],
[
0.8776048563471857, -0.4233383520816991,
1.006711409395973, 0.9692185540781536
],
[
0.823443255069628, -0.3589370029669186,
0.9692185540781536, 1.0067114093959735
]]
parameters = {
'covariance': covariance,
'distribs': {
'feature_01': {
'mean': 5.843333333333334,
'std': 0.8253012917851409
},
'feature_02': {
'mean': 3.0540000000000003,
'std': 0.4321465800705435
},
'feature_03': {
'mean': 3.758666666666666,
'std': 1.7585291834055212
},
'feature_04': {
'mean': 1.1986666666666668,
'std': 0.7606126185881716
}
}
}
# Run
copula = GaussianMultivariate.from_dict(parameters)
# Check
assert (copula.covariance == [[
1.006711409395973, -0.11010327176239865, 0.8776048563471857,
0.823443255069628
],
[
-0.11010327176239865,
1.006711409395972,
-0.4233383520816991,
-0.3589370029669186
],
[
0.8776048563471857,
-0.4233383520816991,
1.006711409395973, 0.9692185540781536
],
[
0.823443255069628,
-0.3589370029669186,
0.9692185540781536,
1.0067114093959735
]]).all()
for name, distrib in copula.distribs.items():
assert copula.distribs[name].to_dict(
) == parameters['distribs'][name]
# This isn't to check the sampling, but that the copula is able to run.
assert copula.sample(10).all().all()
def test_from_dict(self):
"""from_dict generates a new instance from its parameters."""
# Setup
covariance = [[
1.006711409395973, -0.11010327176239865, 0.8776048563471857,
0.823443255069628
],
[
-0.11010327176239865, 1.006711409395972,
-0.4233383520816991, -0.3589370029669186
],
[
0.8776048563471857, -0.4233383520816991,
1.006711409395973, 0.9692185540781536
],
[
0.823443255069628, -0.3589370029669186,
0.9692185540781536, 1.0067114093959735
]]
parameters = {
'covariance':
covariance,
'fitted':
True,
'type':
'copulas.multivariate.gaussian.GaussianMultivariate',
'distribution':
'copulas.univariate.gaussian.GaussianUnivariate',
'columns':
['feature_01', 'feature_02', 'feature_03', 'feature_04'],
'univariates': [{
'type': 'copulas.univariate.gaussian.GaussianUnivariate',
'mean': 5.843333333333334,
'std': 0.8253012917851409,
'fitted': True,
}, {
'type': 'copulas.univariate.gaussian.GaussianUnivariate',
'mean': 3.0540000000000003,
'std': 0.4321465800705435,
'fitted': True,
}, {
'type': 'copulas.univariate.gaussian.GaussianUnivariate',
'mean': 3.758666666666666,
'std': 1.7585291834055212,
'fitted': True,
}, {
'type': 'copulas.univariate.gaussian.GaussianUnivariate',
'mean': 1.1986666666666668,
'std': 0.7606126185881716,
'fitted': True,
}]
}
# Run
copula = GaussianMultivariate.from_dict(parameters)
# Check
assert (copula.covariance == covariance).all()
for a, b in zip(parameters['columns'], copula.columns):
assert a == b
for a, b in zip(parameters['univariates'], copula.univariates):
assert a == b.to_dict()
# This isn't to check the sampling, but that the copula is able to run.
assert copula.sample(10).all().all()
