Applies Batch Normalization over a 2D or 3D input (a mini-batch of 1D inputs with optional additional channel dimension)

Applies Batch Normalization over a 4D input (a mini-batch of 2D inputs with optional additional channel dimension)

Applies Batch Normalization over a 5D input (a mini-batch of 3D inputs with optional additional channel dimension)

A model that applies a 1D convolution over an input signal composed of several input planes

A model that applies a 2D convolution over an input signal composed of several input planes

A model that applies a 3D convolution over an input signal composed of several input planes

A model that applies a 1D transposed convolution operator over an input image composed of several input planes.

A model that applies a 2D transposed convolution operator over an input image composed of several input planes.

A model that applies a 3D transposed convolution operator over an input image composed of several input planes.

A model which during training, randomly zeroes some of the elements of the input tensor with probability p using samples from a Bernoulli distribution.

A model which during training, randomly zero out entire channels. Each channel will be zeroed out independently on every forward call with probability p using samples from a Bernoulli distribution.

A model which during training, randomly zero out entire channels. Each channel will be zeroed out independently on every forward call with probability p using samples from a Bernoulli distribution.

A model that applies a linear transformation to the incoming data: \(y = xA^T + b\)

Long short-term memory (LSTM) recurrent neural network.

Unit cell of a long short-term memory (LSTM) recurrent neural network. Prefer using the RNN class instead, which can combine RNNCells in multiple layers.

Indicates the training or evaluation mode for a model.

Represents a model, primarily a collection of named parameters and sub-models and a function governed by them.

Represents the base class of all models.

Represents a parameter.

Represents a collection of named parameters.

Recurrent neural network.

Unit cell of a recurrent neural network. Prefer using the RNN class instead, which can combine RNNCells in multiple layers.

Variational auto-encoder

Variational auto-encoder base

Variational auto-encoder with multilayer perceptron (MLP) encoder and decoder.

Contains functionality related to generating initial parameter weights for models.