def test_grouped_mr_heart(old_config_path: str, latest_config_path: str):
with open(old_config_path) as file:
old_config = yaml.load(file, Loader=yaml.FullLoader)
with open(latest_config_path) as file:
latest_config = yaml.load(file, Loader=yaml.FullLoader)
updated_config = parse_v011(old_config=old_config)
assert updated_config == latest_config
def config_sanity_check(config: dict) -> dict:
"""
Check if the given config satisfies the requirements.
:param config: entire config.
"""
# check data
data_config = config["dataset"]
if data_config["type"] not in ["paired", "unpaired", "grouped"]:
raise ValueError(
f"data type must be paired / unpaired / grouped, got {type}.")
if data_config["format"] not in ["nifti", "h5"]:
raise ValueError(f"data format must be nifti / h5, got {format}.")
assert "dir" in data_config
for mode in ["train", "valid", "test"]:
assert mode in data_config["dir"].keys()
data_dir = data_config["dir"][mode]
if data_dir is None:
logging.warning(f"Data directory for {mode} is not defined.")
if not (isinstance(data_dir, (str, list)) or data_dir is None):
raise ValueError(
f"data_dir for mode {mode} must be string or list of strings,"
f"got {data_dir}.")
# back compatibility support
config = parse_v011(config)
# check model
if config["train"]["method"] == "conditional":
if data_config["labeled"] is False: # unlabeled
raise ValueError(
"For conditional model, data have to be labeled, got unlabeled data."
)
# loss weights should >= 0
for name in ["image", "label", "regularization"]:
loss_config = config["train"]["loss"][name]
if not isinstance(loss_config, list):
loss_config = [loss_config]
for loss_i in loss_config:
loss_weight = loss_i["weight"]
if loss_weight <= 0:
logging.warning("The %s loss weight %.2f is not positive.",
name, loss_weight)
return config
def config_sanity_check(config: dict) -> dict:
"""
Check if the given config satisfies the requirements.
:param config: entire config.
"""
# back compatibility support
config = parse_v011(config)
# check model
if config["train"]["method"] == "conditional":
if config["dataset"]["train"]["labeled"] is False: # unlabeled
raise ValueError(
"For conditional model, data have to be labeled, got unlabeled data."
)
return config