contrib.seq2seq.AttentionWrapperState

tf.contrib.seq2seq.AttentionWrapperState

class tf.contrib.seq2seq.AttentionWrapperState

Defined in tensorflow/contrib/seq2seq/python/ops/attention_wrapper.py.

namedtuple storing the state of a AttentionWrapper.

Contains:

• cell_state: The state of the wrapped RNNCell at the previous time step.
• attention: The attention emitted at the previous time step.
• time: int32 scalar containing the current time step.
• alignments: The alignment emitted at the previous time step.
• alignment_history: (if enabled) a TensorArray containing alignment matrices from all time steps. Call stack() to convert to a Tensor.

Properties

alignment_history

Alias for field number 4

alignments

Alias for field number 3

attention

Alias for field number 1

cell_state

Alias for field number 0

time

Alias for field number 2

Methods

__new__

__new__(
    _cls,
    cell_state,
    attention,
    time,
    alignments,
    alignment_history
)

Create new instance of AttentionWrapperState(cell_state, attention, time, alignments, alignment_history)

clone

clone(**kwargs)

Clone this object, overriding components provided by kwargs.

Example:

initial_state = attention_wrapper.zero_state(dtype=..., batch_size=...)
initial_state = initial_state.clone(cell_state=encoder_state)

Args:

**kwargs: Any properties of the state object to replace in the returned AttentionWrapperState.

Returns:

A new AttentionWrapperState whose properties are the same as this one, except any overriden properties as provided in kwargs.