def test_sensitivity_norm_list_of_arrays():
class ReshapeToList(Query):
def get(self, data):
return list(np.reshape(data, (20, 30, 40)))
distribution = NormalDistribution(0, 1.5)
sampler = SensitivitySampler()
# L1 norm:
s_max, s_mean = sampler.sample_sensitivity(ReshapeToList(),
L1SensitivityNorm(),
distribution,
n=20 * 30 * 40,
m=285,
gamma=0.33)
assert isinstance(s_max, list)
assert isinstance(s_mean, list)
for i in range(len(s_max)):
assert s_max[i].sum() < 2 * 1.5
assert s_mean[i].sum() < 2 * 1.5
# L2 norm:
s_max, s_mean = sampler.sample_sensitivity(ReshapeToList(),
L2SensitivityNorm(),
distribution,
n=20 * 30 * 40,
m=285,
gamma=0.33)
assert isinstance(s_max, list)
assert isinstance(s_mean, list)
for i in range(len(s_max)):
assert s_max[i].sum() < 2 * 1.5
assert s_mean[i].sum() < 2 * 1.5
def test_sample_sensitivity_gamma():
distribution = NormalDistribution(0, 1)
sampler = SensitivitySampler()
_, mean = sampler.sample_sensitivity(Mean(),
L1SensitivityNorm(),
distribution,
n=100,
gamma=0.33)
assert np.abs(mean - 0) < 0.5
def test_laplace_mechanism():
data_size = 1000
array = NormalDistribution(175, 7).sample(data_size)
federated_array = shfl.private.federated_operation.federate_array(
array, data_size)
federated_array.configure_data_access(LaplaceMechanism(1, 1))
result = federated_array.query()
differences = 0
for i in range(data_size):
if result[i] != array[i]:
differences = differences + 1
assert differences == data_size
assert np.mean(array) - np.mean(result) < 5
def test_normal_distribution():
data_size = 1000
array = NormalDistribution(175, 7).sample(data_size)
assert len(array) == 1000
assert np.abs(np.mean(array) - 175) < 5