Python tensorflow.python.ops.math_ops.div() Examples

def _optimal_step_size(last_step,
                       error_ratio,
                       safety=0.9,
                       ifactor=10.0,
                       dfactor=0.2,
                       order=5,
                       name=None):
  """Calculate the optimal size for the next Runge-Kutta step."""
  with ops.name_scope(
      name, 'optimal_step_size', [last_step, error_ratio]) as scope:
    error_ratio = math_ops.cast(error_ratio, last_step.dtype)
    exponent = math_ops.cast(1 / order, last_step.dtype)
    # this looks more complex than necessary, but importantly it keeps
    # error_ratio in the numerator so we can't divide by zero:
    factor = math_ops.maximum(
        1 / ifactor,
        math_ops.minimum(error_ratio ** exponent / safety, 1 / dfactor))
    return math_ops.div(last_step, factor, name=scope) 

def _length_penalty(sequence_lengths, penalty_factor):
  """Calculates the length penalty. See https://arxiv.org/abs/1609.08144.

  Returns the length penalty tensor:
  ```
  [(5+sequence_lengths)/6]**penalty_factor
  ```
  where all operations are performed element-wise.

  Args:
    sequence_lengths: `Tensor`, the sequence lengths of each hypotheses.
    penalty_factor: A scalar that weights the length penalty.

  Returns:
    If the penalty is `0`, returns the scalar `1.0`.  Otherwise returns
    the length penalty factor, a tensor with the same shape as
    `sequence_lengths`.
  """
  penalty_factor = ops.convert_to_tensor(penalty_factor, name="penalty_factor")
  penalty_factor.set_shape(())  # penalty should be a scalar.
  static_penalty = tensor_util.constant_value(penalty_factor)
  if static_penalty is not None and static_penalty == 0:
    return 1.0
  return math_ops.div((5. + math_ops.to_float(sequence_lengths))
                      ** penalty_factor, (5. + 1.)**penalty_factor) 

def _safe_div(numerator, denominator, name="value"):
  """Computes a safe divide which returns 0 if the denominator is zero.

  Note that the function contains an additional conditional check that is
  necessary for avoiding situations where the loss is zero causing NaNs to
  creep into the gradient computation.

  Args:
    numerator: An arbitrary `Tensor`.
    denominator: `Tensor` whose shape matches `numerator` and whose values are
      assumed to be non-negative.
    name: An optional name for the returned op.

  Returns:
    The element-wise value of the numerator divided by the denominator.
  """
  return array_ops.where(
      math_ops.greater(denominator, 0),
      math_ops.div(numerator, array_ops.where(
          math_ops.equal(denominator, 0),
          array_ops.ones_like(denominator), denominator)),
      array_ops.zeros_like(numerator),
      name=name) 

def _SegmentMinOrMaxGrad(op, grad, is_sorted):
  """Gradient for SegmentMin and (unsorted) SegmentMax. They share similar code."""
  zeros = array_ops.zeros(array_ops.shape(op.inputs[0]),
                          dtype=op.inputs[0].dtype)

  # Get the number of selected (minimum or maximum) elements in each segment.
  gathered_outputs = array_ops.gather(op.outputs[0], op.inputs[1])
  is_selected = math_ops.equal(op.inputs[0], gathered_outputs)
  if is_sorted:
    num_selected = math_ops.segment_sum(math_ops.cast(is_selected, grad.dtype),
                                        op.inputs[1])
  else:
    num_selected = math_ops.unsorted_segment_sum(math_ops.cast(is_selected, grad.dtype),
                                                 op.inputs[1], op.inputs[2])

  # Compute the gradient for each segment. The gradient for the ith segment is
  # divided evenly among the selected elements in that segment.
  weighted_grads = math_ops.div(grad, num_selected)
  gathered_grads = array_ops.gather(weighted_grads, op.inputs[1])

  if is_sorted:
    return array_ops.where(is_selected, gathered_grads, zeros), None
  else:
    return array_ops.where(is_selected, gathered_grads, zeros), None, None 

def tf_safe_div(numerator, denominator):
    """ Computes a safe division which returns 0 if the denominator is zero.
    Note that the function contains an additional conditional check that is
    necessary for avoiding situations where the loss is zero causing NaNs to
    creep into the gradient computation.

    Args:
      numerator: tf.tensor
      denominator: tf.tensor

    Returns: tf.tensor

    """

    return array_ops.where(
        math_ops.greater(denominator, 0),
        math_ops.div(numerator,
                     array_ops.where(
                         math_ops.equal(denominator, 0),
                         array_ops.ones_like(denominator), denominator)),
        array_ops.zeros_like(numerator)) 

def _safe_scalar_div(numerator, denominator, name):
  """Divides two values, returning 0 if the denominator is 0.

  Args:
    numerator: A scalar `float64` `Tensor`.
    denominator: A scalar `float64` `Tensor`.
    name: Name for the returned op.

  Returns:
    0 if `denominator` == 0, else `numerator` / `denominator`
  """
  numerator.get_shape().with_rank_at_most(1)
  denominator.get_shape().with_rank_at_most(1)
  return control_flow_ops.cond(
      math_ops.equal(
          array_ops.constant(0.0, dtype=dtypes.float64), denominator),
      lambda: array_ops.constant(0.0, dtype=dtypes.float64),
      lambda: math_ops.div(numerator, denominator),
      name=name) 

def inference_graph(self, input_data, data_spec=None):
    """Constructs a TF graph for evaluating a random forest.

    Args:
      input_data: A tensor or SparseTensor or placeholder for input data.
      data_spec: A list of tf.dtype values specifying the original types of
        each column.

    Returns:
      The last op in the random forest inference graph.
    """
    data_spec = [constants.DATA_FLOAT] if data_spec is None else data_spec
    probabilities = []
    for i in range(self.params.num_trees):
      with ops.device(self.device_assigner.get_device(i)):
        tree_data = input_data
        if self.params.bagged_features:
          tree_data = self._bag_features(i, input_data)
        probabilities.append(self.trees[i].inference_graph(tree_data,
                                                           data_spec))
    with ops.device(self.device_assigner.get_device(0)):
      all_predict = array_ops.pack(probabilities)
      return math_ops.div(
          math_ops.reduce_sum(all_predict, 0), self.params.num_trees,
          name='probabilities') 

def _MinOrMaxGrad(op, grad):
  """Gradient for Min or Max. Amazingly it's precisely the same code."""
  input_shape = array_ops.shape(op.inputs[0])
  output_shape_kept_dims = math_ops.reduced_shape(input_shape, op.inputs[1])
  y = op.outputs[0]
  y = array_ops.reshape(y, output_shape_kept_dims)
  grad = array_ops.reshape(grad, output_shape_kept_dims)

  # Compute the number of selected (maximum or minimum) elements in each
  # reduction dimension. If there are multiple minimum or maximum elements
  # then the gradient will be divided between them.
  indicators = math_ops.cast(math_ops.equal(y, op.inputs[0]), grad.dtype)
  num_selected = array_ops.reshape(
      math_ops.reduce_sum(indicators, op.inputs[1]), output_shape_kept_dims)

  return [math_ops.div(indicators, num_selected) * grad, None] 

def _length_penalty(sequence_lengths, penalty_factor):
  """Calculates the length penalty. See https://arxiv.org/abs/1609.08144.

  Args:
    sequence_lengths: The sequence length of all hypotheses, a tensor
      of shape [beam_size, vocab_size].
    penalty_factor: A scalar that weights the length penalty.

  Returns:
    The length penalty factor, a tensor fo shape [beam_size].
  """
  penalty_factor = ops.convert_to_tensor(penalty_factor, name="penalty_factor")
  penalty_factor.set_shape(())  # penalty should be a scalar.
  static_penalty = tensor_util.constant_value(penalty_factor)
  if static_penalty is not None and static_penalty == 0:
    return 1.0
  return math_ops.div((5. + math_ops.to_float(sequence_lengths))
                      **penalty_factor, (5. + 1.)**penalty_factor) 

def _safe_div(numerator, denominator, name='safe_div'):
    """Computes a safe divide which returns 0 if the denominator is zero.

    Args:
      numerator: An arbitrary `Tensor`.
      denominator: `Tensor` whose shape matches `numerator`.
      name: An optional name for the returned op.

    Returns:
      The element-wise value of the numerator divided by the denominator.
    """
    return array_ops.where(
        math_ops.equal(denominator, 0),
        array_ops.zeros_like(numerator),
        math_ops.div(numerator, denominator),
        name=name) 

def _sample_n(self, n, seed=None):
    sample_shape = array_ops.concat([[n], array_ops.shape(self.logits)], 0)
    logits = self.logits * array_ops.ones(sample_shape)
    logits_2d = array_ops.reshape(logits, [-1, self.event_size])
    # Uniform variates must be sampled from the open-interval `(0, 1)` rather
    # than `[0, 1)`. To do so, we use `np.finfo(self.dtype.as_numpy_dtype).tiny`
    # because it is the smallest, positive, "normal" number. A "normal" number
    # is such that the mantissa has an implicit leading 1. Normal, positive
    # numbers x, y have the reasonable property that, `x + y >= max(x, y)`. In
    # this case, a subnormal number (i.e., np.nextafter) can cause us to sample
    # 0.
    uniform = random_ops.random_uniform(
        shape=array_ops.shape(logits_2d),
        minval=np.finfo(self.dtype.as_numpy_dtype).tiny,
        maxval=1.,
        dtype=self.dtype,
        seed=seed)
    gumbel = -math_ops.log(-math_ops.log(uniform))
    noisy_logits = math_ops.div(gumbel + logits_2d, self._temperature_2d)
    samples = nn_ops.log_softmax(noisy_logits)
    ret = array_ops.reshape(samples, sample_shape)
    return ret 

def setUp(self):
    self.a = variables.Variable(2.0, name="a")
    self.b = variables.Variable(3.0, name="b")

    self.c = math_ops.multiply(self.a, self.b, name="c")  # Should be 6.0.
    self.d = math_ops.multiply(self.a, self.a, name="d")  # Should be 4.0.

    self.e = math_ops.multiply(self.d, self.c, name="e")  # Should be 24.0.

    self.f_y = constant_op.constant(0.30, name="f_y")
    self.f = math_ops.div(self.b, self.f_y, name="f")  # Should be 10.0.

    # The there nodes x, y and z form a graph with "cross-links" in. I.e., x
    # and y are both direct inputs to z, but x is also a direct input to y.
    self.x = variables.Variable(2.0, name="x")  # Should be 2.0
    self.y = math_ops.negative(self.x, name="y")  # Should be -2.0.

    self.z = math_ops.multiply(self.x, self.y, name="z")  # Should be -4.0.

    self.sess = session.Session()
    self.sess.run(variables.global_variables_initializer())

    self.sess = session.Session()
    self.sess.run(variables.global_variables_initializer()) 

def _safe_scalar_div(numerator, denominator, name):
  """Divides two values, returning 0 if the denominator is 0.

  Args:
    numerator: A scalar `float64` `Tensor`.
    denominator: A scalar `float64` `Tensor`.
    name: Name for the returned op.

  Returns:
    0 if `denominator` == 0, else `numerator` / `denominator`
  """
  numerator.get_shape().with_rank_at_most(1)
  denominator.get_shape().with_rank_at_most(1)
  return control_flow_ops.cond(
      math_ops.equal(
          array_ops.constant(0.0, dtype=dtypes.float64), denominator),
      lambda: array_ops.constant(0.0, dtype=dtypes.float64),
      lambda: math_ops.div(numerator, denominator),
      name=name) 

def _safe_div(numerator, denominator, name="value"):
  """Computes a safe divide which returns 0 if the denominator is zero.

  Note that the function contains an additional conditional check that is
  necessary for avoiding situations where the loss is zero causing NaNs to
  creep into the gradient computation.

  Args:
    numerator: An arbitrary `Tensor`.
    denominator: A `Tensor` whose shape matches `numerator` and whose values are
      assumed to be non-negative.
    name: An optional name for the returned op.

  Returns:
    The element-wise value of the numerator divided by the denominator.
  """
  return math_ops.select(
      math_ops.greater(denominator, 0),
      math_ops.div(numerator, math_ops.select(
          math_ops.equal(denominator, 0),
          array_ops.ones_like(denominator), denominator)),
      array_ops.zeros_like(numerator),
      name=name) 

def _optimal_step_size(last_step,
                       error_ratio,
                       safety=0.9,
                       ifactor=10.0,
                       dfactor=0.2,
                       order=5,
                       name=None):
  """Calculate the optimal size for the next Runge-Kutta step."""
  with ops.name_scope(
      name, 'optimal_step_size', [last_step, error_ratio]) as scope:
    error_ratio = math_ops.cast(error_ratio, last_step.dtype)
    exponent = math_ops.cast(1 / order, last_step.dtype)
    # this looks more complex than necessary, but importantly it keeps
    # error_ratio in the numerator so we can't divide by zero:
    factor = math_ops.maximum(
        1 / ifactor,
        math_ops.minimum(error_ratio ** exponent / safety, 1 / dfactor))
    return math_ops.div(last_step, factor, name=scope) 

def _MinOrMaxGrad(op, grad):
  """Gradient for Min or Max. Amazingly it's precisely the same code."""
  input_shape = array_ops.shape(op.inputs[0])
  output_shape_kept_dims = math_ops.reduced_shape(input_shape, op.inputs[1])
  y = op.outputs[0]
  y = array_ops.reshape(y, output_shape_kept_dims)
  grad = array_ops.reshape(grad, output_shape_kept_dims)

  # Compute the number of selected (maximum or minimum) elements in each
  # reduction dimension. If there are multiple minimum or maximum elements
  # then the gradient will be divided between them.
  indicators = math_ops.cast(math_ops.equal(y, op.inputs[0]), grad.dtype)
  num_selected = array_ops.reshape(
      math_ops.reduce_sum(indicators, op.inputs[1]), output_shape_kept_dims)

  return [math_ops.div(indicators, num_selected) * grad, None] 

def _SegmentMinOrMaxGrad(op, grad):
  """Gradient for SegmentMin and SegmentMax. Both share the same code."""
  zeros = array_ops.zeros(
      array_ops.shape(op.inputs[0]), dtype=op.inputs[0].dtype)

  # Get the number of selected (minimum or maximum) elements in each segment.
  gathered_outputs = array_ops.gather(op.outputs[0], op.inputs[1])
  is_selected = math_ops.equal(op.inputs[0], gathered_outputs)
  num_selected = math_ops.segment_sum(
      math_ops.cast(is_selected, grad.dtype), op.inputs[1])

  # Compute the gradient for each segment. The gradient for the ith segment is
  # divided evenly among the selected elements in that segment.
  weighted_grads = math_ops.div(grad, num_selected)
  gathered_grads = array_ops.gather(weighted_grads, op.inputs[1])

  return array_ops.where(is_selected, gathered_grads, zeros), None 

def _safe_div(numerator, denominator, name="value"):
  """Computes a safe divide which returns 0 if the denominator is zero.

  Note that the function contains an additional conditional check that is
  necessary for avoiding situations where the loss is zero causing NaNs to
  creep into the gradient computation.

  Args:
    numerator: An arbitrary `Tensor`.
    denominator: `Tensor` whose shape matches `numerator` and whose values are
      assumed to be non-negative.
    name: An optional name for the returned op.

  Returns:
    The element-wise value of the numerator divided by the denominator.
  """
  return array_ops.where(
      math_ops.greater(denominator, 0),
      math_ops.div(numerator, array_ops.where(
          math_ops.equal(denominator, 0),
          array_ops.ones_like(denominator), denominator)),
      array_ops.zeros_like(numerator),
      name=name) 

def _safe_scalar_div(numerator, denominator, name):
  """Divides two values, returning 0 if the denominator is 0.

  Args:
    numerator: A scalar `float64` `Tensor`.
    denominator: A scalar `float64` `Tensor`.
    name: Name for the returned op.

  Returns:
    0 if `denominator` == 0, else `numerator` / `denominator`
  """
  numerator.get_shape().with_rank_at_most(1)
  denominator.get_shape().with_rank_at_most(1)
  return control_flow_ops.cond(
      math_ops.equal(
          array_ops.constant(0.0, dtype=dtypes.float64), denominator),
      lambda: array_ops.constant(0.0, dtype=dtypes.float64),
      lambda: math_ops.div(numerator, denominator),
      name=name) 

def _safe_scalar_div(numerator, denominator, name):
  """Divides two values, returning 0 if the denominator is 0.

  Args:
    numerator: A scalar `float64` `Tensor`.
    denominator: A scalar `float64` `Tensor`.
    name: Name for the returned op.

  Returns:
    0 if `denominator` == 0, else `numerator` / `denominator`
  """
  numerator.get_shape().with_rank_at_most(1)
  denominator.get_shape().with_rank_at_most(1)
  return control_flow_ops.cond(
      math_ops.equal(
          array_ops.constant(0.0, dtype=dtypes.float64), denominator),
      lambda: array_ops.constant(0.0, dtype=dtypes.float64),
      lambda: math_ops.div(numerator, denominator),
      name=name) 

def _MinOrMaxGrad(op, grad):
  """Gradient for Min or Max. Amazingly it's precisely the same code."""
  input_shape = array_ops.shape(op.inputs[0])
  output_shape_kept_dims = math_ops.reduced_shape(input_shape, op.inputs[1])
  y = op.outputs[0]
  y = array_ops.reshape(y, output_shape_kept_dims)
  grad = array_ops.reshape(grad, output_shape_kept_dims)

  # Compute the number of selected (maximum or minimum) elements in each
  # reduction dimension. If there are multiple minimum or maximum elements
  # then the gradient will be divided between them.
  indicators = math_ops.cast(math_ops.equal(y, op.inputs[0]), grad.dtype)
  num_selected = array_ops.reshape(
      math_ops.reduce_sum(indicators, op.inputs[1]),
      output_shape_kept_dims)

  return [math_ops.div(indicators, num_selected) * grad, None] 

def _sample_n(self, n, seed=None):
    sample_shape = array_ops.concat(([n], array_ops.shape(self.logits)), 0)
    logits = self.logits * array_ops.ones(sample_shape)
    if logits.get_shape().ndims == 2:
      logits_2d = logits
    else:
      logits_2d = array_ops.reshape(logits, [-1, self.num_classes])
    np_dtype = self.dtype.as_numpy_dtype()
    minval = np.nextafter(np_dtype(0), np_dtype(1))
    uniform = random_ops.random_uniform(shape=array_ops.shape(logits_2d),
                                        minval=minval,
                                        maxval=1,
                                        dtype=self.dtype,
                                        seed=seed)
    gumbel = - math_ops.log(- math_ops.log(uniform))
    noisy_logits = math_ops.div(gumbel + logits_2d, self.temperature)
    samples = nn_ops.log_softmax(noisy_logits)
    ret = array_ops.reshape(samples, sample_shape)
    return ret 

def _safe_div(numerator, denominator, name="value"):
  """Computes a safe divide which returns 0 if the denominator is zero.

  Note that the function contains an additional conditional check that is
  necessary for avoiding situations where the loss is zero causing NaNs to
  creep into the gradient computation.

  Args:
    numerator: An arbitrary `Tensor`.
    denominator: A `Tensor` whose shape matches `numerator` and whose values are
      assumed to be non-negative.
    name: An optional name for the returned op.

  Returns:
    The element-wise value of the numerator divided by the denominator.
  """
  return array_ops.where(
      math_ops.greater(denominator, 0),
      math_ops.div(numerator, array_ops.where(
          math_ops.equal(denominator, 0),
          array_ops.ones_like(denominator), denominator)),
      array_ops.zeros_like(numerator),
      name=name) 

def accuracy(predictions, labels, weights=None, name=None):
  """Computes the percentage of times that predictions matches labels.

  Args:
    predictions: the predicted values, a `Tensor` whose dtype and shape
                 matches 'labels'.
    labels: the ground truth values, a `Tensor` of any shape and
            bool, integer, or string dtype.
    weights: None or `Tensor` of float values to reweight the accuracy.
    name: A name for the operation (optional).

  Returns:
    Accuracy `Tensor`.

  Raises:
    ValueError: if dtypes don't match or
                if dtype is not bool, integer, or string.
  """
  if not (labels.dtype.is_integer or
          labels.dtype in (dtypes.bool, dtypes.string)):
    raise ValueError(
        'Labels should have bool, integer, or string dtype, not %r' %
        labels.dtype)
  if not labels.dtype.is_compatible_with(predictions.dtype):
    raise ValueError('Dtypes of predictions and labels should match. '
                     'Given: predictions (%r) and labels (%r)' %
                     (predictions.dtype, labels.dtype))
  with ops.name_scope(name, 'accuracy', values=[predictions, labels]):
    is_correct = math_ops.cast(
        math_ops.equal(predictions, labels), dtypes.float32)
    if weights is not None:
      is_correct = math_ops.multiply(is_correct, weights)
      num_values = math_ops.multiply(weights, array_ops.ones_like(is_correct))
      return math_ops.div(math_ops.reduce_sum(is_correct),
                          math_ops.reduce_sum(num_values))
    return math_ops.reduce_mean(is_correct) 

def _input_dropout(self,inputs):
        # This implementation of dropout dropouts an entire feature along the time dim
        random_tensor = self._keep_prob
        random_tensor += random_ops.random_uniform([self._batch_size,self._num_inputs],
                                                   dtype=inputs.dtype)
        random_tensor = tf.tile(random_tensor,[1,self._max_unrollings])
        binary_tensor = math_ops.floor(random_tensor)

        ret = math_ops.div(inputs, self._keep_prob) * binary_tensor
        ret.set_shape(inputs.get_shape())

        return ret 

def unit_norm(inputs, dim, epsilon=1e-7, scope=None):
  """Normalizes the given input across the specified dimension to unit length.

  Note that the rank of `input` must be known.

  Args:
    inputs: A `Tensor` of arbitrary size.
    dim: The dimension along which the input is normalized.
    epsilon: A small value to add to the inputs to avoid dividing by zero.
    scope: Optional scope for variable_scope.

  Returns:
    The normalized `Tensor`.

  Raises:
    ValueError: If dim is smaller than the number of dimensions in 'inputs'.
  """
  with variable_scope.variable_scope(scope, 'UnitNorm', [inputs]):
    if not inputs.get_shape():
      raise ValueError('The input rank must be known.')
    input_rank = len(inputs.get_shape().as_list())
    if dim < 0 or dim >= input_rank:
      raise ValueError('dim must be positive but smaller than the input rank.')

    lengths = math_ops.sqrt(
        epsilon + math_ops.reduce_sum(math_ops.square(inputs), dim, True))
    multiples = []
    if dim > 0:
      multiples.append(array_ops.ones([dim], dtypes.int32))
    multiples.append(
        array_ops.strided_slice(array_ops.shape(inputs), [dim], [dim + 1]))
    if dim < (input_rank - 1):
      multiples.append(array_ops.ones([input_rank - 1 - dim], dtypes.int32))
    multiples = array_ops.concat(multiples, 0)
    return math_ops.div(inputs, array_ops.tile(lengths, multiples)) 

def _length_penalty(sequence_lengths, penalty_factor):
  """Calculates the length penalty.

  See https://arxiv.org/abs/1609.08144.

  Returns the length penalty tensor:
  ```
  [(5+sequence_lengths)/6]**penalty_factor
  ```
  where all operations are performed element-wise.

  Args:
    sequence_lengths: `Tensor`, the sequence lengths of each hypotheses.
    penalty_factor: A scalar that weights the length penalty.

  Returns:
    If the penalty is `0`, returns the scalar `1.0`.  Otherwise returns
    the length penalty factor, a tensor with the same shape as
    `sequence_lengths`.
  """
  penalty_factor = ops.convert_to_tensor(penalty_factor, name="penalty_factor")
  penalty_factor.set_shape(())  # penalty should be a scalar.
  static_penalty = tensor_util.constant_value(penalty_factor)
  if static_penalty is not None and static_penalty == 0:
    return 1.0
  return math_ops.div((5. + math_ops.to_float(sequence_lengths))
                      **penalty_factor, (5. + 1.)**penalty_factor) 

def inference_graph(self, input_data, data_spec=None, **inference_args):
    """Constructs a TF graph for evaluating a random forest.

    Args:
      input_data: A tensor or SparseTensor or placeholder for input data.
      data_spec: A list of tf.dtype values specifying the original types of
        each column.
      **inference_args: Keyword arguments to pass through to each tree.

    Returns:
      The last op in the random forest inference graph.
    """
    data_spec = [constants.DATA_FLOAT] if data_spec is None else data_spec
    probabilities = []
    for i in range(self.params.num_trees):
      with ops.device(self.device_assigner.get_device(i)):
        tree_data = input_data
        if self.params.bagged_features:
          tree_data = self._bag_features(i, input_data)
        probabilities.append(self.trees[i].inference_graph(
            tree_data, data_spec, **inference_args))
    with ops.device(self.device_assigner.get_device(0)):
      all_predict = array_ops.pack(probabilities)
      return math_ops.div(
          math_ops.reduce_sum(all_predict, 0), self.params.num_trees,
          name='probabilities') 

def _variational_recurrent_dropout_value(
      self, index, value, noise, keep_prob):
    """Performs dropout given the pre-calculated noise tensor."""
    # uniform [keep_prob, 1.0 + keep_prob)
    random_tensor = keep_prob + noise

    # 0. if [keep_prob, 1.0) and 1. if [1.0, 1.0 + keep_prob)
    binary_tensor = math_ops.floor(random_tensor)
    ret = math_ops.div(value, keep_prob) * binary_tensor
    ret.set_shape(value.get_shape())
    return ret 

def accuracy(predictions, labels, weights=None):
  """Computes the percentage of times that predictions matches labels.

  Args:
    predictions: the predicted values, a `Tensor` whose dtype and shape
                 matches 'labels'.
    labels: the ground truth values, a `Tensor` of any shape and
            bool, integer, or string dtype.
    weights: None or `Tensor` of float values to reweight the accuracy.

  Returns:
    Accuracy `Tensor`.

  Raises:
    ValueError: if dtypes don't match or
                if dtype is not bool, integer, or string.
  """
  if not (labels.dtype.is_integer or
          labels.dtype in (dtypes.bool, dtypes.string)):
    raise ValueError(
        'Labels should have bool, integer, or string dtype, not %r' %
        labels.dtype)
  if not labels.dtype.is_compatible_with(predictions.dtype):
    raise ValueError('Dtypes of predictions and labels should match. '
                     'Given: predictions (%r) and labels (%r)' %
                     (predictions.dtype, labels.dtype))
  with ops.name_scope('accuracy', values=[predictions, labels]):
    is_correct = math_ops.cast(
        math_ops.equal(predictions, labels), dtypes.float32)
    if weights is not None:
      is_correct = math_ops.mul(is_correct, weights)
      num_values = math_ops.mul(weights, array_ops.ones_like(is_correct))
      return math_ops.div(math_ops.reduce_sum(is_correct),
                          math_ops.reduce_sum(num_values))
    return math_ops.reduce_mean(is_correct)