def test_alpha(self, im_shape, input_type):
im = self.get_data(im_shape, input_type)
alpha = [0.5, 0.51]
t = RandGibbsNoise(1.0, alpha)
_ = t(deepcopy(im))
self.assertGreaterEqual(t.sampled_alpha, 0.5)
self.assertLessEqual(t.sampled_alpha, 0.51)
def test_same_result(self, im_shape, input_type):
im = self.get_data(im_shape, input_type)
alpha = [0.5, 0.8]
t = RandGibbsNoise(1.0, alpha)
t.set_random_state(42)
out1 = t(deepcopy(im))
t.set_random_state(42)
out2 = t(deepcopy(im))
assert_allclose(out1, out2, rtol=1e-7, atol=1e-2, type_test="tensor")
def test_same_result(self, im_shape, input_type):
im = self.get_data(im_shape, input_type)
alpha = [0.5, 0.8]
t = RandGibbsNoise(1.0, alpha)
t.set_random_state(42)
out1 = t(deepcopy(im))
t.set_random_state(42)
out2 = t(deepcopy(im))
torch.testing.assert_allclose(out1, out2, rtol=1e-7, atol=0)
self.assertIsInstance(out1, type(im))
def test_same_result(self, im_shape, as_tensor_output, as_tensor_input):
im = self.get_data(im_shape, as_tensor_input)
alpha = [0.5, 0.8]
t = RandGibbsNoise(1.0, alpha, as_tensor_output)
t.set_random_state(42)
out1 = t(deepcopy(im))
t.set_random_state(42)
out2 = t(deepcopy(im))
np.testing.assert_allclose(out1, out2)
self.assertIsInstance(out1, torch.Tensor if as_tensor_output else np.ndarray)
def test_alpha_1(self, im_shape, input_type):
im = self.get_data(im_shape, input_type)
alpha = [1.0, 1.0]
t = RandGibbsNoise(1.0, alpha)
out = t(deepcopy(im))
torch.testing.assert_allclose(0 * im, out, rtol=1e-7, atol=0)
def test_identity(self, im_shape, input_type):
im = self.get_data(im_shape, input_type)
alpha = [0.0, 0.0]
t = RandGibbsNoise(1.0, alpha)
out = t(deepcopy(im))
torch.testing.assert_allclose(im, out, atol=1e-2, rtol=1e-7)
def test_0_prob(self, im_shape, input_type):
im = self.get_data(im_shape, input_type)
alpha = [0.5, 1.0]
t = RandGibbsNoise(0.0, alpha)
out = t(im)
torch.testing.assert_allclose(im, out, rtol=1e-7, atol=0)
def test_alpha_1(self, im_shape, input_type):
im = self.get_data(im_shape, input_type)
alpha = [1.0, 1.0]
t = RandGibbsNoise(1.0, alpha)
out = t(deepcopy(im))
assert_allclose(out, 0 * im, rtol=1e-7, atol=1e-2, type_test="tensor")
def test_0_prob(self, im_shape, input_type):
im = self.get_data(im_shape, input_type)
alpha = [0.5, 1.0]
t = RandGibbsNoise(0.0, alpha)
out = t(im)
assert_allclose(out, im, rtol=1e-7, atol=0, type_test="tensor")
def test_alpha_1(self, im_shape, _, as_tensor_input):
im = self.get_data(im_shape, as_tensor_input)
alpha = [1.0, 1.0]
t = RandGibbsNoise(1.0, alpha)
out = t(deepcopy(im))
np.testing.assert_allclose(0 * im, out)
def test_identity(self, im_shape, _, as_tensor_input):
im = self.get_data(im_shape, as_tensor_input)
alpha = [0.0, 0.0]
t = RandGibbsNoise(1.0, alpha)
out = t(deepcopy(im))
np.testing.assert_allclose(im, out, atol=1e-2)
def test_0_prob(self, im_shape, as_tensor_output, as_tensor_input):
im = self.get_data(im_shape, as_tensor_input)
alpha = [0.5, 1.0]
t = RandGibbsNoise(0.0, alpha, as_tensor_output)
out = t(im)
np.testing.assert_allclose(im, out)