def __init__(self, padding=None):
super(PaddingPropertyHolder, self).__init__()
self.padding = validate_parameter(padding, ALLOWED_PADDINGS, "z")
def __init__(self, interpolation=None):
super(InterpolationPropertyHolder, self).__init__()
self.interpolation = validate_parameter(interpolation,
ALLOWED_INTERPOLATIONS,
"bilinear")
def test_parameter_validation_raises_error_when_default_value_is_wrong_type():
with pytest.raises(TypeError):
slu.validate_parameter(None, {1, 2}, ("10", "inherit"), int)
def test_validate_parameter_raises_value_errors(parameter):
with pytest.raises(ValueError):
slu.validate_parameter(parameter, {1, 2}, 1, basic_type=int)
def test_parameter_validation_raises_error_when_default_type_is_wrong():
with pytest.raises(ValueError):
slu.validate_parameter(None, {1, 2}, (10, "12345"), int)
def test_parameter_validation_raises_error_when_types_dont_match():
with pytest.raises(NotImplementedError):
slu.validate_parameter({1, 2}, 10, int)
def __init__(self, data, fmt, transform_settings=None):
if len(fmt) == 1 and not isinstance(data, tuple):
if not isinstance(data, list):
data = (data, )
else:
raise TypeError
if not isinstance(data, tuple):
raise TypeError
if len(data) != len(fmt):
raise ValueError
if transform_settings is not None:
if not isinstance(transform_settings, dict):
raise TypeError
else:
transform_settings = {}
# Element-wise settings
# If no settings provided for certain items, they will be created
for idx in range(len(data)):
if idx not in transform_settings:
transform_settings[idx] = {}
if fmt[idx] == "I" or fmt[idx] == "M":
val = ("nearest", "strict") if fmt[idx] == "M" else None
if "interpolation" not in transform_settings[idx]:
transform_settings[idx][
"interpolation"] = validate_parameter(
val, ALLOWED_INTERPOLATIONS, "bilinear", str, True)
else:
transform_settings[idx][
"interpolation"] = validate_parameter(
(transform_settings[idx]["interpolation"],
"strict"),
ALLOWED_INTERPOLATIONS,
"bilinear",
str,
True,
)
if "padding" not in transform_settings[idx]:
transform_settings[idx]["padding"] = validate_parameter(
None, ALLOWED_PADDINGS, "z", str, True)
else:
transform_settings[idx]["padding"] = validate_parameter(
(transform_settings[idx]["padding"], "strict"),
ALLOWED_PADDINGS,
"z",
str,
True,
)
else:
if "interpolation" in transform_settings[
idx] or "padding" in transform_settings[idx]:
raise TypeError
if len(data) != len(transform_settings):
raise ValueError
for t in fmt:
if t not in ALLOWED_TYPES:
raise TypeError(
f"The found type was {t}, but needs to be one of {ALLOWED_TYPES}"
)
self.__data = data
self.__fmt = fmt
self.__transform_settings = transform_settings
def test_parameter_validation_raises_error_when_default_value_is_wrong_type():
with pytest.raises(TypeError):
validate_parameter(None, {1, 2}, ('10', 'inherit'), int)
def __init__(self, data, fmt, transform_settings=None):
if len(fmt) == 1 and not isinstance(data, tuple):
if not isinstance(data, list):
data = (data, )
else:
raise TypeError
if not isinstance(data, tuple):
raise TypeError
if len(data) != len(fmt):
raise ValueError
if transform_settings is not None:
if not isinstance(transform_settings, dict):
raise TypeError
else:
transform_settings = {}
# Element-wise settings
# If no settings provided for certain items, they will be created
for idx in range(len(data)):
if idx not in transform_settings:
transform_settings[idx] = {}
if fmt[idx] == "I" or fmt[idx] == "M":
val = ("nearest", "strict") if fmt[idx] == "M" else None
if "interpolation" not in transform_settings[idx]:
transform_settings[idx][
"interpolation"] = validate_parameter(
val, ALLOWED_INTERPOLATIONS, "bilinear", str, True)
else:
transform_settings[idx][
"interpolation"] = validate_parameter(
(transform_settings[idx]["interpolation"],
"strict"),
ALLOWED_INTERPOLATIONS,
"bilinear",
str,
True,
)
if "padding" not in transform_settings[idx]:
transform_settings[idx]["padding"] = validate_parameter(
None, ALLOWED_PADDINGS, "z", str, True)
else:
transform_settings[idx]["padding"] = validate_parameter(
(transform_settings[idx]["padding"], "strict"),
ALLOWED_PADDINGS,
"z",
str,
True,
)
else:
if "interpolation" in transform_settings[
idx] or "padding" in transform_settings[idx]:
raise TypeError
if len(data) != len(transform_settings):
raise ValueError
for t in fmt:
if t not in ALLOWED_TYPES:
raise TypeError
self.__data = data
self.__fmt = fmt
self.__transform_settings = transform_settings
self.__imagenet_mean = torch.tensor(
(0.485, 0.456, 0.406)).view(3, 1, 1)
self.__imagenet_std = torch.tensor((0.229, 0.224, 0.225)).view(3, 1, 1)