autoencoders

Collection of Fully Convolutional Autoencoders.

Classes

Autoencoder

Primitive Model for all convolutional autoencoders.

FCNNAutoencoder

Primitive Model for all fully convolutional autoencoders.

class ashpy.models.convolutional.autoencoders.Autoencoder(layer_spec_input_res, layer_spec_target_res, kernel_size, initial_filters, filters_cap, encoding_dimension, channels)[source]

Bases: tensorflow.python.keras.engine.training.Model

Primitive Model for all convolutional autoencoders.

Examples

  • Direct Usage:

    autoencoder = Autoencoder(
        layer_spec_input_res=(64, 64),
        layer_spec_target_res=(8, 8),
        kernel_size=5,
        initial_filters=32,
        filters_cap=128,
        encoding_dimension=100,
        channels=3,
    )
    
    encoding, reconstruction = autoencoder(tf.zeros((1, 64, 64, 3)))
    print(encoding.shape)
    print(reconstruction.shape)
    
    
    
__init__(layer_spec_input_res, layer_spec_target_res, kernel_size, initial_filters, filters_cap, encoding_dimension, channels)[source]

Instantiate the BaseAutoEncoder.

Parameters
  • layer_spec_input_res (tuple of (int, int)) – Shape of the input tensors.

  • layer_spec_target_res – (tuple of (int, int)): Shape of tensor desired as output of _get_layer_spec().

  • kernel_size (int) – Kernel used by the convolution layers.

  • initial_filters (int) – Numbers of filters to used as a base value.

  • filters_cap (int) – Cap filters to a set amount, in the case of an Encoder is a ceil value AKA the max amount of filters.

  • encoding_dimension (int) – encoding dimension.

  • channels (int) – Number of channels for the reconstructed image.

Returns

None

call(inputs, training=True)[source]

Execute the model on input data.

Parameters
  • inputs (tf.Tensor) – Input tensors.

  • training (bool) – Training flag.

Returns

(encoding, reconstruction) – Pair of tensors.

class ashpy.models.convolutional.autoencoders.FCNNAutoencoder(layer_spec_input_res, layer_spec_target_res, kernel_size, initial_filters, filters_cap, encoding_dimension, channels)[source]

Bases: tensorflow.python.keras.engine.training.Model

Primitive Model for all fully convolutional autoencoders.

Examples

  • Direct Usage:
    autoencoder = FCNNAutoencoder(
        layer_spec_input_res=(64, 64),
        layer_spec_target_res=(8, 8),
        kernel_size=5,
        initial_filters=32,
        filters_cap=128,
        encoding_dimension=100,
        channels=3,
    )
    
    encoding, reconstruction = autoencoder(tf.zeros((1, 64, 64, 3)))
    print(encoding.shape)
    print(reconstruction.shape)
    
    
    
__init__(layer_spec_input_res, layer_spec_target_res, kernel_size, initial_filters, filters_cap, encoding_dimension, channels)[source]

Instantiate the FCNNBaseAutoEncoder.

Parameters
  • layer_spec_input_res (tuple of (int, int)) – Shape of the input tensors.

  • layer_spec_target_res – (tuple of (int, int)): Shape of tensor desired as output of _get_layer_spec().

  • kernel_size (int) – Kernel used by the convolution layers.

  • initial_filters (int) – Numbers of filters to used as a base value.

  • filters_cap (int) – Cap filters to a set amount, in the case of an Encoder is a ceil value AKA the max amount of filters.

  • encoding_dimension (int) – encoding dimension.

  • channels (int) – Number of channels for the reconstructed image.

Returns

None

call(inputs, training=True)[source]

Execute the model on input data.

Parameters
  • inputs (tf.Tensor) – Input tensors.

  • training (bool) – Training flag.

Returns

(encoding, reconstruction) – Pair of tensors.