Module: tfp.experimental.nn

Tools for building neural networks.

Modules

initializers module: Initializer functions for building neural networks.

losses module: Loss functions for neural networks.

util module: Utilitity functions for building neural networks.

Classes

class Affine: Basic affine layer.

class AffineVariationalFlipout: Densely-connected layer class with Flipout estimator.

class AffineVariationalReparameterization: Densely-connected layer class with reparameterization estimator.

class AffineVariationalReparameterizationLocal: Densely-connected layer class with local reparameterization estimator.

class Convolution: Convolution layer.

class ConvolutionTranspose: ConvolutionTranspose layer.

class ConvolutionTransposeVariationalFlipout: ConvolutionTranspose layer class with Flipout estimator.

class ConvolutionTransposeVariationalReparameterization: ConvolutionTranspose layer class with reparameterization estimator.

class ConvolutionV2: Convolution layer.

class ConvolutionVariationalFlipout: Convolution layer class with Flipout estimator.

class ConvolutionVariationalFlipoutV2: Convolution layer class with Flipout estimator.

class ConvolutionVariationalReparameterization: Convolution layer class with reparameterization estimator.

class ConvolutionVariationalReparameterizationV2: Convolution layer class with reparameterization estimator.

class Layer: A callable tf.Module.

class Sequential: A Layer characterized by iteratively given functions.

class VariationalLayer: Base class for all variational layers.