def test_check_params(self):
with self.assertRaises(ValueError):
_ = GaussianAugmentation(augmentation=True, ratio=-1)
with self.assertRaises(ValueError):
_ = GaussianAugmentation(clip_values=(0, 1, 2))
with self.assertRaises(ValueError):
_ = GaussianAugmentation(clip_values=(1, 0))
def test_failure_augmentation_fit_predict(self):
# Assert that value error is raised
with self.assertRaises(ValueError) as context:
_ = GaussianAugmentation(augmentation=True, apply_fit=False, apply_predict=True)
self.assertTrue(
"If `augmentation` is `True`, then `apply_fit` must be `True` and `apply_predict`"
" must be `False`." in str(context.exception)
)
with self.assertRaises(ValueError) as context:
_ = GaussianAugmentation(augmentation=True, apply_fit=False, apply_predict=False)
self.assertIn(
"If `augmentation` is `True`, then `apply_fit` and `apply_predict` can't be both `False`.",
str(context.exception),
)
def test_multiple_size(self):
x = np.arange(12).reshape((4, 3))
x_original = x.copy()
ga = GaussianAugmentation(ratio=3.5)
x_new, _ = ga(x)
self.assertEqual(int(4.5 * x.shape[0]), x_new.shape[0])
# Check that x has not been modified by attack and classifier
self.assertAlmostEqual(float(np.max(np.abs(x_original - x))), 0.0, delta=0.00001)
def test_labels(self):
x = np.arange(12).reshape((4, 3))
y = np.arange(8).reshape((4, 2))
ga = GaussianAugmentation()
x_new, new_y = ga(x, y)
self.assertTrue(x_new.shape[0] == new_y.shape[0] == 8)
self.assertEqual(x_new.shape[1:], x.shape[1:])
self.assertEqual(new_y.shape[1:], y.shape[1:])
def test_no_augmentation(self):
x = np.arange(12).reshape((4, 3))
ga = GaussianAugmentation(augmentation=False)
x_new, _ = ga(x)
self.assertEqual(x.shape, x_new.shape)
self.assertFalse((x == x_new).all())
def test_small_size(self):
x = np.arange(15).reshape((5, 3))
ga = GaussianAugmentation(ratio=0.4)
x_new, _ = ga(x)
self.assertEqual(x_new.shape, (7, 3))