def test_apply_single_many(self):
expected_values = np.array([145.0590587, 354.4284063, 32.746848])
lower = 10.0
upper = 99.0
epsilon = 1.0
anonymizer = DiffPrivSumAnonymizer(epsilon, lower, upper)
self.set_seed()
values = anonymizer.apply(87.0, size=3)
np.testing.assert_almost_equal(values, expected_values)
def test_apply_multiple(self):
expected_values = np.array([145.0590587, 702.4284063])
lower = 10.0
upper = 99.0
epsilon = 1.0
anonymizer = DiffPrivSumAnonymizer(epsilon, lower, upper)
self.set_seed()
values = anonymizer.apply([87.0, 435.0])
np.testing.assert_almost_equal(values, expected_values)
def test_apply_single(self):
expected_value = 145.05905871186388
lower = 10.0
upper = 99.0
epsilon = 1.0
anonymizer = DiffPrivSumAnonymizer(epsilon, lower, upper)
self.set_seed()
value = anonymizer.apply(87.0)
np.testing.assert_almost_equal(value, expected_value)
def test_scale_getter(self):
lower = 10.0
upper = 99.0
epsilon = 1.0
gs_value = np.max([np.abs(lower), np.abs(upper)])
scale = gs_value / epsilon
anonymizer = DiffPrivSumAnonymizer(epsilon, lower, upper)
self.assertEqual(anonymizer.scale, scale)
def test_global_sensitivity_getter(self):
lower = 10.0
upper = 99.0
epsilon = 1.0
gs_value = np.max([np.abs(lower), np.abs(upper)])
anonymizer = DiffPrivSumAnonymizer(epsilon, lower, upper)
np.testing.assert_almost_equal(anonymizer.global_sensitivity.value,
gs_value)
def create_sum_anonymizer(cls, epsilon, lower, upper):
"""
Creates a sum anonymizer instance.
Parameters
----------
epsilon : float
The privacy budget.
lower : float
The lower bound of the data.
upper : float
The upper bound of the data.
Returns
-------
`DiffPrivSumAnonymizer`
A sum anonymizer with the provided privacy budget.
"""
anonymizer = DiffPrivSumAnonymizer(epsilon, lower, upper)
return anonymizer
def test_epsilon_getter(self):
lower = 10.0
upper = 99.0
epsilon = 1.0
anonymizer = DiffPrivSumAnonymizer(epsilon, lower, upper)
self.assertEqual(anonymizer.epsilon, epsilon)