util

Implementation of useful function.

segmentation_rt.util.util.format_log(epoch, iteration, errors, t)

Generic format for print/log.

Parameters
  • epoch (int) – epoch.

  • iteration (int) – iteration.

  • errors (dict[float]) – errors dictionary.

  • t (float) – time.

Returns

message.

Return type

str

segmentation_rt.util.util.get_subjects(path, structures, transform)

Browse the path folder to build a dataset. Folder must contains the subjects with the CT and masks.

Parameters
  • path (str) – root folder.

  • structures (list[str]) – list of structures.

  • transform (tio.transforms.Transform) – transforms to be applied.

Returns

Base TorchIO dataset.

Return type

tio.SubjectsDataset

segmentation_rt.util.util.listdir_full_path(path)

List files in path and return their absolute path.

Parameters

path (str) – path.

Returns

list of absolute path.

Return type

list[str]

segmentation_rt.util.util.print_log(out_f, message)

Writes in log file.

Parameters
  • out_f – I/O stream.

  • message (str) – message to display.

segmentation_rt.util.util.save_image(image, path, width=10, bitdepth=8, start=None, end=None, added_slices_step=5)

Save image in png format.

Parameters
  • image (np.ndarray) – 3D npy array.

  • bitdepth (int) – encoding.

  • path (str) – output folder.

  • width (int) – number of kept slices before the first and after the last non empty slices.

  • added_slices_step (int) – the script saves slices from the first slice to the first non empty slices with this step (same at the end). if this added_slices_step ==0 , no slices are saved.

  • start (int) – first slice (default 0).

  • end (int) – last slice (default -1).