dltk.core package

• dltk.core.io package
  • dltk.core.io.augmentation module
  • dltk.core.io.preprocessing module
  • dltk.core.io.reader module
  • dltk.core.io.sliding_window module
  • Module contents
• dltk.core.modules package
  • dltk.core.modules.activations module
  • dltk.core.modules.base module
  • dltk.core.modules.batch_normalization module
  • dltk.core.modules.bilinear_upsample module
  • dltk.core.modules.convolution module
  • dltk.core.modules.graph_convolution module
  • dltk.core.modules.linear module
  • dltk.core.modules.losses module
  • dltk.core.modules.regularization module
  • dltk.core.modules.residual_units module
  • dltk.core.modules.summaries module
  • dltk.core.modules.tranposed_convolution module
  • Module contents

dltk.core.metrics module

dltk.core.metrics.abs_vol_difference(pred, labels, num_classes)

Calculates the average volume difference of labels and predictions per class

Parameters:

• pred (np.ndarray) – predictions
• labels (np.ndarray) – labels
• num_classes (int) – number of classes to calculate avd for

Returns:

avd per class

Return type:

np.ndarray

dltk.core.metrics.crossentropy(pred, labels, logits=True)

Calculates the crossentropy loss between prediction and labels

Parameters:

• pred (np.ndarray) – prediction of the system
• labels (np.ndarray) – labels
• logits (bool) – flag whether pred are logits or probabilities

Returns:

crossentropy error

Return type:

float

dltk.core.metrics.dice(pred, labels, num_classes)

Calculates the dice score of labels and predictions

Parameters:

• pred (np.ndarray) – predictions
• labels (np.ndarray) – labels
• num_classes (int) – number of classes to calculate avd for

Returns:

dice per class

Return type:

np.ndarray

Module contents