delta.model.expanding_block

delta.model.expanding_block(input_layer: Tensor, skip_layer: Tensor, filters: int, conv2d_parameters: Dict, dropout: float = 0, name: str = 'Expanding') → Tensor

A block of layers for 1 expanding level of the U-Net

Parameters

input_layertf.Tensor

The convolutional layer that is the output of the lower level’s expanding block

skip_layertf.Tensor

The convolutional layer that is the output of this level’s contracting block

filtersint

filters input for the Conv2D layers of the block.

conv2d_parametersdict()

kwargs for the Conv2D layers of the block.

dropoutfloat, optional

Dropout layer rate in the block. Valid range is [0,1). If 0, no dropout layer is added. The default is 0

namestr, optional

Name prefix for the layers in this block. The default is “Expanding”.

Returns

conv3tf.Tensor

Output of this level’s expanding block.