Python tensorflow.python.ops.rnn_cell_impl.RNNCell() Examples

def __init__(self, cell, output_size, activation=None, reuse=None):
    """Create a cell with output projection.

    Args:
      cell: an RNNCell, a projection to output_size is added to it.
      output_size: integer, the size of the output after projection.
      activation: (optional) an optional activation function.
      reuse: (optional) Python boolean describing whether to reuse variables
        in an existing scope.  If not `True`, and the existing scope already has
        the given variables, an error is raised.

    Raises:
      TypeError: if cell is not an RNNCell.
      ValueError: if output_size is not positive.
    """
    super(OutputProjectionWrapper, self).__init__(_reuse=reuse)
    if not _like_rnncell(cell):
      raise TypeError("The parameter cell is not RNNCell.")
    if output_size < 1:
      raise ValueError("Parameter output_size must be > 0: %d." % output_size)
    self._cell = cell
    self._output_size = output_size
    self._activation = activation 

def __init__(self, cell, num_proj, input_size=None):
    """Create a cell with input projection.

    Args:
      cell: an RNNCell, a projection of inputs is added before it.
      num_proj: Python integer.  The dimension to project to.
      input_size: Deprecated and unused.

    Raises:
      TypeError: if cell is not an RNNCell.
    """
    if input_size is not None:
      logging.warn("%s: The input_size parameter is deprecated.", self)
    if not isinstance(cell, RNNCell):
      raise TypeError("The parameter cell is not RNNCell.")
    self._cell = cell
    self._num_proj = num_proj 

def __init__(self, cell, output_size, activation=None, reuse=None):
    """Create a cell with output projection.
    Args:
      cell: an RNNCell, a projection to output_size is added to it.
      output_size: integer, the size of the output after projection.
      activation: (optional) an optional activation function.
      reuse: (optional) Python boolean describing whether to reuse variables
        in an existing scope.  If not `True`, and the existing scope already has
        the given variables, an error is raised.
    Raises:
      TypeError: if cell is not an RNNCell.
      ValueError: if output_size is not positive.
    """
    super(OutputProjectionWrapper, self).__init__(_reuse=reuse)
    if not _like_rnncell(cell):
      raise TypeError("The parameter cell is not RNNCell.")
    if output_size < 1:
      raise ValueError("Parameter output_size must be > 0: %d." % output_size)
    self._cell = cell
    self._output_size = output_size
    self._activation = activation
    self._linear = None 

def __init__(self, cell, output_size, reuse=None):
    """Create a cell with output projection.

    Args:
      cell: an RNNCell, a projection to output_size is added to it.
      output_size: integer, the size of the output after projection.
      reuse: (optional) Python boolean describing whether to reuse variables
        in an existing scope.  If not `True`, and the existing scope already has
        the given variables, an error is raised.

    Raises:
      TypeError: if cell is not an RNNCell.
      ValueError: if output_size is not positive.
    """
    if not isinstance(cell, RNNCell):
      raise TypeError("The parameter cell is not RNNCell.")
    if output_size < 1:
      raise ValueError("Parameter output_size must be > 0: %d." % output_size)
    self._cell = cell
    self._output_size = output_size
    self._reuse = reuse 

def basic_rnn_seq2seq(encoder_inputs,
                      decoder_inputs,
                      cell,
                      dtype=dtypes.float32,
                      scope=None):
  """Basic RNN sequence-to-sequence model.

  This model first runs an RNN to encode encoder_inputs into a state vector,
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell type, but don't share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    dtype: The dtype of the initial state of the RNN cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "basic_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in the final time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope(scope or "basic_rnn_seq2seq"):
    #enc_cell = copy.deepcopy(cell)
    enc_cell = copy.copy(cell)
    _, enc_state = rnn.static_rnn(enc_cell, encoder_inputs, dtype=dtype)
    return rnn_decoder(decoder_inputs, enc_state, cell) 

def basic_rnn_seq2seq(encoder_inputs,
                      decoder_inputs,
                      cell,
                      dtype=dtypes.float32,
                      scope=None):
  """Basic RNN sequence-to-sequence model.

  This model first runs an RNN to encode encoder_inputs into a state vector,
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell type, but don't share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    dtype: The dtype of the initial state of the RNN cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "basic_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in the final time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope(scope or "basic_rnn_seq2seq"):
    #enc_cell = copy.deepcopy(cell)
    enc_cell = copy.copy(cell)
    _, enc_state = rnn.static_rnn(enc_cell, encoder_inputs, dtype=dtype)
    return rnn_decoder(decoder_inputs, enc_state, cell) 

def __init__(self,
               cell,
               embedding_classes,
               embedding_size,
               initializer=None,
               reuse=None):
    """Create a cell with an added input embedding.

    Args:
      cell: an RNNCell, an embedding will be put before its inputs.
      embedding_classes: integer, how many symbols will be embedded.
      embedding_size: integer, the size of the vectors we embed into.
      initializer: an initializer to use when creating the embedding;
        if None, the initializer from variable scope or a default one is used.
      reuse: (optional) Python boolean describing whether to reuse variables
        in an existing scope.  If not `True`, and the existing scope already has
        the given variables, an error is raised.

    Raises:
      TypeError: if cell is not an RNNCell.
      ValueError: if embedding_classes is not positive.
    """
    super(EmbeddingWrapper, self).__init__(_reuse=reuse)
    if not _like_rnncell(cell):
      raise TypeError("The parameter cell is not RNNCell.")
    if embedding_classes <= 0 or embedding_size <= 0:
      raise ValueError("Both embedding_classes and embedding_size must be > 0: "
                       "%d, %d." % (embedding_classes, embedding_size))
    self._cell = cell
    self._embedding_classes = embedding_classes
    self._embedding_size = embedding_size
    self._initializer = initializer 

def basic_rnn_seq2seq(encoder_inputs,
                      decoder_inputs,
                      cell,
                      dtype=dtypes.float32,
                      scope=None):
  """Basic RNN sequence-to-sequence model.

  This model first runs an RNN to encode encoder_inputs into a state vector,
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell type, but don't share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    dtype: The dtype of the initial state of the RNN cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "basic_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in the final time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope(scope or "basic_rnn_seq2seq"):
    #enc_cell = copy.deepcopy(cell)
    enc_cell = copy.copy(cell)
    _, enc_state = rnn.static_rnn(enc_cell, encoder_inputs, dtype=dtype)
    return rnn_decoder(decoder_inputs, enc_state, cell) 

def basic_rnn_seq2seq(encoder_inputs,
                      decoder_inputs,
                      cell,
                      dtype=dtypes.float32,
                      scope=None):
  """Basic RNN sequence-to-sequence model.

  This model first runs an RNN to encode encoder_inputs into a state vector,
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell type, but don't share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    dtype: The dtype of the initial state of the RNN cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "basic_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in the final time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope(scope or "basic_rnn_seq2seq"):
    #enc_cell = copy.deepcopy(cell)
    enc_cell = copy.copy(cell)
    _, enc_state = rnn.static_rnn(enc_cell, encoder_inputs, dtype=dtype)
    return rnn_decoder(decoder_inputs, enc_state, cell) 

def tied_rnn_seq2seq(encoder_inputs,
                     decoder_inputs,
                     cell,
                     loop_function=None,
                     dtype=dtypes.float32,
                     scope=None):
  """RNN sequence-to-sequence model with tied encoder and decoder parameters.

  This model first runs an RNN to encode encoder_inputs into a state vector, and
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell and share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    loop_function: If not None, this function will be applied to i-th output
      in order to generate i+1-th input, and decoder_inputs will be ignored,
      except for the first element ("GO" symbol), see rnn_decoder for details.
    dtype: The dtype of the initial state of the rnn cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "tied_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in each time-step. This is a list
        with length len(decoder_inputs) -- one item for each time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope("combined_tied_rnn_seq2seq"):
    scope = scope or "tied_rnn_seq2seq"
    _, enc_state = rnn.static_rnn(
        cell, encoder_inputs, dtype=dtype, scope=scope)
    variable_scope.get_variable_scope().reuse_variables()
    return rnn_decoder(
        decoder_inputs,
        enc_state,
        cell,
        loop_function=loop_function,
        scope=scope) 

def basic_rnn_seq2seq(encoder_inputs,
                      decoder_inputs,
                      cell,
                      dtype=dtypes.float32,
                      scope=None):
  """Basic RNN sequence-to-sequence model.

  This model first runs an RNN to encode encoder_inputs into a state vector,
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell type, but don't share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    dtype: The dtype of the initial state of the RNN cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "basic_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in the final time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope(scope or "basic_rnn_seq2seq"):
    #enc_cell = copy.deepcopy(cell)
    enc_cell = copy.copy(cell)
    _, enc_state = rnn.static_rnn(enc_cell, encoder_inputs, dtype=dtype)
    return rnn_decoder(decoder_inputs, enc_state, cell) 

def basic_rnn_seq2seq(encoder_inputs,
                      decoder_inputs,
                      cell,
                      dtype=dtypes.float32,
                      scope=None):
  """Basic RNN sequence-to-sequence model.

  This model first runs an RNN to encode encoder_inputs into a state vector,
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell type, but don't share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    dtype: The dtype of the initial state of the RNN cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "basic_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in the final time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope(scope or "basic_rnn_seq2seq"):
    #enc_cell = copy.deepcopy(cell)
    enc_cell = copy.copy(cell)
    _, enc_state = rnn.static_rnn(enc_cell, encoder_inputs, dtype=dtype)
    return rnn_decoder(decoder_inputs, enc_state, cell) 

def tied_rnn_seq2seq(encoder_inputs,
                     decoder_inputs,
                     cell,
                     loop_function=None,
                     dtype=dtypes.float32,
                     scope=None):
  """RNN sequence-to-sequence model with tied encoder and decoder parameters.

  This model first runs an RNN to encode encoder_inputs into a state vector, and
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell and share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    loop_function: If not None, this function will be applied to i-th output
      in order to generate i+1-th input, and decoder_inputs will be ignored,
      except for the first element ("GO" symbol), see rnn_decoder for details.
    dtype: The dtype of the initial state of the rnn cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "tied_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in each time-step. This is a list
        with length len(decoder_inputs) -- one item for each time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope("combined_tied_rnn_seq2seq"):
    scope = scope or "tied_rnn_seq2seq"
    _, enc_state = rnn.static_rnn(
        cell, encoder_inputs, dtype=dtype, scope=scope)
    variable_scope.get_variable_scope().reuse_variables()
    return rnn_decoder(
        decoder_inputs,
        enc_state,
        cell,
        loop_function=loop_function,
        scope=scope) 

def basic_rnn_seq2seq(encoder_inputs,
                      decoder_inputs,
                      cell,
                      dtype=dtypes.float32,
                      scope=None):
  """Basic RNN sequence-to-sequence model.

  This model first runs an RNN to encode encoder_inputs into a state vector,
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell type, but don't share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    dtype: The dtype of the initial state of the RNN cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "basic_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in the final time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope(scope or "basic_rnn_seq2seq"):
    #enc_cell = copy.deepcopy(cell)
    enc_cell = copy.copy(cell)
    _, enc_state = rnn.static_rnn(enc_cell, encoder_inputs, dtype=dtype)
    return rnn_decoder(decoder_inputs, enc_state, cell) 

def tied_rnn_seq2seq(encoder_inputs,
                     decoder_inputs,
                     cell,
                     loop_function=None,
                     dtype=dtypes.float32,
                     scope=None):
  """RNN sequence-to-sequence model with tied encoder and decoder parameters.

  This model first runs an RNN to encode encoder_inputs into a state vector, and
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell and share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    loop_function: If not None, this function will be applied to i-th output
      in order to generate i+1-th input, and decoder_inputs will be ignored,
      except for the first element ("GO" symbol), see rnn_decoder for details.
    dtype: The dtype of the initial state of the rnn cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "tied_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in each time-step. This is a list
        with length len(decoder_inputs) -- one item for each time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope("combined_tied_rnn_seq2seq"):
    scope = scope or "tied_rnn_seq2seq"
    _, enc_state = rnn.static_rnn(
        cell, encoder_inputs, dtype=dtype, scope=scope)
    variable_scope.get_variable_scope().reuse_variables()
    return rnn_decoder(
        decoder_inputs,
        enc_state,
        cell,
        loop_function=loop_function,
        scope=scope) 

def tied_rnn_seq2seq(encoder_inputs,
                     decoder_inputs,
                     cell,
                     loop_function=None,
                     dtype=dtypes.float32,
                     scope=None):
  """RNN sequence-to-sequence model with tied encoder and decoder parameters.

  This model first runs an RNN to encode encoder_inputs into a state vector, and
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell and share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    loop_function: If not None, this function will be applied to i-th output
      in order to generate i+1-th input, and decoder_inputs will be ignored,
      except for the first element ("GO" symbol), see rnn_decoder for details.
    dtype: The dtype of the initial state of the rnn cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "tied_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in each time-step. This is a list
        with length len(decoder_inputs) -- one item for each time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope("combined_tied_rnn_seq2seq"):
    scope = scope or "tied_rnn_seq2seq"
    _, enc_state = rnn.static_rnn(
        cell, encoder_inputs, dtype=dtype, scope=scope)
    variable_scope.get_variable_scope().reuse_variables()
    return rnn_decoder(
        decoder_inputs,
        enc_state,
        cell,
        loop_function=loop_function,
        scope=scope) 

def basic_rnn_seq2seq(encoder_inputs,
                      decoder_inputs,
                      cell,
                      dtype=dtypes.float32,
                      scope=None):
  """Basic RNN sequence-to-sequence model.

  This model first runs an RNN to encode encoder_inputs into a state vector,
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell type, but don't share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    dtype: The dtype of the initial state of the RNN cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "basic_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in the final time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope(scope or "basic_rnn_seq2seq"):
    #enc_cell = copy.deepcopy(cell)
    enc_cell = copy.copy(cell)
    _, enc_state = rnn.static_rnn(enc_cell, encoder_inputs, dtype=dtype)
    return rnn_decoder(decoder_inputs, enc_state, cell) 

def basic_rnn_seq2seq(encoder_inputs,
                      decoder_inputs,
                      cell,
                      dtype=dtypes.float32,
                      scope=None):
  """Basic RNN sequence-to-sequence model.

  This model first runs an RNN to encode encoder_inputs into a state vector,
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell type, but don't share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    dtype: The dtype of the initial state of the RNN cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "basic_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in the final time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope(scope or "basic_rnn_seq2seq"):
    #enc_cell = copy.deepcopy(cell)
    enc_cell = copy.copy(cell)
    _, enc_state = rnn.static_rnn(enc_cell, encoder_inputs, dtype=dtype)
    return rnn_decoder(decoder_inputs, enc_state, cell) 

def tied_rnn_seq2seq(encoder_inputs,
                     decoder_inputs,
                     cell,
                     loop_function=None,
                     dtype=dtypes.float32,
                     scope=None):
  """RNN sequence-to-sequence model with tied encoder and decoder parameters.

  This model first runs an RNN to encode encoder_inputs into a state vector, and
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell and share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    loop_function: If not None, this function will be applied to i-th output
      in order to generate i+1-th input, and decoder_inputs will be ignored,
      except for the first element ("GO" symbol), see rnn_decoder for details.
    dtype: The dtype of the initial state of the rnn cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "tied_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in each time-step. This is a list
        with length len(decoder_inputs) -- one item for each time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope("combined_tied_rnn_seq2seq"):
    scope = scope or "tied_rnn_seq2seq"
    _, enc_state = rnn.static_rnn(
        cell, encoder_inputs, dtype=dtype, scope=scope)
    variable_scope.get_variable_scope().reuse_variables()
    return rnn_decoder(
        decoder_inputs,
        enc_state,
        cell,
        loop_function=loop_function,
        scope=scope) 

def basic_rnn_seq2seq(encoder_inputs,
                      decoder_inputs,
                      cell,
                      dtype=dtypes.float32,
                      scope=None):
  """Basic RNN sequence-to-sequence model.

  This model first runs an RNN to encode encoder_inputs into a state vector,
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell type, but don't share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    dtype: The dtype of the initial state of the RNN cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "basic_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in the final time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope(scope or "basic_rnn_seq2seq"):
    #enc_cell = copy.deepcopy(cell)
    enc_cell = copy.copy(cell)
    _, enc_state = rnn.static_rnn(enc_cell, encoder_inputs, dtype=dtype)
    return rnn_decoder(decoder_inputs, enc_state, cell) 

def __init__(self, cell, zoneout_prob, is_training=True, seed=None):
    if not isinstance(cell, rnn_cell_impl.RNNCell):
      raise TypeError("The parameter cell is not an RNNCell.")
    if (
        isinstance(zoneout_prob, float) and
        not (zoneout_prob >= 0.0 and zoneout_prob <= 1.0)
    ):
      raise ValueError(
          "Parameter zoneout_prob must be between 0 and 1: %d" % zoneout_prob
      )
    self._cell = cell
    self._zoneout_prob = (zoneout_prob, zoneout_prob)
    self._seed = seed
    self._is_training = is_training 

def __init__(self, cell, output_range, norm_op=None):
        """Rescales output range of @cell
            @cell: an RNN cell
            @output_range: range of outputs, e.g. 4 produces outputs in [-2, 2]
            @norm_op: function to map @cell outputs to range [0, 1]
        """
        if not isinstance(cell, RNNCell):
            raise TypeError("The parameter cell is not RNNCell.")
        if output_range < 0:
            raise ValueError("Logit range must be > 0: %d." % output_range)
        self._cell  = cell
        self._range = output_range
        self._norm  = norm_op 

def __init__(self, cell, hq, length, reuse=None):
    """Initialize the cell.

    Args:
      cell: An RNNCell.
      hq: The reference video feature.
      length: The length of the reference video.
      reuse: Whether reuse the parameters.
    """
    super(MatchCellWrapper, self).__init__(_reuse=reuse)
    self._cell = cell
    self._hq = hq
    self._length = length
    self._attn_vec_size = cell.output_size 

def __init__(self, cell,
               couple_carry_transform_gates=True,
               carry_bias_init=1.0):
    """Constructs a `HighwayWrapper` for `cell`.
    Args:
      cell: An instance of `RNNCell`.
      couple_carry_transform_gates: boolean, should the Carry and Transform gate
        be coupled.
      carry_bias_init: float, carry gates bias initialization.
    """
    self._cell = cell
    self._couple_carry_transform_gates = couple_carry_transform_gates
    self._carry_bias_init = carry_bias_init 

def __init__(self,
               cell,
               num_proj,
               activation=None,
               input_size=None,
               reuse=None):
    """Create a cell with input projection.

    Args:
      cell: an RNNCell, a projection of inputs is added before it.
      num_proj: Python integer.  The dimension to project to.
      activation: (optional) an optional activation function.
      input_size: Deprecated and unused.
      reuse: (optional) Python boolean describing whether to reuse variables
        in an existing scope.  If not `True`, and the existing scope already has
        the given variables, an error is raised.

    Raises:
      TypeError: if cell is not an RNNCell.
    """
    super(InputProjectionWrapper, self).__init__(_reuse=reuse)
    if input_size is not None:
      logging.warn("%s: The input_size parameter is deprecated.", self)
    if not _like_rnncell(cell):
      raise TypeError("The parameter cell is not RNNCell.")
    self._cell = cell
    self._num_proj = num_proj
    self._activation = activation 

def __init__(self, cell,
               couple_carry_transform_gates=True,
               carry_bias_init=1.0):
    """Constructs a `HighwayWrapper` for `cell`.

    Args:
      cell: An instance of `RNNCell`.
      couple_carry_transform_gates: boolean, should the Carry and Transform gate
        be coupled.
      carry_bias_init: float, carry gates bias initialization.
    """
    self._cell = cell
    self._couple_carry_transform_gates = couple_carry_transform_gates
    self._carry_bias_init = carry_bias_init 

def basic_rnn_seq2seq(encoder_inputs,
                      decoder_inputs,
                      cell,
                      dtype=dtypes.float32,
                      scope=None):
  """Basic RNN sequence-to-sequence model.

  This model first runs an RNN to encode encoder_inputs into a state vector,
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell type, but don't share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    dtype: The dtype of the initial state of the RNN cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "basic_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in the final time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope(scope or "basic_rnn_seq2seq"):
    enc_cell = copy.deepcopy(cell)
    _, enc_state = rnn.static_rnn(enc_cell, encoder_inputs, dtype=dtype)
    return rnn_decoder(decoder_inputs, enc_state, cell) 

def tied_rnn_seq2seq(encoder_inputs,
                     decoder_inputs,
                     cell,
                     loop_function=None,
                     dtype=dtypes.float32,
                     scope=None):
  """RNN sequence-to-sequence model with tied encoder and decoder parameters.

  This model first runs an RNN to encode encoder_inputs into a state vector, and
  then runs decoder, initialized with the last encoder state, on decoder_inputs.
  Encoder and decoder use the same RNN cell and share parameters.

  Args:
    encoder_inputs: A list of 2D Tensors [batch_size x input_size].
    decoder_inputs: A list of 2D Tensors [batch_size x input_size].
    cell: tf.nn.rnn_cell.RNNCell defining the cell function and size.
    loop_function: If not None, this function will be applied to i-th output
      in order to generate i+1-th input, and decoder_inputs will be ignored,
      except for the first element ("GO" symbol), see rnn_decoder for details.
    dtype: The dtype of the initial state of the rnn cell (default: tf.float32).
    scope: VariableScope for the created subgraph; default: "tied_rnn_seq2seq".

  Returns:
    A tuple of the form (outputs, state), where:
      outputs: A list of the same length as decoder_inputs of 2D Tensors with
        shape [batch_size x output_size] containing the generated outputs.
      state: The state of each decoder cell in each time-step. This is a list
        with length len(decoder_inputs) -- one item for each time-step.
        It is a 2D Tensor of shape [batch_size x cell.state_size].
  """
  with variable_scope.variable_scope("combined_tied_rnn_seq2seq"):
    scope = scope or "tied_rnn_seq2seq"
    _, enc_state = rnn.static_rnn(
        cell, encoder_inputs, dtype=dtype, scope=scope)
    variable_scope.get_variable_scope().reuse_variables()
    return rnn_decoder(
        decoder_inputs,
        enc_state,
        cell,
        loop_function=loop_function,
        scope=scope) 

def __init__(self,
               cell,
               embedding_classes,
               embedding_size,
               initializer=None,
               reuse=None):
    """Create a cell with an added input embedding.
    Args:
      cell: an RNNCell, an embedding will be put before its inputs.
      embedding_classes: integer, how many symbols will be embedded.
      embedding_size: integer, the size of the vectors we embed into.
      initializer: an initializer to use when creating the embedding;
        if None, the initializer from variable scope or a default one is used.
      reuse: (optional) Python boolean describing whether to reuse variables
        in an existing scope.  If not `True`, and the existing scope already has
        the given variables, an error is raised.
    Raises:
      TypeError: if cell is not an RNNCell.
      ValueError: if embedding_classes is not positive.
    """
    super(EmbeddingWrapper, self).__init__(_reuse=reuse)
    if not _like_rnncell(cell):
      raise TypeError("The parameter cell is not RNNCell.")
    if embedding_classes <= 0 or embedding_size <= 0:
      raise ValueError("Both embedding_classes and embedding_size must be > 0: "
                       "%d, %d." % (embedding_classes, embedding_size))
    self._cell = cell
    self._embedding_classes = embedding_classes
    self._embedding_size = embedding_size
    self._initializer = initializer 

def __init__(self,
               cell,
               num_proj,
               activation=None,
               input_size=None,
               reuse=None):
    """Create a cell with input projection.
    Args:
      cell: an RNNCell, a projection of inputs is added before it.
      num_proj: Python integer.  The dimension to project to.
      activation: (optional) an optional activation function.
      input_size: Deprecated and unused.
      reuse: (optional) Python boolean describing whether to reuse variables
        in an existing scope.  If not `True`, and the existing scope already has
        the given variables, an error is raised.
    Raises:
      TypeError: if cell is not an RNNCell.
    """
    super(InputProjectionWrapper, self).__init__(_reuse=reuse)
    if input_size is not None:
      logging.warn("%s: The input_size parameter is deprecated.", self)
    if not _like_rnncell(cell):
      raise TypeError("The parameter cell is not RNNCell.")
    self._cell = cell
    self._num_proj = num_proj
    self._activation = activation
    self._linear = None