def test_copula_ndarray_feature_generation():
n_samples = 200
n_features = 100
input_data = np.random.normal(size=(n_samples, n_features))
generator = syn.CopulaDataGenerator(verbose=True)
generator.fit(input_data, copula=syn.GaussianCopula(),
parameterize_by=syn.QuantileParameterizer(n_quantiles=100))
pickled = pickle.dumps(generator)
generator = pickle.loads(pickled)
n_synthetic_samples = 50
synthetic_data = generator.generate(n_samples=n_synthetic_samples)
assert synthetic_data.shape == (n_synthetic_samples, n_features)
assert isinstance(synthetic_data, np.ndarray)
def test_vine_copula_feature_generation():
n_samples = 200
n_features = 100
input_data = xr.DataArray(np.random.normal(size=(n_samples, n_features)))
generator = syn.CopulaDataGenerator(verbose=True)
ctrl = pv.FitControlsVinecop(family_set=[pv.gaussian], trunc_lvl=1, select_trunc_lvl=False, show_trace=True)
generator.fit(input_data, copula=syn.VineCopula(controls=ctrl),
parameterize_by=syn.QuantileParameterizer(n_quantiles=100))
pickled = pickle.dumps(generator)
generator = pickle.loads(pickled)
n_synthetic_samples = 50
synthetic_data = generator.generate(n_samples=n_synthetic_samples)
assert synthetic_data.shape == (n_synthetic_samples, n_features)
assert isinstance(synthetic_data, xr.DataArray)
assert generator.generate(10, seed=42).equals(generator.generate(10, seed=42))
def test_independent_feature_generation_with_distribution():
n_samples = 20
n_features = 2
input_data = np.random.normal(size=(n_samples, n_features))
generator = syn.CopulaDataGenerator()
dist_names = set(syn.DistributionParameterizer.get_dist_names())
# Remove all very slow distributions
dist_names -= set(['genexpon', 'levy_stable', 'recipinvgauss', 'vonmises', 'kstwo'])
generator.fit(input_data, copula=syn.IndependenceCopula(),
parameterize_by=syn.DistributionParameterizer(dist_names, verbose=True))
pickled = pickle.dumps(generator)
generator = pickle.loads(pickled)
n_synthetic_samples = 50
synthetic_data = generator.generate(n_samples=n_synthetic_samples)
assert synthetic_data.shape == (n_synthetic_samples, n_features)
assert isinstance(synthetic_data, np.ndarray)
def test_copula_dataset_generation():
n_samples = 200
n_features = [10, 20]
input_data = xr.Dataset({
'a': (('sample', 'foo'), np.random.normal(size=(n_samples, n_features[0]))),
'b': (('sample', 'bar'), np.random.normal(size=(n_samples, n_features[1])))
})
generator = syn.CopulaDataGenerator()
generator.fit(input_data, copula=syn.GaussianCopula(),
parameterize_by=syn.QuantileParameterizer(n_quantiles=100))
pickled = pickle.dumps(generator)
generator = pickle.loads(pickled)
n_synthetic_samples = 50
synthetic_data = generator.generate(n_samples=n_synthetic_samples)
assert synthetic_data['a'].shape == (n_synthetic_samples, n_features[0])
assert synthetic_data['b'].shape == (n_synthetic_samples, n_features[1])
assert generator.generate(10, seed=42).equals(generator.generate(10, seed=42))
def test_vine_copula_with_categorical():
a = np.array([[0, 1, 2], [3, 4, 5]])
b = np.array([6, 7])
input_data = xr.Dataset({
'a': (('sample', 'foo'), a),
'b': (('sample'), b)
})
generator = syn.CopulaDataGenerator(verbose=True)
ctrl = pv.FitControlsVinecop(family_set=[pv.gaussian], trunc_lvl=1, select_trunc_lvl=False, show_trace=True)
generator.fit(input_data, types={ 'b': 'cat' }, copula=syn.VineCopula(controls=ctrl))
pickled = pickle.dumps(generator)
generator = pickle.loads(pickled)
n_synthetic_samples = 50
synthetic_data = generator.generate(n_samples=n_synthetic_samples)
assert synthetic_data['a'].shape == (n_synthetic_samples, 3)
assert synthetic_data['b'].shape == (n_synthetic_samples,)
assert generator.generate(10, seed=42).equals(generator.generate(10, seed=42))
def main():
import pyvinecopulib as pv
n_samples = 100
n_features = 50
n_synthetic_samples = 500
input_data = np.random.normal(size=(n_samples, n_features))
generator = syn.CopulaDataGenerator()
t0 = time.time()
generator.fit(input_data, syn.GaussianCopula())
generator.generate(n_synthetic_samples)
print(f'synthia: {time.time() - t0}s')
t0 = time.time()
generator.fit(
input_data,
syn.VineCopula(
pv.FitControlsVinecop(family_set=[pv.gaussian],
trunc_lvl=1,
select_trunc_lvl=False)))
generator.generate(n_synthetic_samples)
print(f'pyvinecopulib: {time.time() - t0}s')