Python tensorflow.python.framework.ops.convert_to_tensor() Examples

def set_global_step(self, new_global_step, name=None):
    """Sets the global time step of the accumulator.

    The operation logs a warning if we attempt to set to a time step that is
    lower than the accumulator's own time step.

    Args:
      new_global_step: Value of new time step. Can be a variable or a constant
      name: Optional name for the operation.

    Returns:
      Operation that sets the accumulator's time step.
    """
    return gen_data_flow_ops.accumulator_set_global_step(
        self._accumulator_ref,
        math_ops.to_int64(ops.convert_to_tensor(new_global_step)),
        name=name) 

def size_internal(input, name=None, optimize=True, out_type=dtypes.int32):
  # pylint: disable=redefined-builtin,protected-access
  """Returns the size of a tensor.

  Args:
    input: A `Tensor` or `SparseTensor`.
    name: A name for the operation (optional).
    optimize: if true, encode the size as a constant when possible.
    out_type: (Optional) The specified output type of the operation
      (`int32` or `int64`). Defaults to tf.int32.

  Returns:
    A `Tensor` of type `out_type`.
  """
  with ops.name_scope(name, "Size", [input]) as name:
    if isinstance(
        input, (sparse_tensor.SparseTensor, sparse_tensor.SparseTensorValue)):
      return gen_math_ops._prod(
          gen_math_ops.cast(input.dense_shape, out_type), 0, name=name)
    else:
      input_tensor = ops.convert_to_tensor(input)
      input_shape = input_tensor.get_shape()
      if optimize and input_shape.is_fully_defined():
        return constant(input_shape.num_elements(), out_type, name=name)
      return gen_array_ops.size(input, name=name, out_type=out_type) 

def _lower_bound(inputs, bound, name=None):
    """Same as tf.maximum, but with helpful gradient for inputs < bound.

    The gradient is overwritten so that it is passed through if the input is not
    hitting the bound. If it is, only gradients that push `inputs` higher than
    the bound are passed through. No gradients are passed through to the bound.

    Args:
      inputs: input tensor
      bound: lower bound for the input tensor
      name: name for this op

    Returns:
      tf.maximum(inputs, bound)
    """
    with ops.name_scope(name, 'GDNLowerBound', [inputs, bound]) as scope:
      inputs = ops.convert_to_tensor(inputs, name='inputs')
      bound = ops.convert_to_tensor(bound, name='bound')
      with ops.get_default_graph().gradient_override_map(
          {'Maximum': 'GDNLowerBound'}):
        return math_ops.maximum(inputs, bound, name=scope) 

def flatten(inputs, outputs_collections=None, scope=None):
  """Flattens the input while maintaining the batch_size.

    Assumes that the first dimension represents the batch.

  Args:
    inputs: A tensor of size [batch_size, ...].
    outputs_collections: Collection to add the outputs.
    scope: Optional scope for name_scope.

  Returns:
    A flattened tensor with shape [batch_size, k].
  Raises:
    ValueError: If inputs rank is unknown or less than 2.
  """
  with ops.name_scope(scope, 'Flatten', [inputs]) as sc:
    inputs = ops.convert_to_tensor(inputs)
    outputs = core_layers.flatten(inputs)
    return utils.collect_named_outputs(outputs_collections, sc, outputs) 

def rank_internal(input, name=None, optimize=True):
  # pylint: disable=redefined-builtin
  """Returns the rank of a tensor.

  Args:
    input: A `Tensor` or `SparseTensor`.
    name: A name for the operation (optional).
    optimize: if true, encode the rank as a constant when possible.

  Returns:
    A `Tensor` of type `int32`.
  """
  with ops.name_scope(name, "Rank", [input]) as name:
    if isinstance(
        input, (sparse_tensor.SparseTensor, sparse_tensor.SparseTensorValue)):
      return gen_array_ops.size(input.dense_shape, name=name)
    else:
      input_tensor = ops.convert_to_tensor(input)
      input_shape = input_tensor.get_shape()
      if optimize and input_shape.ndims is not None:
        return constant(input_shape.ndims, dtypes.int32, name=name)
      return gen_array_ops.rank(input, name=name) 

def dense_to_sparse(tensor, eos_token=0, outputs_collections=None, scope=None):
  """Converts a dense tensor into a sparse tensor.

  An example use would be to convert dense labels to sparse ones
  so that they can be fed to the ctc_loss.

  Args:
     tensor: An `int` `Tensor` to be converted to a `Sparse`.
     eos_token: An integer. It is part of the target label that signifies the
       end of a sentence.
     outputs_collections: Collection to add the outputs.
     scope: Optional scope for name_scope.
  """
  with variable_scope.variable_scope(scope, 'dense_to_sparse', [tensor]) as sc:
    tensor = ops.convert_to_tensor(tensor)
    indices = array_ops.where(
        math_ops.not_equal(tensor, constant_op.constant(eos_token,
                                                        tensor.dtype)))
    values = array_ops.gather_nd(tensor, indices)
    shape = array_ops.shape(tensor, out_type=dtypes.int64)
    outputs = sparse_tensor.SparseTensor(indices, values, shape)
    return utils.collect_named_outputs(outputs_collections, sc.name, outputs) 

def random_flip_left_right(image, bboxes, seed=None):
    """Random flip left-right of an image and its bounding boxes.
    """
    def flip_bboxes(bboxes):
        """Flip bounding boxes coordinates.
        """
        bboxes = tf.stack([bboxes[:, 0], 1 - bboxes[:, 3],
                           bboxes[:, 2], 1 - bboxes[:, 1]], axis=-1)
        return bboxes

    # Random flip. Tensorflow implementation.
    with tf.name_scope('random_flip_left_right'):
        image = ops.convert_to_tensor(image, name='image')
        _Check3DImage(image, require_static=False)
        uniform_random = random_ops.random_uniform([], 0, 1.0, seed=seed)
        mirror_cond = math_ops.less(uniform_random, .5)
        # Flip image.
        result = control_flow_ops.cond(mirror_cond,
                                       lambda: array_ops.reverse_v2(image, [1]),
                                       lambda: image)
        # Flip bboxes.
        bboxes = control_flow_ops.cond(mirror_cond,
                                       lambda: flip_bboxes(bboxes),
                                       lambda: bboxes)
        return fix_image_flip_shape(image, result), bboxes 

def shape_internal(input, name=None, optimize=True, out_type=dtypes.int32):
  # pylint: disable=redefined-builtin
  """Returns the shape of a tensor.

  Args:
    input: A `Tensor` or `SparseTensor`.
    name: A name for the operation (optional).
    optimize: if true, encode the shape as a constant when possible.
    out_type: (Optional) The specified output type of the operation
      (`int32` or `int64`). Defaults to tf.int32.

  Returns:
    A `Tensor` of type `out_type`.

  """
  with ops.name_scope(name, "Shape", [input]) as name:
    if isinstance(
        input, (sparse_tensor.SparseTensor, sparse_tensor.SparseTensorValue)):
      return gen_math_ops.cast(input.dense_shape, out_type)
    else:
      input_tensor = ops.convert_to_tensor(input)
      input_shape = input_tensor.get_shape()
      if optimize and input_shape.is_fully_defined():
        return constant(input_shape.as_list(), out_type, name=name)
      return gen_array_ops.shape(input, name=name, out_type=out_type) 

def apply_grad(self, grad, local_step=0, name=None):
    """Attempts to apply a gradient to the accumulator.

    The attempt is silently dropped if the gradient is stale, i.e., local_step
    is less than the accumulator's global time step.

    Args:
      grad: The gradient tensor to be applied.
      local_step: Time step at which the gradient was computed.
      name: Optional name for the operation.

    Returns:
      The operation that (conditionally) applies a gradient to the accumulator.

    Raises:
      ValueError: If grad is of the wrong shape
    """
    grad = ops.convert_to_tensor(grad, self._dtype)
    grad.get_shape().assert_is_compatible_with(self._shape)
    local_step = math_ops.to_int64(ops.convert_to_tensor(local_step))
    return gen_data_flow_ops.accumulator_apply_gradient(
        self._accumulator_ref, local_step=local_step, gradient=grad, name=name) 

def binary_cross_entropy(preds, targets, name=None):
    """Computes binary cross entropy given `preds`.

    For brevity, let `x = `, `z = targets`.  The logistic loss is

        loss(x, z) = - sum_i (x[i] * log(z[i]) + (1 - x[i]) * log(1 - z[i]))

    Args:
        preds: A `Tensor` of type `float32` or `float64`.
        targets: A `Tensor` of the same type and shape as `preds`.
    """
    eps = 1e-12
    with ops.op_scope([preds, targets], name, "bce_loss") as name:
        preds = ops.convert_to_tensor(preds, name="preds")
        targets = ops.convert_to_tensor(targets, name="targets")
        return tf.reduce_mean(-(targets * tf.log(preds + eps) +
                              (1. - targets) * tf.log(1. - preds + eps))) 

def shape_list(x):
    """Returns **static** shape of the input Tensor whenever possible.

    Args:
        x: A Tensor.

    Returns:
        - If the rank of :attr:`x` is unknown, returns the dynamic shape: \
        `tf.shape(x)`
        - Otherwise, returns a list of dims, each of which is either an `int` \
        whenever it can be statically determined, or a scalar Tensor otherwise.
    """
    x = tf.convert_to_tensor(x)
    # If unknown rank, return dynamic shape
    if x.get_shape().dims is None:
        return tf.shape(x)
    static = x.get_shape().as_list()
    shape = tf.shape(x)
    ret = []
    for i, dim in enumerate(static):
        if dim is None:
            dim = shape[i]
        ret.append(dim)
    return ret 

def __init__(self, sample_fn, sample_shape, sample_dtype,
                 start_inputs, end_fn, next_inputs_fn=None):
        """Initializer.

        Args:
          sample_fn: A callable that takes `outputs` and emits tensor `sample_ids`.
          sample_shape: Either a list of integers, or a 1-D Tensor of type `int32`,
            the shape of the each sample in the batch returned by `sample_fn`.
          sample_dtype: the dtype of the sample returned by `sample_fn`.
          start_inputs: The initial batch of inputs.
          end_fn: A callable that takes `sample_ids` and emits a `bool` vector
            shaped `[batch_size]` indicating whether each sample is an end token.
          next_inputs_fn: (Optional) A callable that takes `sample_ids` and returns
            the next batch of inputs. If not provided, `sample_ids` is used as the
            next batch of inputs.
        """
        self._sample_fn = sample_fn
        self._end_fn = end_fn
        self._sample_shape = tensor_shape.TensorShape(sample_shape)
        self._sample_dtype = sample_dtype
        self._next_inputs_fn = next_inputs_fn
        self._batch_size = array_ops.shape(start_inputs)[0]
        self._start_inputs = ops.convert_to_tensor(
            start_inputs, name="start_inputs") 

def _tile_batch(t, multiplier):
    """Core single-tensor implementation of tile_batch."""
    t = ops.convert_to_tensor(t, name="t")
    shape_t = tf.shape(t)
    if t.shape.ndims is None or t.shape.ndims < 1:
        raise ValueError("t must have statically known rank")
    tiling = [1] * (t.shape.ndims + 1)
    tiling[1] = multiplier
    tiled_static_batch_size = (
        t.shape[0].value * multiplier if t.shape[0].value is not None else None)
    tiled = tf.tile(tf.expand_dims(t, 1), tiling)
    tiled = tf.reshape(
        tiled, tf.concat(([shape_t[0] * multiplier], shape_t[1:]), 0))
    tiled.set_shape(
        tensor_shape.TensorShape(
            [tiled_static_batch_size]).concatenate(t.shape[1:]))
    return tiled 

def bias_add_v1(value, bias, name=None):
  """Adds `bias` to `value`.

  This is a deprecated version of bias_add and will soon to be removed.

  This is (mostly) a special case of `tf.add` where `bias` is restricted to 1-D.
  Broadcasting is supported, so `value` may have any number of dimensions.
  Unlike `tf.add`, the type of `bias` is allowed to differ from `value` in the
  case where both types are quantized.

  Args:
    value: A `Tensor` with type `float`, `double`, `int64`, `int32`, `uint8`,
      `int16`, `int8`, `complex64`, or `complex128`.
    bias: A 1-D `Tensor` with size matching the last dimension of `value`.
      Must be the same type as `value` unless `value` is a quantized type,
      in which case a different quantized type may be used.
    name: A name for the operation (optional).

  Returns:
    A `Tensor` with the same type as `value`.
  """
  with ops.name_scope(name, "BiasAddV1", [value, bias]) as name:
    value = ops.convert_to_tensor(value, name="input")
    bias = ops.convert_to_tensor(bias, dtype=value.dtype, name="bias")
    return gen_nn_ops._bias_add_v1(value, bias, name=name) 

def crelu(features, name=None):
  """Computes Concatenated ReLU.

  Concatenates a ReLU which selects only the positive part of the activation
  with a ReLU which selects only the *negative* part of the activation.
  Note that as a result this non-linearity doubles the depth of the activations.
  Source: [Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear Units. W. Shang, et al.](https://arxiv.org/abs/1603.05201) 

  Args:
    features: A `Tensor` with type `float`, `double`, `int32`, `int64`, `uint8`,
      `int16`, or `int8`.
    name: A name for the operation (optional).

  Returns:
    A `Tensor` with the same type as `features`.
  """
  with ops.name_scope(name, "CRelu", [features]) as name:
    features = ops.convert_to_tensor(features, name="features")
    c = array_ops.concat([features, -features], -1, name=name)
    return gen_nn_ops.relu(c) 

def binary_cross_entropy(preds, targets, name=None):
	"""Computes binary cross entropy given `preds`.

	For brevity, let `x = `, `z = targets`.  The logistic loss is

		loss(x, z) = - sum_i (x[i] * log(z[i]) + (1 - x[i]) * log(1 - z[i]))

	Args:
		preds: A `Tensor` of type `float32` or `float64`.
		targets: A `Tensor` of the same type and shape as `preds`.
	"""
	eps = 1e-12
	with ops.op_scope([preds, targets], name, "bce_loss") as name:
		preds = ops.convert_to_tensor(preds, name="preds")
		targets = ops.convert_to_tensor(targets, name="targets")
		return tf.reduce_mean(-(targets * tf.log(preds + eps) +
							  (1. - targets) * tf.log(1. - preds + eps))) 

def xw_plus_b(x, weights, biases, name=None):  # pylint: disable=invalid-name
  """Computes matmul(x, weights) + biases.

  Args:
    x: a 2D tensor.  Dimensions typically: batch, in_units
    weights: a 2D tensor.  Dimensions typically: in_units, out_units
    biases: a 1D tensor.  Dimensions: out_units
    name: A name for the operation (optional).  If not specified
      "xw_plus_b" is used.

  Returns:
    A 2-D Tensor computing matmul(x, weights) + biases.
    Dimensions typically: batch, out_units.
  """
  with ops.name_scope(name, "xw_plus_b", [x, weights, biases]) as name:
    x = ops.convert_to_tensor(x, name="x")
    weights = ops.convert_to_tensor(weights, name="weights")
    biases = ops.convert_to_tensor(biases, name="biases")
    mm = math_ops.matmul(x, weights)
    return bias_add(mm, biases, name=name) 

def xw_plus_b_v1(x, weights, biases, name=None):  # pylint: disable=invalid-name
  """Computes matmul(x, weights) + biases.

  This is a deprecated version of that will soon be removed.

  Args:
    x: a 2D tensor.  Dimensions typically: batch, in_units
    weights: a 2D tensor.  Dimensions typically: in_units, out_units
    biases: a 1D tensor.  Dimensions: out_units
    name: A name for the operation (optional).  If not specified
      "xw_plus_b_v1" is used.

  Returns:
    A 2-D Tensor computing matmul(x, weights) + biases.
    Dimensions typically: batch, out_units.
  """
  with ops.name_scope(name, "xw_plus_b_v1", [x, weights, biases]) as name:
    x = ops.convert_to_tensor(x, name="x")
    weights = ops.convert_to_tensor(weights, name="weights")
    biases = ops.convert_to_tensor(biases, name="biases")
    mm = math_ops.matmul(x, weights)
    return bias_add_v1(mm, biases, name=name) 

def shape_list(x):
  """Return list of dims, statically where possible."""
  x = tf.convert_to_tensor(x)

  # If unknown rank, return dynamic shape
  if x.get_shape().dims is None:
    return tf.shape(x)

  static = x.get_shape().as_list()
  shape = tf.shape(x)

  ret = []
  for i in range(len(static)):
    dim = static[i]
    if dim is None:
      dim = shape[i]
    ret.append(dim)
  return ret 

def _prepare(self):
    learning_rate = self._call_if_callable(self._learning_rate)
    self._learning_rate_tensor = ops.convert_to_tensor(
        learning_rate, name="learning_rate")
    momentum = self._call_if_callable(self._momentum)
    self._momentum_tensor = ops.convert_to_tensor(momentum, name="momentum") 

def max_pool(value, ksize, strides, padding, data_format="NHWC", name=None):
  """Performs the max pooling on the input.

  Args:
    value: A 4-D `Tensor` with shape `[batch, height, width, channels]` and
      type `tf.float32`.
    ksize: A list of ints that has length >= 4.  The size of the window for
      each dimension of the input tensor.
    strides: A list of ints that has length >= 4.  The stride of the sliding
      window for each dimension of the input tensor.
    padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm.
      See the @{tf.nn.convolution$comment here}
    data_format: A string. 'NHWC' and 'NCHW' are supported.
    name: Optional name for the operation.

  Returns:
    A `Tensor` with type `tf.float32`.  The max pooled output tensor.
  """
  with ops.name_scope(name, "MaxPool", [value]) as name:
    value = ops.convert_to_tensor(value, name="input")
    return gen_nn_ops._max_pool(value,
                                ksize=ksize,
                                strides=strides,
                                padding=padding,
                                data_format=data_format,
                                name=name) 

def bias_add(value, bias, data_format=None, name=None):
  """Adds `bias` to `value`.

  This is (mostly) a special case of `tf.add` where `bias` is restricted to 1-D.
  Broadcasting is supported, so `value` may have any number of dimensions.
  Unlike `tf.add`, the type of `bias` is allowed to differ from `value` in the
  case where both types are quantized.

  Args:
    value: A `Tensor` with type `float`, `double`, `int64`, `int32`, `uint8`,
      `int16`, `int8`, `complex64`, or `complex128`.
    bias: A 1-D `Tensor` with size matching the last dimension of `value`.
      Must be the same type as `value` unless `value` is a quantized type,
      in which case a different quantized type may be used.
    data_format: A string. 'NHWC' and 'NCHW' are supported.
    name: A name for the operation (optional).

  Returns:
    A `Tensor` with the same type as `value`.
  """
  with ops.name_scope(name, "BiasAdd", [value, bias]) as name:
    value = ops.convert_to_tensor(value, name="input")
    bias = ops.convert_to_tensor(bias, dtype=value.dtype, name="bias")
    return gen_nn_ops._bias_add(value, bias, data_format=data_format, name=name)


# pylint: disable=protected-access 

def avg_pool(value, ksize, strides, padding, data_format="NHWC", name=None):
  """Performs the average pooling on the input.

  Each entry in `output` is the mean of the corresponding size `ksize`
  window in `value`.

  Args:
    value: A 4-D `Tensor` of shape `[batch, height, width, channels]` and type
      `float32`, `float64`, `qint8`, `quint8`, or `qint32`.
    ksize: A list of ints that has length >= 4.
      The size of the window for each dimension of the input tensor.
    strides: A list of ints that has length >= 4.
      The stride of the sliding window for each dimension of the
      input tensor.
    padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm.
      See the @{tf.nn.convolution$comment here}
    data_format: A string. 'NHWC' and 'NCHW' are supported.
    name: Optional name for the operation.

  Returns:
    A `Tensor` with the same type as `value`.  The average pooled output tensor.
  """
  with ops.name_scope(name, "AvgPool", [value]) as name:
    value = ops.convert_to_tensor(value, name="input")
    return gen_nn_ops._avg_pool(value,
                                ksize=ksize,
                                strides=strides,
                                padding=padding,
                                data_format=data_format,
                                name=name) 

def __init__(self, batch_size, max_iters, model_path, data_ob, sample_path, log_dir, learning_rate, l1_w, per_w, recon_w):

        self.batch_size = batch_size
        self.max_iters = max_iters
        self.pixel_model_path = model_path
        self.data_ob = data_ob
        self.sample_path = sample_path
        self.log_dir = log_dir
        self.learning_rate = learning_rate
        self.log_vars = []
        self.channel = data_ob.channel
        self.shape = data_ob.shape
        self.class_nums = 2
        self.output_size = data_ob.image_size
        self.l1_w = l1_w
        self.per_w = per_w
        self.recon_w = recon_w

        self.dom_1_images = tf.placeholder(tf.float32, [batch_size, self.output_size, self.output_size, self.channel])
        self.dom_2_images = tf.placeholder(tf.float32, [batch_size, self.output_size, self.output_size, self.channel])

        self.dataset = tf.data.Dataset.from_tensor_slices((convert_to_tensor(self.data_ob.dom_1_train_data_list, dtype=tf.string),
                                                           convert_to_tensor(self.data_ob.dom_2_train_data_list, dtype=tf.string)))

        self.dataset = self.dataset.shuffle(buffer_size=len(self.data_ob.dom_1_train_data_list))

        self.dataset = self.dataset.map(lambda filename1, filename2: tuple(
            tf.py_func(self._read_by_function, [filename1, filename2], [tf.float32, tf.float32])), num_parallel_calls=32)
        self.dataset = self.dataset.repeat(50000)
        self.dataset = self.dataset.apply(tf.contrib.data.batch_and_drop_remainder(batch_size))
        self.iterator = tf.data.Iterator.from_structure(self.dataset.output_types, self.dataset.output_shapes)

        self.next_images1, self.next_images2 = self.iterator.get_next()
        self.train_init_op = self.iterator.make_initializer(self.dataset)
        self.domain_label = tf.placeholder(tf.int32, [batch_size])

        self.lr_decay = tf.placeholder(tf.float32, None, name='lr_decay') 

def _prepare(self):
        self._lr_t = ops.convert_to_tensor(self._lr)
        self._beta1_t = ops.convert_to_tensor(self._beta1)
        self._beta2_t = ops.convert_to_tensor(self._beta2)
        self._epsilon_t = ops.convert_to_tensor(self._epsilon) 

def _prepare(self):
    learning_rate = self._call_if_callable(self._learning_rate)
    self._learning_rate_tensor = ops.convert_to_tensor(
        learning_rate, name="learning_rate")
    momentum = self._call_if_callable(self._momentum)
    self._momentum_tensor = ops.convert_to_tensor(momentum, name="momentum") 

def _create_slots(self, var_list):
    for v in var_list:
      self._zeros_slot(v, "momentum", self._name)
      self._zeros_slot(v, "gbar", self._name)
      g_tensor = ops.convert_to_tensor(v)
      gain_init = self._initial_gain * array_ops.ones_like(g_tensor)
      _ = self._get_or_make_slot(v, self._initial_scale * array_ops.ones((1)),
                                 "lr_scale", self._name)
      _ = self._get_or_make_slot(v, gain_init, "gain", self._name)
      _ = self._get_or_make_slot(v, array_ops.zeros((1)), "counter", self._name) 

def _prepare(self):
    self._lr_t = ops.convert_to_tensor(self._lr, name="learning_rate")
    self._beta1_t = ops.convert_to_tensor(self._beta1, name="beta1")
    self._beta2_t = ops.convert_to_tensor(self._beta2, name="beta2") 

def _prepare(self):
        self._lr_t = ops.convert_to_tensor(self._lr)
        self._base_lr_t = ops.convert_to_tensor(self._lr)
        self._beta1_t = ops.convert_to_tensor(self._beta1)
        self._beta2_t = ops.convert_to_tensor(self._beta2)
        self._epsilon_t = ops.convert_to_tensor(self._epsilon) 

def padded_cross_entropy_factored(factored_logits,
                                  labels,
                                  label_smoothing,
                                  weights_fn=weights_nonzero,
                                  reduce_sum=True):
  """Memory-efficient computation of smoothing cross-entropy.

  Avoids realizing the entire logits matrix at once.

  Args:
    factored_logits: a `FactoredTensor` representing a Tensor
       with shape `[batch, timesteps, vocab_size]`.
    labels: an integer `Tensor` with shape `[batch, timesteps]`.
    label_smoothing: a floating point `Scalar`.
    weights_fn: A function from labels to weights.
    reduce_sum: a Boolean, whether to sum at the end or not.

  Returns:
    loss_numerator: a `Scalar`.  Sum of losses.
    loss_denominator: a `Scalar.  The number of non-padding target tokens.
  """
  a = factored_logits.a
  b = factored_logits.b
  confidence = 1.0 - label_smoothing
  with tf.name_scope("padded_cross_entropy_factored", values=[a, b, labels]):
    labels_flat = tf.reshape(labels, [-1])
    a_flat = tf.reshape(a, [-1, shape_list(b)[1]])
    xent = smoothing_cross_entropy_factored(a_flat, b, labels_flat,
                                            tf.convert_to_tensor(confidence))
    xent = tf.reshape(xent, shape_list(labels))
    weights = weights_fn(labels)
    if not reduce_sum:
      return xent * weights, weights
    return tf.reduce_sum(xent * weights), tf.reduce_sum(weights)