def setUp(self):
self.graph = Graph()
self.operator_fill = OpFillMaskArray(graph=self.graph)
self.operator_identity = OpArrayPiper(graph=self.graph)
self.operator_fill.InputArray.meta.axistags = vigra.AxisTags("txyzc")
self.operator_identity.Input.connect(self.operator_fill.Output)
def setUp(self):
self.graph = Graph()
self.operator_fill = OpFillMaskArray(graph=self.graph)
self.operator_identity = OpArrayPiper(graph=self.graph)
self.operator_fill.InputArray.meta.axistags = vigra.AxisTags("txyzc")
self.operator_identity.Input.connect(self.operator_fill.Output)
def setUp(self):
self.graph = Graph()
self.operator_fill = OpFillMaskArray(graph=self.graph)
self.operator_fill.InputArray.meta.axistags = vigra.AxisTags("txyzc")
class TestOpFillMaskArray(object):
def setUp(self):
self.graph = Graph()
self.operator_fill = OpFillMaskArray(graph=self.graph)
self.operator_fill.InputArray.meta.axistags = vigra.AxisTags("txyzc")
def test1(self):
# Generate a random dataset and see if it we get the right masking from the operator.
data = numpy.random.random((4, 5, 6, 7, 3)).astype(numpy.float32)
mask = numpy.zeros(data.shape, dtype=bool)
# Mask borders of the expected output.
left_slicing = (mask.ndim - 1) * (slice(None),) + (slice(None, 1),)
right_slicing = (mask.ndim - 1) * (slice(None),) + (slice(-1, None),)
for i in xrange(mask.ndim):
left_slicing = left_slicing[-1:] + left_slicing[:-1]
right_slicing = right_slicing[-1:] + right_slicing[:-1]
mask[left_slicing] = True
mask[right_slicing] = True
data = numpy.ma.masked_array(
data,
mask=mask,
fill_value=data.dtype.type(numpy.nan),
shrink=False
)
expected_output = data.filled(numpy.nan)
# Provide input read all output.
self.operator_fill.InputArray.setValue(data)
self.operator_fill.InputFillValue.setValue(numpy.nan)
output = self.operator_fill.Output[None].wait()
assert not isinstance(output, numpy.ma.masked_array)
assert (
(expected_output == output) |
(numpy.isnan(expected_output) & numpy.isnan(output))
).all()
def test2(self):
# Generate a dataset and grab chunks of it from the operator. The result should be the same as above.
data = numpy.random.random((4, 5, 6, 7, 3)).astype(numpy.float32)
mask = numpy.zeros(data.shape, dtype=bool)
# Mask borders of the expected output.
left_slicing = (mask.ndim - 1) * (slice(None),) + (slice(None, 1),)
right_slicing = (mask.ndim - 1) * (slice(None),) + (slice(-1, None),)
for i in xrange(mask.ndim):
left_slicing = left_slicing[-1:] + left_slicing[:-1]
right_slicing = right_slicing[-1:] + right_slicing[:-1]
mask[left_slicing] = True
mask[right_slicing] = True
data = numpy.ma.masked_array(
data,
mask=mask,
fill_value=data.dtype.type(numpy.nan),
shrink=False
)
expected_output = data.filled(numpy.nan)
# Create array to store results. Don't keep original data.
output = expected_output.copy()
output[:] = 0
# Provide input and grab chunks.
self.operator_fill.InputArray.setValue(data)
self.operator_fill.InputFillValue.setValue(numpy.nan)
output[:2] = self.operator_fill.Output[:2].wait()
output[2:] = self.operator_fill.Output[2:].wait()
assert not isinstance(output, numpy.ma.masked_array)
assert (
(expected_output == output) |
(numpy.isnan(expected_output) & numpy.isnan(output))
).all()
def test3(self):
# Generate a random dataset and see if it we get the right masking from the operator.
data = numpy.random.random((4, 5, 6, 7, 3)).astype(numpy.float32)
mask = numpy.zeros(data.shape, dtype=bool)
# Mask borders of the expected output.
left_slicing = (mask.ndim - 1) * (slice(None),) + (slice(None, 1),)
right_slicing = (mask.ndim - 1) * (slice(None),) + (slice(-1, None),)
for i in xrange(mask.ndim):
left_slicing = left_slicing[-1:] + left_slicing[:-1]
right_slicing = right_slicing[-1:] + right_slicing[:-1]
mask[left_slicing] = True
mask[right_slicing] = True
data = numpy.ma.masked_array(
data,
mask=mask,
fill_value=data.dtype.type(numpy.nan),
shrink=False
)
expected_output = data.filled(numpy.nan)
# Provide input read all output.
self.operator_fill.InputArray.setValue(data)
output = self.operator_fill.Output[None].wait()
assert not isinstance(output, numpy.ma.masked_array)
assert (
(expected_output == output) |
(numpy.isnan(expected_output) & numpy.isnan(output))
).all()
def test4(self):
# Generate a dataset and grab chunks of it from the operator. The result should be the same as above.
data = numpy.random.random((4, 5, 6, 7, 3)).astype(numpy.float32)
mask = numpy.zeros(data.shape, dtype=bool)
# Mask borders of the expected output.
left_slicing = (mask.ndim - 1) * (slice(None),) + (slice(None, 1),)
right_slicing = (mask.ndim - 1) * (slice(None),) + (slice(-1, None),)
for i in xrange(mask.ndim):
left_slicing = left_slicing[-1:] + left_slicing[:-1]
right_slicing = right_slicing[-1:] + right_slicing[:-1]
mask[left_slicing] = True
mask[right_slicing] = True
data = numpy.ma.masked_array(
data,
mask=mask,
fill_value=data.dtype.type(numpy.nan),
shrink=False
)
expected_output = data.filled(numpy.nan)
# Create array to store results. Don't keep original data.
output = expected_output.copy()
output[:] = 0
# Provide input and grab chunks.
self.operator_fill.InputArray.setValue(data)
output[:2] = self.operator_fill.Output[:2].wait()
output[2:] = self.operator_fill.Output[2:].wait()
assert not isinstance(output, numpy.ma.masked_array)
assert (
(expected_output == output) |
(numpy.isnan(expected_output) & numpy.isnan(output))
).all()
def tearDown(self):
# Take down operators
self.operator_fill.InputArray.disconnect()
self.operator_fill.Output.disconnect()
self.operator_fill.cleanUp()
def setUp(self):
self.graph = Graph()
self.operator_fill = OpFillMaskArray(graph=self.graph)
self.operator_fill.InputArray.meta.axistags = vigra.AxisTags("txyzc")
class TestOpFillMaskArray(object):
def setUp(self):
self.graph = Graph()
self.operator_fill = OpFillMaskArray(graph=self.graph)
self.operator_fill.InputArray.meta.axistags = vigra.AxisTags("txyzc")
def test1(self):
# Generate a random dataset and see if it we get the right masking from the operator.
data = numpy.random.random((4, 5, 6, 7, 3)).astype(numpy.float32)
mask = numpy.zeros(data.shape, dtype=bool)
# Mask borders of the expected output.
left_slicing = (mask.ndim - 1) * (slice(None), ) + (slice(None, 1), )
right_slicing = (mask.ndim - 1) * (slice(None), ) + (slice(-1, None), )
for i in xrange(mask.ndim):
left_slicing = left_slicing[-1:] + left_slicing[:-1]
right_slicing = right_slicing[-1:] + right_slicing[:-1]
mask[left_slicing] = True
mask[right_slicing] = True
data = numpy.ma.masked_array(data,
mask=mask,
fill_value=data.dtype.type(numpy.nan),
shrink=False)
expected_output = data.filled(numpy.nan)
# Provide input read all output.
self.operator_fill.InputArray.setValue(data)
self.operator_fill.InputFillValue.setValue(numpy.nan)
output = self.operator_fill.Output[None].wait()
assert not isinstance(output, numpy.ma.masked_array)
assert ((expected_output == output) |
(numpy.isnan(expected_output) & numpy.isnan(output))).all()
def test2(self):
# Generate a dataset and grab chunks of it from the operator. The result should be the same as above.
data = numpy.random.random((4, 5, 6, 7, 3)).astype(numpy.float32)
mask = numpy.zeros(data.shape, dtype=bool)
# Mask borders of the expected output.
left_slicing = (mask.ndim - 1) * (slice(None), ) + (slice(None, 1), )
right_slicing = (mask.ndim - 1) * (slice(None), ) + (slice(-1, None), )
for i in xrange(mask.ndim):
left_slicing = left_slicing[-1:] + left_slicing[:-1]
right_slicing = right_slicing[-1:] + right_slicing[:-1]
mask[left_slicing] = True
mask[right_slicing] = True
data = numpy.ma.masked_array(data,
mask=mask,
fill_value=data.dtype.type(numpy.nan),
shrink=False)
expected_output = data.filled(numpy.nan)
# Create array to store results. Don't keep original data.
output = expected_output.copy()
output[:] = 0
# Provide input and grab chunks.
self.operator_fill.InputArray.setValue(data)
self.operator_fill.InputFillValue.setValue(numpy.nan)
output[:2] = self.operator_fill.Output[:2].wait()
output[2:] = self.operator_fill.Output[2:].wait()
assert not isinstance(output, numpy.ma.masked_array)
assert ((expected_output == output) |
(numpy.isnan(expected_output) & numpy.isnan(output))).all()
def test3(self):
# Generate a random dataset and see if it we get the right masking from the operator.
data = numpy.random.random((4, 5, 6, 7, 3)).astype(numpy.float32)
mask = numpy.zeros(data.shape, dtype=bool)
# Mask borders of the expected output.
left_slicing = (mask.ndim - 1) * (slice(None), ) + (slice(None, 1), )
right_slicing = (mask.ndim - 1) * (slice(None), ) + (slice(-1, None), )
for i in xrange(mask.ndim):
left_slicing = left_slicing[-1:] + left_slicing[:-1]
right_slicing = right_slicing[-1:] + right_slicing[:-1]
mask[left_slicing] = True
mask[right_slicing] = True
data = numpy.ma.masked_array(data,
mask=mask,
fill_value=data.dtype.type(numpy.nan),
shrink=False)
expected_output = data.filled(numpy.nan)
# Provide input read all output.
self.operator_fill.InputArray.setValue(data)
output = self.operator_fill.Output[None].wait()
assert not isinstance(output, numpy.ma.masked_array)
assert ((expected_output == output) |
(numpy.isnan(expected_output) & numpy.isnan(output))).all()
def test4(self):
# Generate a dataset and grab chunks of it from the operator. The result should be the same as above.
data = numpy.random.random((4, 5, 6, 7, 3)).astype(numpy.float32)
mask = numpy.zeros(data.shape, dtype=bool)
# Mask borders of the expected output.
left_slicing = (mask.ndim - 1) * (slice(None), ) + (slice(None, 1), )
right_slicing = (mask.ndim - 1) * (slice(None), ) + (slice(-1, None), )
for i in xrange(mask.ndim):
left_slicing = left_slicing[-1:] + left_slicing[:-1]
right_slicing = right_slicing[-1:] + right_slicing[:-1]
mask[left_slicing] = True
mask[right_slicing] = True
data = numpy.ma.masked_array(data,
mask=mask,
fill_value=data.dtype.type(numpy.nan),
shrink=False)
expected_output = data.filled(numpy.nan)
# Create array to store results. Don't keep original data.
output = expected_output.copy()
output[:] = 0
# Provide input and grab chunks.
self.operator_fill.InputArray.setValue(data)
output[:2] = self.operator_fill.Output[:2].wait()
output[2:] = self.operator_fill.Output[2:].wait()
assert not isinstance(output, numpy.ma.masked_array)
assert ((expected_output == output) |
(numpy.isnan(expected_output) & numpy.isnan(output))).all()
def tearDown(self):
# Take down operators
self.operator_fill.InputArray.disconnect()
self.operator_fill.Output.disconnect()
self.operator_fill.cleanUp()
