def run_preprocessing(self, num_threads):
if os.path.isdir(
join(self.preprocessed_output_folder, "gt_segmentations")):
shutil.rmtree(
join(self.preprocessed_output_folder, "gt_segmentations"))
shutil.copytree(
join(self.folder_with_cropped_data, "gt_segmentations"),
join(self.preprocessed_output_folder, "gt_segmentations"))
normalization_schemes = self.plans['normalization_schemes']
use_nonzero_mask_for_normalization = self.plans['use_mask_for_norm']
intensityproperties = self.plans['dataset_properties'][
'intensityproperties']
preprocessor = GenericPreprocessor(normalization_schemes,
use_nonzero_mask_for_normalization,
self.transpose_forward,
intensityproperties)
target_spacings = [
i["current_spacing"] for i in self.plans_per_stage.values()
]
if self.plans['num_stages'] > 1 and not isinstance(
num_threads, (list, tuple)):
num_threads = (8, num_threads)
elif self.plans['num_stages'] == 1 and isinstance(
num_threads, (list, tuple)):
num_threads = num_threads[-1]
preprocessor.run(target_spacings, self.folder_with_cropped_data,
self.preprocessed_output_folder,
self.plans['data_identifier'], num_threads)
def generate_data(self, data_path):
# form list of file names
# expected_num_modalities = self.trainer.plans['num_modalities']
# # check input folder integrity
# case_ids = check_input_folder_and_return_caseIDs(data_path, expected_num_modalities)
# all_files = subfiles(data_path, suffix=".nii.gz", join=False, sort=True)
# list_of_lists = [[join(data_path, i) for i in all_files if i[:len(j)].startswith(j) and
# len(i) == (len(j) + 12)] for j in case_ids]
# part_id = 0
# num_parts = 1
file_name = [
'/home/zhiyuan/project/data/nnUNet_raw/nnUNet_raw_data/Task004_Hippocampus/imagesTs/hippocampus_392_0000.nii.gz'
]
data, seg, properties = ImageCropper.crop_from_list_of_files(file_name)
plans = self.trainer.plans
if plans.get('transpose_forward') is None or plans.get(
'transpose_backward') is None:
print(
"WARNING! You seem to have data that was preprocessed with a previous version of nnU-Net. "
"You should rerun preprocessing. We will proceed and assume that both transpose_foward "
"and transpose_backward are [0, 1, 2]. If that is not correct then weird things will happen!"
)
plans['transpose_forward'] = [0, 1, 2]
plans['transpose_backward'] = [0, 1, 2]
# 1. From preprocessing.GenericPreprocessor.preprocess_test_case
preprocessor = GenericPreprocessor(
plans['normalization_schemes'], plans['use_mask_for_norm'],
plans['transpose_forward'],
plans['dataset_properties']['intensityproperties'])
data = data.transpose(
(0, *[i + 1 for i in preprocessor.transpose_forward]))
seg = seg.transpose(
(0, *[i + 1 for i in preprocessor.transpose_forward]))
data, seg, properties = preprocessor.resample_and_normalize(
data,
plans['plans_per_stage'][self.trainer.stage]['current_spacing'],
properties,
seg,
force_separate_z=None)
output_data = data.astype(np.float32)
return output_data, seg, properties
def preprocess_patient(self, input_files):
"""
Used to predict new unseen data. Not used for the preprocessing of the training/test data
:param input_files:
:return:
"""
from nnunet.preprocessing.preprocessing import GenericPreprocessor, PreprocessorFor2D
if self.threeD:
preprocessor = GenericPreprocessor(self.normalization_schemes, self.use_mask_for_norm,
self.intensity_properties)
else:
preprocessor = PreprocessorFor2D(self.normalization_schemes, self.use_mask_for_norm,
self.intensity_properties)
d, s, properties = preprocessor.preprocess_test_case(input_files,
self.plans['plans_per_stage'][self.stage]['current_spacing'])
return d, s, properties