def get_default_configuration(network,
task,
network_trainer,
plans_identifier=default_plans_identifier,
search_in=(nnunet.__path__[0], "training",
"network_training"),
base_module='nnunet.training.network_training'):
assert network in ['2d', '3d_lowres', '3d_fullres', '3d_cascade_fullres'], \
"network can only be one of the following: \'3d\', \'3d_lowres\', \'3d_fullres\', \'3d_cascade_fullres\'"
dataset_directory = join(preprocessing_output_dir, task)
if network == '2d':
plans_file = join(preprocessing_output_dir, task,
plans_identifier + "_plans_2D.pkl")
else:
plans_file = join(preprocessing_output_dir, task,
plans_identifier + "_plans_3D.pkl")
plans = load_pickle(plans_file)
possible_stages = list(plans['plans_per_stage'].keys())
if (network == '3d_cascade_fullres'
or network == "3d_lowres") and len(possible_stages) == 1:
raise RuntimeError(
"3d_lowres/3d_cascade_fullres only applies if there is more than one stage. This task does "
"not require the cascade. Run 3d_fullres instead")
if network == '2d' or network == "3d_lowres":
stage = 0
else:
stage = possible_stages[-1]
trainer_class = recursive_find_trainer([join(*search_in)],
network_trainer,
current_module=base_module)
output_folder_name = join(network_training_output_dir, network, task,
network_trainer + "__" + plans_identifier)
print("###############################################")
print("I am running the following nnUNet: %s" % network)
print("My trainer class is: ", trainer_class)
print("For that I will be using the following configuration:")
summarize_plans(plans_file)
print("I am using stage %d from these plans" % stage)
if (network == '2d'
or len(possible_stages) > 1) and not network == '3d_lowres':
batch_dice = True
print("I am using batch dice + CE loss")
else:
batch_dice = False
print("I am using sample dice + CE loss")
print("\nI am using data from this folder: ",
join(dataset_directory, plans['data_identifier']))
print("###############################################")
return plans_file, output_folder_name, dataset_directory, batch_dice, stage, trainer_class
parser = argparse.ArgumentParser()
parser.add_argument("network_trainer")
parser.add_argument("task")
parser.add_argument("fold", type=int)
args = parser.parse_args()
trainerclass = args.network_trainer
task = args.task
fold = args.fold
plans_file, folder_with_preprocessed_data, output_folder_name, dataset_directory, batch_dice, stage = \
get_default_configuration("3d_lowres", task)
trainer_class = recursive_find_trainer(
[os.path.join(nnunet.__path__[0], "training", "network_training")],
trainerclass, "nnunet.training.network_training")
if trainer_class is None:
raise RuntimeError(
"Could not find trainer class in nnunet.training.network_training")
else:
assert issubclass(
trainer_class, nnUNetTrainer
), "network_trainer was found but is not derived from nnUNetTrainer"
trainer = trainer_class(plans_file,
fold,
folder_with_preprocessed_data,
output_folder=output_folder_name,
dataset_directory=dataset_directory,
def get_default_configuration(network,
task,
network_trainer,
plans_identifier=default_plans_identifier,
search_in=(nnunet.__path__[0], "training",
"network_training"),
base_module='nnunet.training.network_training'):
assert network in ['2d', '3d_lowres', '3d_fullres', '3d_cascade_fullres'], \
"network can only be one of the following: \'3d\', \'3d_lowres\', \'3d_fullres\', \'3d_cascade_fullres\'"
dataset_directory = join(preprocessing_output_dir, task)
if network == '2d':
plans_file = join(preprocessing_output_dir, task,
plans_identifier + "_plans_2D.pkl")
else:
plans_file = join(preprocessing_output_dir, task + '/',
plans_identifier + "_plans_3D.pkl")
plans = load_pickle(plans_file)
"""
plans_file content:
dict_keys(['preprocessed_data_folder', 'min_size_per_class', 'num_classes', 'keep_only_largest_region', 'original_sizes', 'original_spacings', 'all_classes', 'modalities', 'base_num_features', 'min_region_size_per_class', 'data_identifier', 'use_mask_for_norm', 'num_modalities', 'normalization_schemes', 'plans_per_stage', 'num_stages', 'dataset_properties', 'list_of_npz_files'])
"""
possible_stages = list(plans['plans_per_stage'].keys())
"""
file_data['plans_per_stage'][0]
{'pool_op_kernel_sizes': [[2, 2, 1], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2]], 'conv_kernel_sizes': [[3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'num_pool_per_axis': [5, 5, 4], 'batch_size': 2, 'original_spacing': array([0.78162497, 0.78162497, 3. ]), 'current_spacing': array([1.5530063, 1.5530063, 3.090903 ]), 'do_dummy_2D_data_aug': False, 'patch_size': array([160, 160, 80]), 'median_patient_size_in_voxels': array([263, 263, 113])}
file_data['plans_per_stage'][1]
{'pool_op_kernel_sizes': [[2, 2, 1], [2, 2, 1], [2, 2, 2], [2, 2, 2], [2, 2, 2]], 'conv_kernel_sizes': [[3, 3, 1], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'num_pool_per_axis': [5, 5, 3], 'batch_size': 2, 'original_spacing': array([0.78162497, 0.78162497, 3. ]), 'current_spacing': array([0.78162497, 0.78162497, 3. ]), 'do_dummy_2D_data_aug': False, 'patch_size': array([192, 192, 48]), 'median_patient_size_in_voxels': array([523, 523, 116])}
"""
if (network == '3d_cascade_fullres'
or network == "3d_lowres") and len(possible_stages) == 1:
raise RuntimeError(
"3d_lowres/3d_cascade_fullres only applies if there is more than one stage. This task does "
"not require the cascade. Run 3d_fullres instead")
if network == '2d' or network == "3d_lowres":
stage = 0
else:
stage = possible_stages[-1]
trainer_class = recursive_find_trainer([join(*search_in)],
network_trainer,
current_module=base_module)
output_folder_name = join(network_training_output_dir + '/', network + '/',
task + '/',
network_trainer + "__" + plans_identifier + '/')
print("###############################################")
print("I am running the following nnUNet: %s" % network)
print("My trainer class is: ", trainer_class)
print("For that I will be using the following configuration:")
summarize_plans(plans_file)
print("I am using stage %d from these plans" % stage)
if (network == '2d'
or len(possible_stages) > 1) and not network == '3d_lowres':
batch_dice = True
print("I am using batch dice + CE loss")
else:
batch_dice = False
print("I am using sample dice + CE loss")
print("\nI am using data from this folder: ",
join(dataset_directory, plans['data_identifier']))
print("###############################################")
return plans_file, output_folder_name, dataset_directory, batch_dice, stage, trainer_class