) # expected: [img_size_z*num_images, img_size_x, vol_size_y, img_size_t, n_channels]
logging.info(
'============================================================\n')
# =================================================================================
# ==== If slicing preprocessing is enabled, do preprocessing of masked the data now
# =================================================================================
if preprocess_enabled == "slice":
logging.info(
'============================================================')
logging.info('Loading training data from: ' + project_data_root)
data_tr = data_freiburg_numpy_to_preprocessed_hdf5.load_cropped_data_sliced(
basepath=project_data_root,
idx_start=0,
idx_end=10,
train_test='train')
images_tr = data_tr['sliced_images_train']
logging.info(type(images_tr))
logging.info(
'Shape of training images: %s' % str(images_tr.shape)
) # expected: [img_size_z*num_images, img_size_x, vol_size_y, img_size_t, n_channels]
logging.info(
'============================================================')
logging.info('Loading validation data from: ' + project_data_root)
data_vl = data_freiburg_numpy_to_preprocessed_hdf5.load_cropped_data_sliced(
basepath=project_data_root,
idx_start=10,
idx_end=13,
vae_network = VariationalAutoencoder
model = VAEModel(vae_network, config, model_name, log_dir)
# Load the vae model as our baseline
path = os.path.join(project_code_root, config["model_directory"])
model.load_from_path(path, config["model_name"] , config["latest_model_epoch"])
# ==================================================================================================================================================================
# ============== LOAD THE DATA =====================================================================================================================================
# ==================================================================================================================================================================
if preprocess_enabled == "slice":
logging.info('============================================================')
logging.info('Loading training data from: ' + project_data_root)
data_tr = data_freiburg_numpy_to_preprocessed_hdf5.load_cropped_data_sliced(basepath = project_data_root,
idx_start = config['train_data_start_idx'],
idx_end = config['train_data_end_idx'],
train_test='train')
images_tr_sl = data_tr['sliced_images_train']
logging.info(type(images_tr_sl))
logging.info('Shape of training images: %s' %str(images_tr_sl.shape)) # expected: [img_size_z*num_images, img_size_x, vol_size_y, img_size_t, n_channels]
logging.info('============================================================')
logging.info('Loading validation data from: ' + project_data_root)
data_vl = data_freiburg_numpy_to_preprocessed_hdf5.load_cropped_data_sliced(basepath = project_data_root,
idx_start = config['validation_data_start_idx'],
idx_end = config['validation_data_end_idx'],
train_test='validation')
images_vl_sl = data_vl['sliced_images_validation']
logging.info('Shape of validation images: %s' %str(images_vl_sl.shape)) # expected: [img_size_z*num_images, img_size_x, vol_size_y, img_size_t, n_channels]
logging.info('============================================================\n')
