def randomize(self,
label: np.ndarray,
image: Optional[np.ndarray] = None) -> None:
self.spatial_size = fall_back_tuple(self.spatial_size,
default=label.shape[1:])
self.centers = generate_pos_neg_label_crop_centers(
label, self.spatial_size, self.num_samples, self.pos_ratio, image,
self.image_threshold, self.R)
def randomize(
self,
label: np.ndarray,
fg_indices: Optional[np.ndarray] = None,
bg_indices: Optional[np.ndarray] = None,
image: Optional[np.ndarray] = None,
) -> None:
self.spatial_size = fall_back_tuple(self.spatial_size,
default=label.shape[1:])
if np.greater(self.spatial_size, label.shape[1:]).any():
self.centers = [
None,
] * self.num_samples
return
if fg_indices is None or bg_indices is None:
fg_indices_, bg_indices_ = map_binary_to_indices(
label, image, self.image_threshold)
else:
fg_indices_ = fg_indices
bg_indices_ = bg_indices
self.centers = generate_pos_neg_label_crop_centers(
self.spatial_size,
self.num_samples,
self.pos_ratio,
label.shape[1:],
fg_indices_,
bg_indices_,
self.R,
)
self.offset_centers = []
for center in self.centers:
if 0 < self.offset <= 1:
offset = [
self.R.randint(self.offset * sz // 2) *
self.R.choice([1, -1]) for sz in self.spatial_size
]
elif self.offset > 1:
offset = [
self.R.randint(self.offset) * self.R.choice([1, -1])
for sz in self.spatial_size
]
else:
offset = [
0,
] * len(self.spatial_size)
self.offset_centers.append(
[int(c + b) for c, b in zip(center, offset)])
self.centers = self.offset_centers
def randomize(
self,
label: np.ndarray,
fg_indices: Optional[np.ndarray] = None,
bg_indices: Optional[np.ndarray] = None,
image: Optional[np.ndarray] = None,
) -> None:
self.spatial_size = fall_back_tuple(self.spatial_size, default=label.shape[1:])
if fg_indices is None or bg_indices is None:
fg_indices_, bg_indices_ = map_binary_to_indices(label, image, self.image_threshold)
else:
fg_indices_ = fg_indices
bg_indices_ = bg_indices
self.centers = generate_pos_neg_label_crop_centers(
self.spatial_size, self.num_samples, self.pos_ratio, label.shape[1:], fg_indices_, bg_indices_, self.R
)
def randomize(self, label, image):
self.centers = generate_pos_neg_label_crop_centers(
label, self.size, self.num_samples, self.pos_ratio, image,
self.image_threshold, self.R)
def randomize(self, label, image) -> None: # type: ignore # see issue #495
self.spatial_size = fall_back_tuple(self.spatial_size, default=label.shape[1:])
self.centers = generate_pos_neg_label_crop_centers(
label, self.spatial_size, self.num_samples, self.pos_ratio, image, self.image_threshold, self.R
)
def randomize(
self,
label: np.ndarray,
fg_indices: Optional[np.ndarray] = None,
bg_indices: Optional[np.ndarray] = None,
image: Optional[np.ndarray] = None,
) -> None:
self.spatial_size = fall_back_tuple(self.spatial_size,
default=label.shape[1:])
if np.greater(self.spatial_size, label.shape[1:]).any():
self.centers = [
None,
] * self.num_samples
return
# Select subregion to assure valid roi
valid_start = np.floor_divide(self.spatial_size, 2)
# add 1 for random
valid_end = np.subtract(label.shape[1:] + np.array(1),
self.spatial_size / np.array(2)).astype(
np.uint16)
# int generation to have full range on upper side, but subtract unfloored size/2 to prevent rounded range
# from being too high
for i in range(
len(valid_start)
): # need this because np.random.randint does not work with same start and end
if valid_start[i] == valid_end[i]:
valid_end[i] += 1
def _correct_centers(center_ori: List[np.ndarray],
valid_start: np.ndarray,
valid_end: np.ndarray) -> List[np.ndarray]:
for i, c in enumerate(center_ori):
center_i = c
if c < valid_start[i]:
center_i = valid_start[i]
if c >= valid_end[i]:
center_i = valid_end[i] - 1
center_ori[i] = center_i
return center_ori
if fg_indices is None or bg_indices is None:
fg_indices_, bg_indices_ = map_binary_to_indices(
label, image, self.image_threshold)
else:
fg_indices_ = fg_indices
bg_indices_ = bg_indices
self.centers = generate_pos_neg_label_crop_centers(
self.spatial_size,
self.num_samples,
self.pos_ratio,
label.shape[1:],
fg_indices_,
bg_indices_,
self.R,
)
self.offset_centers = []
for center in self.centers:
if 0 < self.offset <= 1:
offset = [
self.R.randint(self.offset * sz // 2) *
self.R.choice([1, -1]) for sz in self.spatial_size
]
elif self.offset > 1:
offset = [
self.R.randint(self.offset) * self.R.choice([1, -1])
for sz in self.spatial_size
]
else:
offset = [
0,
] * len(self.spatial_size)
# print('Offset: ', offset, "Center: ", center)
offset_centers = [int(c + b) for c, b in zip(center, offset)]
self.offset_centers.append(
_correct_centers(offset_centers, valid_start, valid_end))
self.centers = self.offset_centers
def randomize(self, label):
self.centers = generate_pos_neg_label_crop_centers(label, self.size, self.num_samples, self.pos_ratio, self.R)
def test_type_shape(self, input_data, expected_type, expected_count,
expected_shape):
result = generate_pos_neg_label_crop_centers(**input_data)
self.assertIsInstance(result, expected_type)
self.assertEqual(len(result), expected_count)
self.assertEqual(len(result[0]), expected_shape)
