def test_average_precision_real_data(self, ovule_label):
label = ovule_label[64:128, 64:128, 64:128]
ltb = LabelToAffinities((1, 2, 4, 6), aggregate_affinities=True)
pred = ltb(label)
label = torch.tensor(label.reshape((1, 1) + label.shape).astype('int64'))
pred = torch.tensor(np.expand_dims(pred, 0))
ap = BoundaryAveragePrecision()
assert ap(pred, label) > 0.5
def test_average_precision_real_data(self):
l_file = 'resources/sample_patch.h5'
with h5py.File(l_file, 'r') as f:
label = f['big_label'][...]
ltb = LabelToAffinities((1, 2, 4, 6), aggregate_affinities=True)
pred = ltb(label)
# don't compare instances smaller than 25K voxels
ap = BoundaryAveragePrecision(min_instance_size=25000)
assert ap(pred, label) > 0.5
def test_average_precision_synthethic_data(self):
input = np.zeros((64, 200, 200), dtype=np.int)
for i in range(40, 200, 40):
input[:, :, i:i + 2] = 1
for i in range(40, 200, 40):
input[:, i:i + 2, :] = 1
for i in range(40, 64, 40):
input[i:i + 2, :, :] = 1
target = measure.label(np.logical_not(input).astype(np.int), background=0)
input = torch.tensor(input.reshape((1, 1) + input.shape))
target = torch.tensor(target.reshape((1, 1) + target.shape))
ap = BoundaryAveragePrecision()
# expect perfect AP
assert ap(input, target) == 1.0
def test_average_precision_synthethic_data(self):
input = np.zeros((64, 200, 200), dtype=np.int)
for i in range(40, 200, 40):
input[:, :, i:i + 2] = 1
for i in range(40, 200, 40):
input[:, i:i + 2, :] = 1
for i in range(40, 64, 40):
input[i:i + 2, :, :] = 1
target = measure.label(np.logical_not(input).astype(np.int),
background=0)
input = np.expand_dims(input, axis=0)
ap = BoundaryAveragePrecision()
# expect perfect AP
assert ap(input, target) == 1.0