`namedlist` storing the state of a `attention_wrapper`.
attention_wrapper_state(
object,
cell_state,
attention,
alignments,
alignment_history,
attention_state
)Model or layer object
The state of the wrapped RNNCell at the previous time step.
The attention emitted at the previous time step.
A single or tuple of Tensor(s) containing the alignments emitted at the previous time step for each attention mechanism.
(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.
A single or tuple of nested objects containing attention mechanism state for each attention mechanism. The objects may contain Tensors or TensorArrays.
None