if we run with -c then we need to set the correct lr for the first epoch, otherwise it will run the first continued epoch with self.i

if we run with -c then we need to set the correct lr for the first epoch, otherwise it will run the first continued epoch with self.i

if we run with -c then we need to set the correct lr for the first epoch, otherwise it will run the first continued epoch with self.i

faster and uses less ram than save_segmentation_nifti_from_softmax, but maybe less precise and also does not support softmax export (which is

Similarly to BraTS 2017 - BraTS 2019, each participant will be ranked for each of the X test cases. Each case includes 3 regions of evaluatio

Set evaluation metrics

Set the reference segmentation.

Set the test segmentation.

Compute binary focal loss between target and output logits. Source https://github.com/BloodAxe/pytorch-toolbelt See :class:`~pytorch_tool

We now take a large number of random combinations and perform sanity checks :return:

Return result as pandas DataFrame.

TN + FP

TN + FN

TP + FN

TP + FP

TN / (TN + FP)

We need to wrap this because we need to enforce self.network.do_ds = False for prediction

We need to wrap this because we need to enforce self.network.do_ds = False for prediction

We need to wrap this because we need to enforce self.network.do_ds = False for prediction :param do_mirroring: :param use_tr

We need to wrap this because we need to enforce self.network.do_ds = False for prediction

This is used for test set preprocessing :return: