data and seg must already have been transposed by transpose_forward. properties are the un-transposed values (spacing etc) :p

TP / (TP + FN)

Set the labels. :param labels= may be a dictionary (int->str), a set (of ints), a tuple (of ints) or a list (of ints). Labels will onl

TN / (TN + FP)

used to swap between folds for inference (ensemble of models from cross-validation) DO NOT USE DURING TRAINING AS THIS WILL NOT UPDAT

Build the Pyramid Pooling Module.

This is the basic data loader for 3D networks. It uses preprocessed data as produced by my (Fabian) preprocessing. You can load the d

Compute focal loss for binary classification problem.

This one does fully convolutional inference. No sliding window

Implementation of FPN head for segmentation models according to: http://presentations.cocodataset.org/COCO17-Stuff-FAIR.pdf Args:

inspired by https://stackoverflow.com/questions/1855095/how-to-create-a-zip-archive-of-a-directory-in-python

:param directory: must have the 5 fold_X subfolders as well as a postprocessing.json and plans.pkl :param output_directory: :return:

This script will run recursively through all subfolders of current_folder and copy all files that end with suffix with some automatically gen

:param data: :param seg: :param nonzero_label: this will be written into the segmentation map :return:

Display Configuration values.

attempting to reimplement the brats evaluation scheme assumes edema=1, non_enh=2, enh=3 :param arr: :param arr_gt: :return:

FN / (TN + FN)

Can be implemented, does not have to :return: