def test_memory(self, input_param, expected_key_size):
input_data = dict()
for i in range(50):
input_data[str(i)] = [time.time()] * 100000
result = DeleteItemsd(**input_param)(input_data)
self.assertEqual(len(result.keys()), expected_key_size)
self.assertGreaterEqual(
sys.getsizeof(input_data) * float(expected_key_size) /
len(input_data), sys.getsizeof(result))
def test_memory(self, input_param, expected_key_size):
input_data = {"image": {}} if "sep" in input_param else {}
for i in range(50):
if "sep" in input_param:
input_data["image"][str(i)] = [time.time()] * 100000
else:
input_data[str(i)] = [time.time()] * 100000
result = DeleteItemsd(**input_param)(input_data)
if "sep" in input_param:
self.assertEqual(len(result["image"].keys()), expected_key_size)
else:
self.assertEqual(len(result.keys()), expected_key_size)
self.assertGreaterEqual(
sys.getsizeof(input_data) * float(expected_key_size) / len(input_data), sys.getsizeof(result)
)
def get_xforms(mode="train",
keys=("image", "label"),
keys2=("image", "label", "synthesis"),
path_synthesis='',
decreasing_sequence='',
scans_syns=[],
texture=[],
GEN=15):
"""returns a composed transform for train/val/infer."""
xforms = [
LoadImaged(keys),
AddChanneld(keys),
Orientationd(keys, axcodes="LPS"),
Spacingd(keys,
pixdim=(1.25, 1.25, 5.0),
mode=("bilinear", "nearest")[:len(keys)]),
ScaleIntensityRanged(keys[0],
a_min=-1000.0,
a_max=500.0,
b_min=0.0,
b_max=1.0,
clip=True),
]
if mode == "train":
xforms.extend([
SpatialPadd(keys, spatial_size=(192, 192, -1),
mode="reflect"), # ensure at least 192x192
RandAffined(
keys,
prob=0.15,
rotate_range=(
0.05, 0.05, None
), # 3 parameters control the transform on 3 dimensions
scale_range=(0.1, 0.1, None),
mode=("bilinear", "nearest"),
as_tensor_output=False,
),
RandCropByPosNegLabeld(keys,
label_key=keys[1],
spatial_size=(192, 192, 16),
num_samples=3),
RandGaussianNoised(keys[0], prob=0.15, std=0.01),
RandFlipd(keys, spatial_axis=0, prob=0.5),
RandFlipd(keys, spatial_axis=1, prob=0.5),
RandFlipd(keys, spatial_axis=2, prob=0.5),
])
dtype = (np.float32, np.uint8)
if mode == "synthesis":
print('DOING SYNTHESIS')
xforms.extend([
CopyItemsd(keys, 1, names=['image_1', 'label_1']),
# PrintTypesShapes(keys, '======SHAPE LOAD'),
SpatialPadd(keys, spatial_size=(192, 192, -1),
mode="reflect"), # ensure at least 192x192
RandCropByPosNegLabeld(keys,
label_key=keys[1],
spatial_size=(192, 192, 16),
num_samples=3),
TransCustom(keys,
path_synthesis,
read_cea_aug_slice2,
pseudo_healthy_with_texture,
scans_syns,
decreasing_sequence,
GEN=GEN,
POST_PROCESS=True,
mask_outer_ring=True,
texture=np.empty(shape=(456, 456)),
new_value=.5),
RandAffined(
keys2,
prob=0.15,
rotate_range=(
0.05, 0.05, None
), # 3 parameters control the transform on 3 dimensions
scale_range=(0.1, 0.1, None),
mode=("bilinear", "nearest", "bilinear"),
# mode=("bilinear", "nearest"),
as_tensor_output=False),
RandGaussianNoised((keys2[0], keys2[2]), prob=0.15, std=0.01),
# RandGaussianNoised(keys[0], prob=0.15, std=0.01),
RandFlipd(keys, spatial_axis=0, prob=0.5),
RandFlipd(keys, spatial_axis=1, prob=0.5),
RandFlipd(keys, spatial_axis=2, prob=0.5),
TransCustom2(keys2),
SpatialPadd(keys, spatial_size=(192, 192, -1), mode="reflect"),
DeleteItemsd(('image_1', 'label_1')),
# PrintTypesShapes(('image_1', 'label_1','synthetic_lesion', 'synthetic_label'), '=syn='),
])
dtype = (np.float32, np.uint8)
if mode == "val":
dtype = (np.float32, np.uint8)
if mode == "infer":
dtype = (np.float32, )
xforms.extend([CastToTyped(keys, dtype=dtype), EnsureTyped(keys)])
return monai.transforms.Compose(xforms)
def test_re(self, input_param):
input_data = {"image": [1, 2, 3], PostFix.meta(): {"0008|0005": 1, "0008|1050": 2, "0008test": 3}}
result = DeleteItemsd(**input_param)(input_data)
self.assertEqual(result[PostFix.meta()]["0008test"], 3)
self.assertTrue(len(result[PostFix.meta()]), 1)