RNN cell composed sequentially of multiple simple cells.

API for TGIF dataset

Basic LSTM recurrent network cell. The implementation is based on: http://arxiv.org/pdf/1409.2329v5.pdf. It does not allow cell clipping, a proj

RNN cell composed sequentially of multiple simple cells.

Basic attention cell wrapper. Implementation based on https://arxiv.org/abs/1409.0473.

Bidirectional GridLstm cell. The bidirection connection is only used in the frequency direction, which hence doesn't affect the time direction's

Wraps step execution in an XLA JIT scope.

1D Convolutional LSTM recurrent network cell. https://arxiv.org/pdf/1506.04214v1.pdf

2D Convolutional LSTM recurrent network cell. https://arxiv.org/pdf/1506.04214v1.pdf

3D Convolutional LSTM recurrent network cell. https://arxiv.org/pdf/1506.04214v1.pdf

Convolutional LSTM recurrent network cell. https://arxiv.org/pdf/1506.04214v1.pdf

Long short-term memory unit (LSTM) recurrent network cell. The default non-peephole implementation is based on: http://www.bioinf.jku.at/publi

Operator adding dropout to inputs and outputs of the given cell.

Operator adding input embedding to the given cell. Note: in many cases it may be more efficient to not use this wrapper, but instead concatenate

Abstract object representing a fused RNN cell. A fused RNN cell represents the entire RNN expanded over the time dimension. In effect, this repre

This is an adaptor for RNNCell classes to be used with `FusedRNNCell`.

Group LSTM cell (G-LSTM). The implementation is based on: https://arxiv.org/abs/1703.10722 O. Kuchaiev and B. Ginsburg "Factorization Tri

r"""Block GRU cell implementation. Deprecated: use GRUBlockCellV2 instead. The implementation is based on: http://arxiv.org/abs/1406.1078 Com

Temporary GRUBlockCell impl with a different variable naming scheme. Only differs from GRUBlockCell by variable names.

Grid Long short-term memory unit (LSTM) recurrent network cell. The default is based on: Nal Kalchbrenner, Ivo Danihelka and Alex Graves "G

RNNCell wrapper that adds highway connection on cell input and output. Based on: R. K. Srivastava, K. Greff, and J. Schmidhuber, "Highway netwo

Operator adding an input projection to the given cell. Note: in many cases it may be more efficient to not use this wrapper, but instead concaten

Intersection Recurrent Neural Network (+RNN) cell. Architecture with coupled recurrent gate as well as coupled depth gate, designed to improve in

Basic LSTM recurrent network cell. The implementation is based on: http://arxiv.org/abs/1409.2329. We add `forget_bias` (default: 1) to the bias

FusedRNNCell implementation of LSTM. This is an extremely efficient LSTM implementation, that uses a single TF op for the entire LSTM. It should

This is a helper class that provides housekeeping for LSTM cells. This may be useful for alternative LSTM and similar type of cells. The subclass

LSTM unit with layer normalization and recurrent dropout. This class adds layer normalization and recurrent dropout to a basic LSTM unit. Layer n

LSTM unit with layer normalization and recurrent dropout. This class adds layer normalization and recurrent dropout to a basic LSTM unit. Layer n