Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix cupy pytest errors #368

Merged
merged 3 commits into from
May 15, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions tests/test_dask_image/test_ndfilters/test__threshold.py
Original file line number Diff line number Diff line change
Expand Up @@ -98,6 +98,7 @@ def test_threshold_local_generic(simple_test_image):
assert np.allclose(unchanged.compute(), simple_test_image.compute())
assert np.allclose(out.compute(), ref)


def test_threshold_local_generic_invalid(simple_test_image):
expected_error_message = "Must include a valid function to use as "
"the 'param' keyword argument."
Expand All @@ -117,6 +118,7 @@ def test_nan_blocksize(simple_test_image, method, block_size, error_type):
with pytest.raises(error_type):
threshold_local(simple_test_image, block_size, method=method)


def test_invalid_threshold_method(simple_test_image):
with pytest.raises(ValueError):
threshold_local(simple_test_image, 3, method='invalid')
211 changes: 113 additions & 98 deletions tests/test_dask_image/test_ndfilters/test_cupy_threshold.py
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,20 @@

cupy = pytest.importorskip("cupy", minversion="5.0.0")

@pytest.fixture
def simple_test_image():
image = da.from_array(cupy.array(
[[0, 0, 1, 3, 5],
[0, 1, 4, 3, 4],
[1, 2, 5, 4, 1],
[2, 4, 5, 2, 1],
[4, 5, 1, 0, 0]], dtype=int), chunks=(5, 5))
return image


# ==================================================
# Test Threshold Filters
# ==================================================

@pytest.mark.cupy
@pytest.mark.parametrize('block_size', [
Expand All @@ -18,101 +32,102 @@
da.from_array(np.array([3, 3]), chunks=1),
da.from_array(np.array([3, 3]), chunks=2),
])
class TestSimpleImage:
def setup(self):
self.image = da.from_array(cupy.array(
[[0, 0, 1, 3, 5],
[0, 1, 4, 3, 4],
[1, 2, 5, 4, 1],
[2, 4, 5, 2, 1],
[4, 5, 1, 0, 0]], dtype=int), chunks=(5, 5))

def test_threshold_local_gaussian(self, block_size):
ref = np.array(
[[False, False, False, False, True],
[False, False, True, False, True],
[False, False, True, True, False],
[False, True, True, False, False],
[True, True, False, False, False]]
)
out = threshold_local(self.image, block_size, method='gaussian')
cupy.testing.assert_array_equal(ref, (self.image > out).compute())

out = threshold_local(self.image, block_size, method='gaussian',
param=1./3.)
cupy.testing.assert_array_equal(ref, (self.image > out).compute())

def test_threshold_local_mean(self, block_size):
ref = cupy.array(
[[False, False, False, False, True],
[False, False, True, False, True],
[False, False, True, True, False],
[False, True, True, False, False],
[True, True, False, False, False]]
)
out = threshold_local(self.image, block_size, method='mean')
cupy.testing.assert_array_equal(ref, (self.image > out).compute())

def test_threshold_local_median(self, block_size):
ref = cupy.array(
[[False, False, False, False, True],
[False, False, True, False, False],
[False, False, True, False, False],
[False, False, True, True, False],
[False, True, False, False, False]]
)
out = threshold_local(self.image, block_size, method='median')
cupy.testing.assert_array_equal(ref, (self.image > out).compute())


class TestGenericFilter:
def setup(self):
self.image = da.from_array(cupy.array(
[[0, 0, 1, 3, 5],
[0, 1, 4, 3, 4],
[1, 2, 5, 4, 1],
[2, 4, 5, 2, 1],
[4, 5, 1, 0, 0]], dtype=int), chunks=(5, 5))

def test_threshold_local_generic(self):
ref = cupy.array(
[[1., 7., 16., 29., 37.],
[5., 14., 23., 30., 30.],
[13., 24., 30., 29., 21.],
[25., 29., 28., 19., 10.],
[34., 31., 23., 10., 4.]]
)
my_sum = cupy.ReductionKernel(
'T x', 'T out', 'x', 'a + b', 'out = a', '0', 'my_sum')
unchanged = threshold_local(self.image, 1, method='generic', param=my_sum) # noqa: E501
out = threshold_local(self.image, 3, method='generic', param=my_sum)
assert cupy.allclose(unchanged.compute(), self.image.compute())
assert cupy.allclose(out.compute(), ref)

def test_threshold_local_generic_invalid(self):
expected_error_message = "Must include a valid function to use as "
"the 'param' keyword argument."
with pytest.raises(ValueError) as e:
threshold_local(self.image, 3, method='generic', param='sum')
assert e == expected_error_message


class TestInvalidArguments:
def setup(self):
self.image = da.from_array(cupy.array(
[[0, 0, 1, 3, 5],
[0, 1, 4, 3, 4],
[1, 2, 5, 4, 1],
[2, 4, 5, 2, 1],
[4, 5, 1, 0, 0]], dtype=int), chunks=(5, 5))

@pytest.mark.parametrize("method, block_size, error_type", [
('median', cupy.nan, TypeError),
])
def test_nan_blocksize(self, method, block_size, error_type):
with pytest.raises(error_type):
threshold_local(self.image, block_size, method=method)

def test_invalid_threshold_method(self):
with pytest.raises(ValueError):
threshold_local(self.image, 3, method='invalid')
def test_threshold_local_gaussian(simple_test_image, block_size):
ref = np.array(
[[False, False, False, False, True],
[False, False, True, False, True],
[False, False, True, True, False],
[False, True, True, False, False],
[True, True, False, False, False]]
)
out = threshold_local(simple_test_image, block_size, method='gaussian')
cupy.testing.assert_array_equal(ref, (simple_test_image > out).compute())

out = threshold_local(simple_test_image, block_size, method='gaussian',
param=1./3.)
cupy.testing.assert_array_equal(ref, (simple_test_image > out).compute())


@pytest.mark.cupy
@pytest.mark.parametrize('block_size', [
3,
[3, 3],
np.array([3, 3]),
da.from_array(np.array([3, 3]), chunks=1),
da.from_array(np.array([3, 3]), chunks=2),
])
def test_threshold_local_mean(simple_test_image, block_size):
ref = cupy.array(
[[False, False, False, False, True],
[False, False, True, False, True],
[False, False, True, True, False],
[False, True, True, False, False],
[True, True, False, False, False]]
)
out = threshold_local(simple_test_image, block_size, method='mean')
cupy.testing.assert_array_equal(ref, (simple_test_image > out).compute())


@pytest.mark.cupy
@pytest.mark.parametrize('block_size', [
3,
[3, 3],
np.array([3, 3]),
da.from_array(np.array([3, 3]), chunks=1),
da.from_array(np.array([3, 3]), chunks=2),
])
def test_threshold_local_median(simple_test_image, block_size):
ref = cupy.array(
[[False, False, False, False, True],
[False, False, True, False, False],
[False, False, True, False, False],
[False, False, True, True, False],
[False, True, False, False, False]]
)
out = threshold_local(simple_test_image, block_size, method='median')
cupy.testing.assert_array_equal(ref, (simple_test_image > out).compute())


# ==================================================
# Test Generic Filters
# ==================================================

def test_threshold_local_generic(simple_test_image):
ref = cupy.array(
[[1., 7., 16., 29., 37.],
[5., 14., 23., 30., 30.],
[13., 24., 30., 29., 21.],
[25., 29., 28., 19., 10.],
[34., 31., 23., 10., 4.]]
)
my_sum = cupy.ReductionKernel(
'T x', 'T out', 'x', 'a + b', 'out = a', '0', 'my_sum')
unchanged = threshold_local(simple_test_image, 1, method='generic', param=my_sum) # noqa: E501
out = threshold_local(simple_test_image, 3, method='generic', param=my_sum)
assert cupy.allclose(unchanged.compute(), simple_test_image.compute())
assert cupy.allclose(out.compute(), ref)


def test_threshold_local_generic_invalid(simple_test_image):
expected_error_message = "Must include a valid function to use as "
"the 'param' keyword argument."
with pytest.raises(ValueError) as e:
threshold_local(simple_test_image, 3, method='generic', param='sum')
assert e == expected_error_message


# ==================================================
# Test Invalid Arguments
# ==================================================

@pytest.mark.parametrize("method, block_size, error_type", [
('median', cupy.nan, TypeError),
])
def test_nan_blocksize(simple_test_image, method, block_size, error_type):
with pytest.raises(error_type):
threshold_local(simple_test_image, block_size, method=method)


def test_invalid_threshold_method(simple_test_image):
with pytest.raises(ValueError):
threshold_local(simple_test_image, 3, method='invalid')