Io loader

The IO Loader functions of AUCMEDI allow loading samples from datasets in different file formats.

These functions are called internally via the DataGenerator.

IO_loader Functions

Interface	Description
image_loader()	Image Loader for image loading via Pillow.
sitk_loader()	SimpleITK Loader for loading NIfTI (nii) or Metafile (mha) formats.
numpy_loader()	NumPy Loader for image loading of .npy files.
cache_loader()	Cache Loader for passing already loaded images.

Parameters defined in **kwargs are passed down to IO_loader functions.

Example

# Import required libraries
from aucmedi import *

# Initialize input data reader
ds = input_interface(interface="csv",
                     path_imagedir="dataset/images/",
                     path_data="dataset/annotations.csv",
                     ohe=False, col_sample="ID", col_class="diagnosis")
(samples, class_ohe, nclasses, class_names, image_format) = ds

# Initialize DataGenerator with by default using image_loader
data_gen = DataGenerator(samples, "dataset/images/", labels=class_ohe,
                         image_format=image_format, resize=None)

# Initialize DataGenerator with manually selected image_loader
from aucmedi.data_processing.io_loader import image_loader
data_gen = DataGenerator(samples, "dataset/images/", labels=class_ohe,
                         image_format=image_format, resize=None,
                         loader=image_loader)