delta.model

Click on the signatures below to get a complete description.

Module for model definitions and loss/metrics functions definitions

@author: jblugagne

Functions

class_weighted_categorical_crossentropy(...)	Generate class-weighted categorical cross-entropy loss function.
contracting_block(input_layer, filters, ...)	A block of layers for 1 contracting level of the U-Net
expanding_block(input_layer, skip_layer, ...)	A block of layers for 1 expanding level of the U-Net
pixelwise_weighted_binary_crossentropy_seg(...)	Pixel-wise weighted binary cross-entropy loss.
pixelwise_weighted_binary_crossentropy_track(...)	Pixel-wise weighted binary cross-entropy loss.
unet([input_size, final_activation, ...])	Generic U-Net declaration.
unet_rois([input_size, levels])	ROIs segmentation U-Net.
unet_seg([pretrained_weights, input_size, ...])	Cell segmentation U-Net definition function.
unet_track([pretrained_weights, input_size, ...])	Tracking U-Net definition function.
unstack_acc(y_true, y_pred)	Unstacks the mask from the weights in the output tensor for segmentation and computes binary accuracy