unet

UNET implementations.

Functions

FUNet

Functional UNET Implementation

Classes

SUNet	Semantic UNet
UNet	UNet Architecture

ashpy.models.convolutional.unet.FUNet(input_res, min_res, kernel_size, initial_filters, filters_cap, channels, input_channels=3, use_dropout_encoder=True, use_dropout_decoder=True, dropout_prob=0.3, encoder_non_linearity=<class 'tensorflow.python.keras.layers.advanced_activations.LeakyReLU'>, decoder_non_linearity=<class 'tensorflow.python.keras.layers.advanced_activations.ReLU'>, last_activation=<function tanh>, use_attention=False)

Functional UNET Implementation

class ashpy.models.convolutional.unet.SUNet(input_res, min_res, kernel_size, initial_filters, filters_cap, channels, use_dropout_encoder=True, use_dropout_decoder=True, dropout_prob=0.3, encoder_non_linearity=<class 'tensorflow.python.keras.layers.advanced_activations.LeakyReLU'>, decoder_non_linearity=<class 'tensorflow.python.keras.layers.advanced_activations.ReLU'>, use_attention=False)

Bases: ashpy.models.convolutional.unet.UNet

Semantic UNet

class ashpy.models.convolutional.unet.UNet(input_res, min_res, kernel_size, initial_filters, filters_cap, channels, use_dropout_encoder=True, use_dropout_decoder=True, dropout_prob=0.3, encoder_non_linearity=<class 'tensorflow.python.keras.layers.advanced_activations.LeakyReLU'>, decoder_non_linearity=<class 'tensorflow.python.keras.layers.advanced_activations.ReLU'>, normalization_layer=<class 'ashpy.layers.instance_normalization.InstanceNormalization'>, last_activation=<function tanh>, use_attention=False)

Bases: ashpy.models.convolutional.interfaces.Conv2DInterface

UNet Architecture

Used in Image-to-Image Translation with Conditional Adversarial Nets 1

Examples

• Direct Usage:

  x = tf.ones((1, 512, 512, 3))
  u_net = UNet(input_res = 512,
               min_res=4,
               kernel_size=4,
               initial_filters=64,
               filters_cap=512,
               channels=3)
  y = u_net(x)
  print(y.shape)
  print(len(u_net.trainable_variables)>0)
  

  (1, 512, 512, 3)
  True
  

1

Image-to-Image Translation with Conditional Adversarial Nets https://arxiv.org/abs/1611.04076

__init__(input_res, min_res, kernel_size, initial_filters, filters_cap, channels, use_dropout_encoder=True, use_dropout_decoder=True, dropout_prob=0.3, encoder_non_linearity=<class 'tensorflow.python.keras.layers.advanced_activations.LeakyReLU'>, decoder_non_linearity=<class 'tensorflow.python.keras.layers.advanced_activations.ReLU'>, normalization_layer=<class 'ashpy.layers.instance_normalization.InstanceNormalization'>, last_activation=<function tanh>, use_attention=False)

Parameters

• input_res (int) – input resolution

• min_res (int) – minimum resolution reached after decode

• kernel_size (int) – kernel size used in the network

• initial_filters (int) – number of filter of the initial convolution

• filters_cap (int) – maximum number of filters

• channels (int) – number of output channels

• use_dropout_encoder (bool) – whether to use dropout in the encoder module

• use_dropout_decoder (bool) – whether to use dropout in the decoder module

• dropout_prob (float) – probability of dropout

• encoder_non_linearity (Type[Layer]) – non linearity of encoder

• decoder_non_linearity (Type[Layer]) – non linearity of decoder

• last_activation (<module 'tensorflow.python.keras.api._v2.keras.activations' from '/home/docs/checkouts/readthedocs.org/user_builds/ashpy/envs/v0.1.3/lib/python3.7/site-packages/tensorflow/python/keras/api/_v2/keras/activations/__init__.py'>) – last activation function, tanh or softmax (for semantic images)

• use_attention (bool) – whether to use attention

call(inputs, training=False)

Execute the model on input data.

Parameters

• inputs (tf.Tensor) – Input tensor(s).

• training (bool) – Training flag.

• return_features (bool) – If True returns the features.

Returns

tf.Tensor – The model output.

get_decoder_block(filters, use_bn=True, use_dropout=False, use_attention=False)

Returns a block to be used in the decoder part of the UNET :param filters: number of filters :param use_bn: whether to use batch normalization :param use_dropout: whether to use dropout :param use_attention: whether to use attention

Returns

A block to be used in the decoder part

get_encoder_block(filters, use_bn=True, use_attention=False)

Returns a block to be used in the encoder part of the UNET :param filters: number of filters :param use_bn: whether to use batch normalization :param use_attention: whether to use attention

Returns

A block to be used in the encoder part