|View source on GitHub|
RNN that accepts a state saver for time-truncated RNN calculation. (deprecated)
tf.nn.static_state_saving_rnn( cell, inputs, state_saver, state_name, sequence_length=None, scope=None )
cell: An instance of
inputs: A length T list of inputs, each a
state_saver: A state saver object with methods
state_name: Python string or tuple of strings. The name to use with the state_saver. If the cell returns tuples of states (i.e.,
cell.state_sizeis a tuple) then
state_nameshould be a tuple of strings having the same length as
cell.state_size. Otherwise it should be a single string.
sequence_length: (optional) An int32/int64 vector size [batch_size]. See the documentation for rnn() for more details about sequence_length.
scope: VariableScope for the created subgraph; defaults to "rnn".
A pair (outputs, state) where: outputs is a length T list of outputs (one for each input) states is the final state
cellis not an instance of RNNCell.
Noneor an empty list, or if the arity and type of
state_namedoes not match that of