def nii(input_folder, output_folder):
"""
Take a folder, recursively traverse through all its subfolders, convert each subfolder into a NII file
:param input_folder:
:param output_folder:
:return:
"""
os.chdir(output_folder)
create("nii")
logger.info("Generating NII files:")
path_nii = os.path.join(output_folder, "nii")
DICOM_convert.to_nii(input_folder, path_nii)
DICOM_convert.fix_series(path_nii)
logger.info("Nii files generated!")
def raw_sorted(input_folder, output_folder):
"""
FLATCOPY (with timestampe to avoid naming duplications) to a an output folder with raw prefix, then SORT THEM with proper protocole series acquisition number.
:param input_folder:
:param output_folder:
:return:
"""
os.chdir(output_folder)
os.chdir(output_folder)
create("raw_sorted")
logger.info("Generating sorted raw files:")
path_raw_sorted = os.path.join(output_folder, "raw_sorted")
DICOM_sort.into_folder(input_folder, path_raw_sorted)
DICOM_convert.fix_series(path_raw_sorted)
logger.info("Sorted raw files completed!")
def DICOM_universal_convert_default(
input_root_folder,
name_default_folder=f"DICOM_UniversalConvert-{unique_name()}"):
"""
Create a new folder with
:param input_root_folder:
:return:
"""
logger.info("Begin converting folder: " + input_root_folder)
# Generate the potential name for the new path.
path_result = f"{input_root_folder}-{name_default_folder}"
# intellgently create the folder if needed be.
create(path_result)
# Trigger the subject specific convertion.
DICOM_converter.DICOM_universal_convert(input_root_folder, path_result)
logger.info(f"Finished converting: {input_root_folder}")
def DICOMOBJ_converter(
input_root_folder,
default_folder=f"DICOM_UniversalConvert-{unique_name()}"):
"""
Create a new folder with
:param input_root_folder:
:return:
"""
path_list = DICOM_converter.DICOMOBJ_finder(input_root_folder)
for path in path_list:
logger.info("Begin converting folder: " + path)
# Get higher folder path.
path_source = os.path.dirname(path)
# Generate the potential name for the new path.
path_result = os.path.join(path_source, default_folder)
# intellgently create the folder if needed be.
create(path_result)
# Trigger the subject specific convertion.
DICOM_converter.DICOM_universal_convert(path, path_result)
logger.info(f"Finished converting: {path}")
from imageio import imsave
from pathlib import Path
path_current_file = Path(os.path.realpath(__file__))
path_module = path_current_file.parents[
2] # this file is located 3 folders deep: /, /model, /model/Kaggle_DCGAN_Dogs
print(f"{path_module}")
sys.path.append(f"{path_module}")
# Some convenience function I already prebuilt: github.com/dyt811/PythonUtils/
from PythonUtils.PUFile import unique_name
from PythonUtils.PUFolder import recursive_list, create
path_log_run = path_module / Path("logs") / unique_name()
create(path_log_run
) # create the training specific folder if it doesn't exist already.
# dimensions_noise = 100
"""
Largely inspired from source: https://github.com/DataSnaek/DCGAN-Keras
"""
class DCGAN:
def __init__(self, path_discriminator, path_generator, path_output,
img_size):
# Key image properties.
self.img_size = img_size # default x and y
self.channels = 3
def __init__(
self,
input_shape,
output_classes,
train_data_path: Path,
optimizer="adam",
loss="mse",
metrics=None,
checkpoint_metric="val_loss",
checkpoint_metric_mode="min",
):
# Use these settings per constructor input.
self.input_shape = input_shape
self.output_classes = output_classes
self.loss = loss
self.optimizer = optimizer
# Default metrics
if metrics is None:
self.metrics = ([
"mae",
"mse",
"mape",
"cosine",
loss_SSIM,
loss_mae_diff_SSIM_composite,
loss_mse_diff_SSIM_composite,
f1_metric,
], )
else:
self.metrics = metrics
self.checkpoint_metric = checkpoint_metric
self.checkpoint_metric_mode = checkpoint_metric_mode
self.train_data = None
self.path_train_data: Path = train_data_path
self.test_data = None # fixme: independent data used for testing (optional?)
self.callbacks_list = None
self.model = None
self.path_prediction = None
# Default step and epoch size. Easily overwritten during the run stage.
self.size_step = 256
self.size_epoch = 500
# Dynamically generate model input_path.
this_file = os.path.realpath(__file__)
project_root = get_abspath(
this_file, 2
) # todo: this folder path is hard coded. Needs to be generalized.
# Log path.
self.path_log, self.path_model = get_paths(project_root)
# Log run path.
self.path_log_run = os.path.join(self.path_log,
unique_name() + __name__)
# Create the Log run path.
create(self.path_log_run)
self.model_stage = stage.Initialized