attention_wrapper_state: Attention Wrapper State

In tfaddons: Interface to 'TensorFlow SIG Addons'

Description

'namedlist' storing the state of a 'attention_wrapper'.

Usage

attention_wrapper_state(
  object,
  cell_state,
  attention,
  alignments,
  alignment_history,
  attention_state
)

Arguments

object	Model or layer object
cell_state	The state of the wrapped RNNCell at the previous time step.
attention	The attention emitted at the previous time step.
alignments	A single or tuple of Tensor(s) containing the alignments emitted at the previous time step for each attention mechanism.
alignment_history	(if enabled) a single or tuple of TensorArray(s) containing alignment matrices from all time steps for each attention mechanism. Call stack() on each to convert to a Tensor.
attention_state	A single or tuple of nested objects containing attention mechanism state for each attention mechanism. The objects may contain Tensors or TensorArrays.

Value

None

tfaddons documentation built on July 2, 2020, 2:12 a.m.