def test_different_results(self):
X = np.random.rand(10, 5)
ss1 = StandardScaler(bounds=(0, 1))
ss1.fit(X)
ss2 = StandardScaler(bounds=(0, 1))
ss2.fit(X)
self.assertFalse(
np.allclose(ss1.mean_, ss2.mean_),
"Arrays %s and %s should be the same" % (ss1.mean_, ss2.mean_))
self.assertFalse(
np.allclose(ss1.var_, ss2.var_),
"Arrays %s and %s should be the same" % (ss1.var_, ss2.var_))
def test_refit(self):
X = np.random.rand(10, 5)
dp_ss = StandardScaler(bounds=(0, 1), epsilon=1)
dp_ss.fit(X)
dp_ss.partial_fit(X)
self.assertIsNotNone(dp_ss)
def test_functionality(self):
X = np.random.rand(10, 5)
ss = StandardScaler(bounds=(0, 1))
ss.fit(X)
self.assertIsNotNone(ss.transform(X))
self.assertIsNotNone(ss.inverse_transform(X))
self.assertIsNotNone(ss.fit_transform(X))
def test_accountant(self):
from diffprivlib.accountant import BudgetAccountant
acc = BudgetAccountant()
X = np.random.rand(10, 5)
ss = StandardScaler(epsilon=1, bounds=(0, 1), accountant=acc)
ss.fit(X)
self.assertEqual((1, 0), acc.total())
with BudgetAccountant(1.5, 0) as acc2:
ss = StandardScaler(epsilon=1, bounds=(0, 1))
ss.fit(X)
self.assertEqual((1, 0), acc2.total())
with self.assertRaises(BudgetError):
ss.fit(X)
self.assertEqual((1, 0), acc.total())
def test_do_nothing(self):
X = np.random.rand(10, 5)
dp_ss = StandardScaler(bounds=(0, 1),
epsilon=1,
with_std=False,
with_mean=False)
dp_ss.fit(X)
self.assertIsNotNone(dp_ss)
def test_sample_weights(self):
X = np.random.rand(10, 5)
dp_ss = StandardScaler(bounds=(0, 1), epsilon=1)
try:
self.assertWarns(DiffprivlibCompatibilityWarning,
dp_ss.fit,
X,
sample_weight=1)
except TypeError:
pass
def test_inf_epsilon(self):
X = np.random.rand(10, 5)
dp_ss = StandardScaler(bounds=(0, 1), epsilon=float("inf"))
dp_ss.fit(X)
sk_ss = sk_pp.StandardScaler()
sk_ss.fit(X)
self.assertTrue(
np.allclose(dp_ss.mean_, sk_ss.mean_),
"Arrays %s and %s should be the same" % (dp_ss.mean_, sk_ss.mean_))
self.assertTrue(
np.allclose(dp_ss.var_, sk_ss.var_),
"Arrays %s and %s should be the same" % (dp_ss.var_, sk_ss.var_))
self.assertTrue(np.all(dp_ss.n_samples_seen_ == sk_ss.n_samples_seen_))
def test_similar_results(self):
global_seed(314159)
X = np.random.rand(100000, 5)
dp_ss = StandardScaler(bounds=(0, 1), epsilon=float("inf"))
dp_ss.fit(X)
sk_ss = sk_pp.StandardScaler()
sk_ss.fit(X)
self.assertTrue(
np.allclose(dp_ss.mean_, sk_ss.mean_, rtol=1, atol=1e-4),
"Arrays %s and %s should be close" % (dp_ss.mean_, sk_ss.mean_))
self.assertTrue(
np.allclose(dp_ss.var_, sk_ss.var_, rtol=1, atol=1e-4),
"Arrays %s and %s should be close" % (dp_ss.var_, sk_ss.var_))
self.assertTrue(np.all(dp_ss.n_samples_seen_ == sk_ss.n_samples_seen_))
def test_no_range(self):
X = np.random.rand(10, 5)
ss = StandardScaler()
with self.assertWarns(PrivacyLeakWarning):
ss.fit(X)
def test_not_none(self):
ss = StandardScaler()
self.assertIsNotNone(ss)