def test_process_before_load(self):
'''
Cannot process before loading data
'''
feat = SectorFeature('test')
self.assertFalse(feat.loaded)
self.assertRaises(ValueError, lambda: feat.process())
def test_process_with_dvh(self):
'''
Can list specific DVH values to compute
'''
# Set up a schema validator
com_schema = {
'bounds':
And(np.ndarray, lambda x: x.shape == (1, 2)),
'mean':
And([float], lambda x: len(x) == 1),
'min':
And([float], lambda x: len(x) == 1),
'max':
And([float], lambda x: len(x) == 1),
'dvh':
And([np.ndarray], lambda x: np.all([d.shape == (6, 2) for d in x]),
lambda x: len(x) == 1)
}
validator = Schema(com_schema)
# Compute features
dvh_vals = [0, 0.2, 0.4, 0.6, 0.8, 1]
feat = SectorFeature('test',
mask=self.mask,
dose=self.dg,
dvh=dvh_vals)
output = feat.process()
# Check the output schema
try:
self.assertTrue(validator.validate(output) is not SchemaError)
except SchemaError:
self.fail('Output does not match given schema')
def test_load_feature(self):
'''
Can initialize with mask and dose grid
'''
feat = SectorFeature('test')
feat.load(self.mask, self.dg)
self.assertTrue(feat.mask is not None)
self.assertTrue(feat.dose is not None)
self.assertTrue(feat.loaded)
def test_process_dose(self):
'''
DVH feature class computes mean, min, max, and dvh
'''
v = Schema([DoseMask])
feat = SectorFeature('test', mask=self.mask, dose=self.dg)
output = feat.process_dose()
try:
self.assertTrue(v.validate(output) is not SchemaError)
except SchemaError:
self.fail('Output does not match given schema')
def test_process_mask(self):
'''
DVH feature class has trivial mask processing
'''
v = Schema([Mask])
feat = SectorFeature('test', mask=self.mask, dose=self.dg)
output = feat.process_mask()
try:
self.assertTrue(v.validate(output) is not SchemaError)
except SchemaError:
self.fail('Output does not match given schema')
def test_create_with_featuretype(self):
'''
Can specify a certain feature type
'''
feat = SectorFeature('test', feature_type='sample_type')
self.assertFalse(feat.loaded)
self.assertEqual(feat.type, 'sample_type')
def test_create_feature_with_values(self):
'''
Can initialize with mask and dose grid
'''
feat = SectorFeature('test', mask=self.mask, dose=self.dg)
self.assertTrue(feat.mask is not None)
self.assertTrue(feat.dose is not None)
self.assertTrue(feat.loaded)
def test_check_elements(self):
'''
Make sure the Feature object has all the required attributes
'''
feat = SectorFeature('test')
elem_list = [
'id', 'type', 'output', 'dose', 'mask', 'feature_mask',
'feature_dosemask', 'loaded'
]
self.assertTrue(isinstance(feat, SectorFeature))
for elem in elem_list:
self.assertTrue(hasattr(feat, elem))
def __helper_test_process(self, a, n):
'''
Helper method to validate slice feature
'''
# Create a schema validator
com_schema = {
'bounds': And(np.ndarray, lambda x: x.shape == (n, 2)),
'mean': And([float], lambda x: len(x) == n),
'min': And([float], lambda x: len(x) == n),
'max': And([float], lambda x: len(x) == n),
'dvh': And([np.ndarray], lambda x: len(x) == n)
}
validator = Schema(com_schema)
# Compute features
feat = SectorFeature('test', mask=self.mask, dose=self.dg, angles=a)
output = feat.process()
# Check the output schema
try:
self.assertTrue(validator.validate(output) is not SchemaError)
except SchemaError:
self.fail('Output does not match given schema')
def test_values_before_load(self):
'''
Cannot get values before loading data
'''
feat = SectorFeature('test')
self.assertRaises(ValueError, lambda: feat.values)
def test_bad_angles_2(self):
'''
Initialize with malformed angles
'''
self.assertRaises(ValueError,
lambda: SectorFeature('test', angles=[[0, 90, 180]]))
def test_good_angles_2(self):
'''
Initialize with malformed angles
'''
feat = SectorFeature('test', angles=[[0, 90], [180, 270]])
self.assertEqual(feat.angles.shape, (2, 2))