def test_bgr(self, device):
a_val: float = 1.
x_rgb = torch.rand(3, 4, 4).to(device)
x_bgr = kornia.rgb_to_bgr(x_rgb)
x_rgba = kornia.rgb_to_rgba(x_rgb, a_val)
x_rgba_new = kornia.bgr_to_rgba(x_bgr, a_val)
assert_allclose(x_rgba, x_rgba_new)
def test_unit(self, device, dtype, aval):
data = torch.tensor([[[[1., 1.], [1., 1.]], [[2., 2.], [2., 2.]],
[[3., 3.], [3., 3.]]]],
device=device,
dtype=dtype) # Bx3x2x2
expected = torch.tensor(
[[[[1.0, 1.0], [1.0, 1.0]], [[2.0, 2.0], [2.0, 2.0]],
[[3.0, 3.0], [3.0, 3.0]], [[aval, aval], [aval, aval]]]],
device=device,
dtype=dtype) # Bx4x2x2
assert_allclose(kornia.rgb_to_rgba(data, aval), expected)
def test_single(self, device):
data = torch.tensor([[[1., 1.], [1., 1.]], [[2., 2.], [2., 2.]],
[[3., 3.], [3., 3.]]]) # 3x2x2
data = data.to(device)
aval: float = 0.4
expected = torch.tensor([[[1.0, 1.0], [1.0, 1.0]],
[[2.0, 2.0], [2.0, 2.0]],
[[3.0, 3.0], [3.0, 3.0]],
[[0.4, 0.4], [0.4, 0.4]]]) # 4x2x2
expected = expected.to(device)
assert_allclose(kornia.rgb_to_rgba(data, aval), expected)
def test_unit_aval_th(self, device, dtype, aval):
data = torch.tensor(
[[[[1.0, 1.0], [1.0, 1.0]], [[2.0, 2.0], [2.0, 2.0]],
[[3.0, 3.0], [3.0, 3.0]]]],
device=device,
dtype=dtype) # Bx3x2x2
expected = torch.tensor(
[[
[[1.0, 1.0], [1.0, 1.0]],
[[2.0, 2.0], [2.0, 2.0]],
[[3.0, 3.0], [3.0, 3.0]],
[[aval, aval], [aval, aval]],
]],
device=device,
dtype=dtype,
) # Bx4x2x2
aval = torch.full_like(data[:, :1], aval) # Bx1xHxW
assert_close(kornia.rgb_to_rgba(data, aval), expected)
def test_batch(self, device):
data = torch.tensor([[[[1., 1.], [1., 1.]], [[2., 2.], [2., 2.]],
[[3., 3.], [3., 3.]]],
[[[1., 1.], [1., 1.]], [[2., 2.], [2., 2.]],
[[3., 3.], [3., 3.]]]]) # 2x3x2x2
data = data.to(device)
aval: float = 45.
expected = torch.tensor([[[[1.0, 1.0], [1.0, 1.0]],
[[2.0, 2.0], [2.0, 2.0]],
[[3.0, 3.0], [3.0, 3.0]],
[[45., 45.], [45., 45.]]],
[[[1.0, 1.0], [1.0, 1.0]],
[[2.0, 2.0], [2.0, 2.0]],
[[3.0, 3.0], [3.0, 3.0]],
[[45., 45.], [45., 45.]]]])
expected = expected.to(device)
assert_allclose(kornia.rgb_to_rgba(data, aval), expected)
def test_back_and_forth_rgb(self, device):
a_val: float = 1.
x_rgb = torch.rand(3, 4, 4).to(device)
x_rgba = kornia.rgb_to_rgba(x_rgb, a_val)
x_rgb_new = kornia.rgba_to_rgb(x_rgba)
assert_allclose(x_rgb, x_rgb_new)
def test_smoke(self, device):
data = torch.rand(3, 4, 4).to(device)
assert kornia.rgb_to_rgba(data, 0.).shape == (4, 4, 4)
