def test_exception(self, device, dtype):
input = torch.ones(1, 1, 3, 4, device=device, dtype=dtype)
kernel = torch.ones(3, 3, device=device, dtype=dtype)
with pytest.raises(TypeError):
assert open([0.], kernel)
with pytest.raises(TypeError):
assert open(input, [0.])
with pytest.raises(ValueError):
test = torch.ones(2, 3, 4, device=device, dtype=dtype)
assert open(test, kernel)
with pytest.raises(ValueError):
test = torch.ones(2, 3, 4, device=device, dtype=dtype)
assert open(input, test)
def test_exception(self, dev_str, dtype_str, call):
input_ = ivy.ones((1, 1, 3, 4), dev_str=dev_str, dtype_str=dtype_str)
kernel = ivy.ones((3, 3), dev_str=dev_str, dtype_str=dtype_str)
with pytest.raises(ValueError):
test = ivy.ones((2, 3, 4), dev_str=dev_str, dtype_str=dtype_str)
assert open(test, kernel)
with pytest.raises(ValueError):
test = ivy.ones((2, 3, 4), dev_str=dev_str, dtype_str=dtype_str)
assert open(input_, test)
if call is not helpers.torch_call:
return
with pytest.raises(TypeError):
assert open([0.], kernel)
with pytest.raises(TypeError):
assert open(input_, [0.])
def test_value(self, device, dtype):
input = torch.tensor(
[[0.5, 1., 0.3], [0.7, 0.3, 0.8], [0.4, 0.9, 0.2]],
device=device,
dtype=dtype)[None, None, :, :]
kernel = torch.tensor([[0., 1., 0.], [1., 1., 1.], [0., 1., 0.]],
device=device,
dtype=dtype)
expected = torch.tensor(
[[0.5, 0.5, 0.3], [0.5, 0.3, 0.3], [0.4, 0.4, 0.2]],
device=device,
dtype=dtype)[None, None, :, :]
assert_allclose(open(input, kernel), expected)
def top_hat(tensor: torch.Tensor, kernel: torch.Tensor) -> torch.Tensor:
r"""
Returns the top hat tranformation of an image,
(that means, image - opened_image) applying the same kernel in each channel.
The kernel must have 2 dimensions, each one defined by an odd number.
See :class:`~kornia.morphology.open` for details.
Args
tensor (torch.Tensor): Image with shape :math:`(B, C, H, W)`.
kernel (torch.Tensor): Structuring element with shape :math:`(H, W)`.
Returns:
torch.Tensor: Top hat transformated image with shape :math:`(B, C, H, W)`.
Example:
>>> tensor = torch.rand(1, 3, 5, 5)
>>> kernel = torch.ones(3, 3)
>>> top_hat_img = top_hat(tensor, kernel)
"""
if not isinstance(tensor, torch.Tensor):
raise TypeError("Input type is not a torch.Tensor. Got {}".format(
type(tensor)))
if len(tensor.shape) != 4:
raise ValueError("Input size must have 4 dimensions. Got {}".format(
tensor.dim()))
if not isinstance(kernel, torch.Tensor):
raise TypeError("Kernel type is not a torch.Tensor. Got {}".format(
type(kernel)))
if len(kernel.shape) != 2:
raise ValueError("Kernel size must have 2 dimensions. Got {}".format(
kernel.dim()))
return tensor - open(tensor, kernel)
def top_hat(tensor: ivy.Array, kernel: ivy.Array) -> ivy.Array:
r"""Returns the top hat tranformation of an image.
That means, (image - opened_image) applying the same kernel in each channel.
The kernel must have 2 dimensions, each one defined by an odd number.
See :class:`~kornia.morphology.open` for details.
Args:
tensor (ivy.Array): Image with shape :math:`(B, C, H, W)`.
kernel (ivy.Array): Structuring element with shape :math:`(H, W)`.
Returns:
ivy.Array: Top hat transformated image with shape :math:`(B, C, H, W)`.
Example:
>>> tensor = ivy.random_uniform(shape=(1, 3, 5, 5))
>>> kernel = ivy.ones((3, 3))
>>> top_hat_img = top_hat(tensor, kernel)
"""
if ivy.backend == 'torch' and not isinstance(tensor, ivy.Array):
raise TypeError("Input type is not a ivy.Array. Got {}".format(
type(tensor)))
if len(tensor.shape) != 4:
raise ValueError("Input size must have 4 dimensions. Got {}".format(
ivy.get_num_dims(tensor)))
if ivy.backend == 'torch' and not isinstance(kernel, ivy.Array):
raise TypeError("Kernel type is not a ivy.Array. Got {}".format(
type(kernel)))
if len(kernel.shape) != 2:
raise ValueError("Kernel size must have 2 dimensions. Got {}".format(
ivy.get_num_dims(kernel)))
return tensor - open(tensor, kernel)
def test_cardinality(self, dev_str, dtype_str, call, shape, kernel):
img = ivy.ones(shape, dtype_str=dtype_str, dev_str=dev_str)
krnl = ivy.ones(kernel, dtype_str=dtype_str, dev_str=dev_str)
assert open(img, krnl).shape == shape
def test_cardinality(self, device, dtype, shape, kernel):
img = torch.ones(shape, device=device, dtype=dtype)
krnl = torch.ones(kernel, device=device, dtype=dtype)
assert open(img, krnl).shape == shape