Python tensorflow.sparse_to_dense() Examples

def one_hot_encoding(labels, num_classes, scope=None):
  """Transform numeric labels into onehot_labels.

  Args:
    labels: [batch_size] target labels.
    num_classes: total number of classes.
    scope: Optional scope for name_scope.
  Returns:
    one hot encoding of the labels.
  """
  with tf.name_scope(scope, 'OneHotEncoding', [labels]):
    batch_size = labels.get_shape()[0]
    indices = tf.expand_dims(tf.range(0, batch_size), 1)
    labels = tf.cast(tf.expand_dims(labels, 1), indices.dtype)
    concated = tf.concat(axis=1, values=[indices, labels])
    onehot_labels = tf.sparse_to_dense(
        concated, tf.stack([batch_size, num_classes]), 1.0, 0.0)
    onehot_labels.set_shape([batch_size, num_classes])
    return onehot_labels 

def one_hot_encoding(labels, num_classes, scope=None):
  """Transform numeric labels into onehot_labels.

  Args:
    labels: [batch_size] target labels.
    num_classes: total number of classes.
    scope: Optional scope for name_scope.
  Returns:
    one hot encoding of the labels.
  """
  with tf.name_scope(scope, 'OneHotEncoding', [labels]):
    batch_size = labels.get_shape()[0]
    indices = tf.expand_dims(tf.range(0, batch_size), 1)
    labels = tf.cast(tf.expand_dims(labels, 1), indices.dtype)
    concated = tf.concat([indices, labels], 1)
    onehot_labels = tf.sparse_to_dense(
        concated, tf.pack([batch_size, num_classes]), 1.0, 0.0)
    onehot_labels.set_shape([batch_size, num_classes])
    return onehot_labels 

def nms_return_masks(self, X):
    config = self.config
    prob, box = X # [K], [K,4]
    output_shape = tf.shape(prob)
    # [K]
    ids = tf.reshape(tf.where(prob > config.result_score_thres), [-1])
    prob_ = tf.gather(prob, ids)
    box_ = tf.gather(box, ids)
    # NMS
    selection = tf.image.non_max_suppression(
        box_, prob_, max_output_size=config.result_per_im,
        iou_threshold=config.fastrcnn_nms_iou_thres)
    selection = tf.to_int32(tf.gather(ids, selection))
    sorted_selection = -tf.nn.top_k(-selection, k=tf.size(selection))[0]

    mask = tf.sparse_to_dense(
        sparse_indices=sorted_selection,
        output_shape=output_shape,
        sparse_values=True,
        default_value=False)

    return mask 

def one_hot_encoding(labels, num_classes, scope=None):
  """Transform numeric labels into onehot_labels.

  Args:
    labels: [batch_size] target labels.
    num_classes: total number of classes.
    scope: Optional scope for name_scope.
  Returns:
    one hot encoding of the labels.
  """
  with tf.name_scope(scope, 'OneHotEncoding', [labels]):
    batch_size = labels.get_shape()[0]
    indices = tf.expand_dims(tf.range(0, batch_size), 1)
    labels = tf.cast(tf.expand_dims(labels, 1), indices.dtype)
    concated = tf.concat(axis=1, values=[indices, labels])
    onehot_labels = tf.sparse_to_dense(
        concated, tf.stack([batch_size, num_classes]), 1.0, 0.0)
    onehot_labels.set_shape([batch_size, num_classes])
    return onehot_labels 

def parse_example_batch(serialized):
  """Parses a batch of tf.Example protos.

  Args:
    serialized: A 1-D string Tensor; a batch of serialized tf.Example protos.
  Returns:
    encode: A SentenceBatch of encode sentences.
    decode_pre: A SentenceBatch of "previous" sentences to decode.
    decode_post: A SentenceBatch of "post" sentences to decode.
  """
  features = tf.parse_example(
    serialized,
    features={"features": tf.VarLenFeature(dtype=tf.int64)}
  )
  features = features["features"]

  def _sparse_to_batch(sparse):
    ids = tf.sparse_tensor_to_dense(sparse)  # Padding with zeroes.
    mask = tf.sparse_to_dense(sparse.indices, sparse.dense_shape,
                              tf.ones_like(sparse.values, dtype=tf.int32))
    return SentenceBatch(ids=ids, mask=mask)

  return _sparse_to_batch(features) 

def one_hot_encoding(labels, num_classes, scope=None):
  """Transform numeric labels into onehot_labels.

  Args:
    labels: [batch_size] target labels.
    num_classes: total number of classes.
    scope: Optional scope for name_scope.
  Returns:
    one hot encoding of the labels.
  """
  with tf.name_scope(scope, 'OneHotEncoding', [labels]):
    batch_size = labels.get_shape()[0]
    indices = tf.expand_dims(tf.range(0, batch_size), 1)
    labels = tf.cast(tf.expand_dims(labels, 1), indices.dtype)
    concated = tf.concat([indices, labels], 1)
    onehot_labels = tf.sparse_to_dense(
        concated, tf.pack([batch_size, num_classes]), 1.0, 0.0)
    onehot_labels.set_shape([batch_size, num_classes])
    return onehot_labels 

def kSparse(self, x, topk):
        print 'run regular k-sparse'
        dim = int(x.get_shape()[1])
        if topk > dim:
            warnings.warn('Warning: topk should not be larger than dim: %s, found: %s, using %s' % (dim, topk, dim))
            topk = dim

        k = dim - topk
        values, indices = tf.nn.top_k(-x, k) # indices will be [[0, 1], [2, 1]], values will be [[6., 2.], [5., 4.]]

        # We need to create full indices like [[0, 0], [0, 1], [1, 2], [1, 1]]
        my_range = tf.expand_dims(tf.range(0, tf.shape(indices)[0]), 1)  # will be [[0], [1]]
        my_range_repeated = tf.tile(my_range, [1, k])  # will be [[0, 0], [1, 1]]

        full_indices = tf.stack([my_range_repeated, indices], axis=2) # change shapes to [N, k, 1] and [N, k, 1], to concatenate into [N, k, 2]
        full_indices = tf.reshape(full_indices, [-1, 2])

        to_reset = tf.sparse_to_dense(full_indices, tf.shape(x), tf.reshape(values, [-1]), default_value=0., validate_indices=False)

        res = tf.add(x, to_reset)

        return res 

def one_hot_encoding(labels, num_classes, scope=None):
  """Transform numeric labels into onehot_labels.

  Args:
    labels: [batch_size] target labels.
    num_classes: total number of classes.
    scope: Optional scope for op_scope.
  Returns:
    one hot encoding of the labels.
  """
  with tf.op_scope([labels], scope, 'OneHotEncoding'):
    batch_size = labels.get_shape()[0]
    indices = tf.expand_dims(tf.range(0, batch_size), 1)
    labels = tf.cast(tf.expand_dims(labels, 1), indices.dtype)
    concated = tf.concat(1, [indices, labels])
    onehot_labels = tf.sparse_to_dense(
        concated, tf.pack([batch_size, num_classes]), 1.0, 0.0)
    onehot_labels.set_shape([batch_size, num_classes])
    return onehot_labels 

def one_hot_encoding(labels, num_classes, scope=None):
  """Transform numeric labels into onehot_labels.

  Args:
    labels: [batch_size] target labels.
    num_classes: total number of classes.
    scope: Optional scope for op_scope.
  Returns:
    one hot encoding of the labels.
  """
  with tf.op_scope([labels], scope, 'OneHotEncoding'):
    batch_size = labels.get_shape()[0]
    indices = tf.expand_dims(tf.range(0, batch_size), 1)
    labels = tf.cast(tf.expand_dims(labels, 1), indices.dtype)
    concated = tf.concat(1, [indices, labels])
    onehot_labels = tf.sparse_to_dense(
        concated, tf.pack([batch_size, num_classes]), 1.0, 0.0)
    onehot_labels.set_shape([batch_size, num_classes])
    return onehot_labels 

def one_hot_encoding(labels, num_classes, scope=None):
  """Transform numeric labels into onehot_labels.

  Args:
    labels: [batch_size] target labels.
    num_classes: total number of classes.
    scope: Optional scope for name_scope.
  Returns:
    one hot encoding of the labels.
  """
  with tf.name_scope(scope, 'OneHotEncoding', [labels]):
    batch_size = labels.get_shape()[0]
    indices = tf.expand_dims(tf.range(0, batch_size), 1)
    labels = tf.cast(tf.expand_dims(labels, 1), indices.dtype)
    concated = tf.concat([indices, labels], 1)
    onehot_labels = tf.sparse_to_dense(
        concated, tf.pack([batch_size, num_classes]), 1.0, 0.0)
    onehot_labels.set_shape([batch_size, num_classes])
    return onehot_labels 

def cartesian_product(a, b, axis):
    a_rank = tf.rank(a)
    a_dim = tf.shape(a)[axis]    
    b_rank = tf.rank(b)
    b_dim = tf.shape(b)[axis]
    
    axis_a_repeat = tf.sparse_to_dense(sparse_indices=[axis+1], sparse_values=[b_dim], output_shape=[a_rank+1], default_value=1)
    tile_a = tf.tile(tf.expand_dims(a, axis+1), axis_a_repeat)
    
    axis_b_repeat = tf.sparse_to_dense(sparse_indices=[axis], sparse_values=[a_dim], output_shape=[b_rank+1], default_value=1)
    tile_b = tf.tile(tf.expand_dims(b, axis), axis_b_repeat)
    
    cart_prod = tf.concat([tile_a, tile_b], axis=-1)

    #Defining the last dimension of resulting tensor (originally undefined)
    last_dim = int(a.get_shape()[-1]) + int(b.get_shape()[-1])
    cart_prod_shape = list(cart_prod.get_shape())
    cart_prod_shape[-1] = last_dim
    cart_prod.set_shape(cart_prod_shape)    
    
    return cart_prod 

def one_hot_encoding(labels, num_classes, scope=None):
  """Transform numeric labels into onehot_labels.

  Args:
    labels: [batch_size] target labels.
    num_classes: total number of classes.
    scope: Optional scope for name_scope.
  Returns:
    one hot encoding of the labels.
  """
  with tf.name_scope(scope, 'OneHotEncoding', [labels]):
    batch_size = labels.get_shape()[0]
    indices = tf.expand_dims(tf.range(0, batch_size), 1)
    labels = tf.cast(tf.expand_dims(labels, 1), indices.dtype)
    concated = tf.concat(axis=1, values=[indices, labels])
    onehot_labels = tf.sparse_to_dense(
        concated, tf.stack([batch_size, num_classes]), 1.0, 0.0)
    onehot_labels.set_shape([batch_size, num_classes])
    return onehot_labels 

def indices_to_dense_vector(indices,
                            size,
                            indices_value=1.,
                            default_value=0,
                            dtype=tf.float32):
  """Creates dense vector with indices set to specific value and rest to zeros.

  This function exists because it is unclear if it is safe to use
    tf.sparse_to_dense(indices, [size], 1, validate_indices=False)
  with indices which are not ordered.
  This function accepts a dynamic size (e.g. tf.shape(tensor)[0])

  Args:
    indices: 1d Tensor with integer indices which are to be set to
        indices_values.
    size: scalar with size (integer) of output Tensor.
    indices_value: values of elements specified by indices in the output vector
    default_value: values of other elements in the output vector.
    dtype: data type.

  Returns:
    dense 1D Tensor of shape [size] with indices set to indices_values and the
        rest set to default_value.
  """
  size = tf.to_int32(size)
  zeros = tf.ones([size], dtype=dtype) * default_value
  values = tf.ones_like(indices, dtype=dtype) * indices_value

  return tf.dynamic_stitch([tf.range(size), tf.to_int32(indices)],
                           [zeros, values]) 

def findBoundaries(planes, planeMasks):
    height = int(planeMasks.shape[0])
    width = int(planeMasks.shape[1])
    
    planesD = tf.norm(planes, axis=1, keep_dims=True)
    planesD = tf.clip_by_value(planesD, 1e-5, 10)
    planesNormal = planes / planesD

    ND = tf.expand_dims(planesNormal, 0) * tf.expand_dims(planesD, 1)
    ND_diff = tf.reshape(ND - tf.transpose(ND, [1, 0, 2]), [-1, 3])
    coefX, coefY, coefZ = tf.unstack(ND_diff, axis=1)

    pixels = []
    focalLength = 517.97
    urange = tf.range(width, dtype=tf.float32) / focalLength
    ones = tf.ones(urange.shape)
    vs = (coefX * urange + coefY * ones) / coefZ
    pixels.append(tf.stack([tf.floor(vs), urange], axis=1))
    pixels.append(tf.stack([tf.ceil(vs), urange], axis=1))
    
    vrange = tf.range(height, dtype=tf.float32) / focalLength
    ones = tf.ones(vrange.shape)
    us = -(coefY * ones - coefZ * vrange) / coefX
    pixels.append(tf.stack([vrange, tf.floor(us)], axis=1))
    pixels.append(tf.stack([vrange, tf.ceil(us)], axis=1))

    v, u = tf.unstack(pixels, axis=1)
    validMask = tf.logical_and(tf.less(u, width), tf.less(v, height))
    validMask = tf.logical_and(validMask, tf.greater_equal(u, 0))
    validMask = tf.logical_and(validMask, tf.greater_equal(v, 0))
    
    pixels *= tf.expand_dims(invalidMask, -1)
    
    boundary = tf.sparse_to_dense(pixels, output_shape=[height, width], sparse_values=1)
    return boundary 

def get_config(self):
        config = {'topk': self.topk, 'ctype': self.ctype}
        base_config = super(KCompetitive, self).get_config()
        return dict(list(base_config.items()) + list(config.items()))

    # def k_comp_sigm(self, x, topk):
    #     print 'run k_comp_sigm'
    #     dim = int(x.get_shape()[1])
    #     if topk > dim:
    #         warnings.warn('topk should not be larger than dim: %s, found: %s, using %s' % (dim, topk, dim))
    #         topk = dim

    #     values, indices = tf.nn.top_k(x, topk) # indices will be [[0, 1], [2, 1]], values will be [[6., 2.], [5., 4.]]

    #     # We need to create full indices like [[0, 0], [0, 1], [1, 2], [1, 1]]
    #     my_range = tf.expand_dims(tf.range(0, K.shape(indices)[0]), 1)  # will be [[0], [1]]
    #     my_range_repeated = tf.tile(my_range, [1, topk])  # will be [[0, 0], [1, 1]]

    #     full_indices = tf.stack([my_range_repeated, indices], axis=2) # change shapes to [N, k, 1] and [N, k, 1], to concatenate into [N, k, 2]
    #     full_indices = tf.reshape(full_indices, [-1, 2])

    #     to_reset = tf.sparse_to_dense(full_indices, tf.shape(x), tf.reshape(values, [-1]), default_value=0., validate_indices=False)

    #     batch_size = tf.to_float(tf.shape(x)[0])
    #     tmp = 1 * batch_size * tf.reduce_sum(x - to_reset, 1, keep_dims=True) / topk

    #     res = tf.sparse_to_dense(full_indices, tf.shape(x), tf.reshape(tf.add(values, tmp), [-1]), default_value=0., validate_indices=False)

    #     return res 

def _count_matrix_input(self, filenames, submatrix_rows, submatrix_cols):
    """Creates ops that read submatrix shards from disk."""
    random.shuffle(filenames)
    filename_queue = tf.train.string_input_producer(filenames)
    reader = tf.WholeFileReader()
    _, serialized_example = reader.read(filename_queue)
    features = tf.parse_single_example(
        serialized_example,
        features={
            'global_row': tf.FixedLenFeature([submatrix_rows], dtype=tf.int64),
            'global_col': tf.FixedLenFeature([submatrix_cols], dtype=tf.int64),
            'sparse_local_row': tf.VarLenFeature(dtype=tf.int64),
            'sparse_local_col': tf.VarLenFeature(dtype=tf.int64),
            'sparse_value': tf.VarLenFeature(dtype=tf.float32)
        })

    global_row = features['global_row']
    global_col = features['global_col']

    sparse_local_row = features['sparse_local_row'].values
    sparse_local_col = features['sparse_local_col'].values
    sparse_count = features['sparse_value'].values

    sparse_indices = tf.concat(
        axis=1, values=[tf.expand_dims(sparse_local_row, 1),
                        tf.expand_dims(sparse_local_col, 1)])

    count = tf.sparse_to_dense(sparse_indices, [submatrix_rows, submatrix_cols],
                               sparse_count)

    return global_row, global_col, count 

def indices_to_dense_vector(indices,
                            size,
                            indices_value=1.,
                            default_value=0,
                            dtype=tf.float32):
  """Creates dense vector with indices set to specific value and rest to zeros.

  This function exists because it is unclear if it is safe to use
    tf.sparse_to_dense(indices, [size], 1, validate_indices=False)
  with indices which are not ordered.
  This function accepts a dynamic size (e.g. tf.shape(tensor)[0])

  Args:
    indices: 1d Tensor with integer indices which are to be set to
        indices_values.
    size: scalar with size (integer) of output Tensor.
    indices_value: values of elements specified by indices in the output vector
    default_value: values of other elements in the output vector.
    dtype: data type.

  Returns:
    dense 1D Tensor of shape [size] with indices set to indices_values and the
        rest set to default_value.
  """
  size = tf.to_int32(size)
  zeros = tf.ones([size], dtype=dtype) * default_value
  values = tf.ones_like(indices, dtype=dtype) * indices_value

  return tf.dynamic_stitch([tf.range(size), tf.to_int32(indices)],
                           [zeros, values]) 

def indices_to_dense_vector(indices,
                            size,
                            indices_value=1.,
                            default_value=0,
                            dtype=tf.float32):
  """Creates dense vector with indices set to specific value and rest to zeros.

  This function exists because it is unclear if it is safe to use
    tf.sparse_to_dense(indices, [size], 1, validate_indices=False)
  with indices which are not ordered.
  This function accepts a dynamic size (e.g. tf.shape(tensor)[0])

  Args:
    indices: 1d Tensor with integer indices which are to be set to
        indices_values.
    size: scalar with size (integer) of output Tensor.
    indices_value: values of elements specified by indices in the output vector
    default_value: values of other elements in the output vector.
    dtype: data type.

  Returns:
    dense 1D Tensor of shape [size] with indices set to indices_values and the
        rest set to default_value.
  """
  size = tf.to_int32(size)
  zeros = tf.ones([size], dtype=dtype) * default_value
  values = tf.ones_like(indices, dtype=dtype) * indices_value

  return tf.dynamic_stitch([tf.range(size), tf.to_int32(indices)],
                           [zeros, values]) 

def loss(logits, labels, batch_size=None):
  """Adds all losses for the model.

  Note the final loss is not returned. Instead, the list of losses are collected
  by slim.losses. The losses are accumulated in tower_loss() and summed to
  calculate the total loss.

  Args:
    logits: List of logits from inference(). Each entry is a 2-D float Tensor.
    labels: Labels from distorted_inputs or inputs(). 1-D tensor
            of shape [batch_size]
    batch_size: integer
  """
  if not batch_size:
    batch_size = FLAGS.batch_size

  # Reshape the labels into a dense Tensor of
  # shape [FLAGS.batch_size, num_classes].
  sparse_labels = tf.reshape(labels, [batch_size, 1])
  indices = tf.reshape(tf.range(batch_size), [batch_size, 1])
  concated = tf.concat(1, [indices, sparse_labels])
  num_classes = logits[0].get_shape()[-1].value
  dense_labels = tf.sparse_to_dense(concated,
                                    [batch_size, num_classes],
                                    1.0, 0.0)

  # Cross entropy loss for the main softmax prediction.
  slim.losses.cross_entropy_loss(logits[0],
                                 dense_labels,
                                 label_smoothing=0.1,
                                 weight=1.0)

  # Cross entropy loss for the auxiliary softmax head.
  slim.losses.cross_entropy_loss(logits[1],
                                 dense_labels,
                                 label_smoothing=0.1,
                                 weight=0.4,
                                 scope='aux_loss') 

def loss(logits, labels):
  """Add L2Loss to all the trainable variables.

  Add summary for for "Loss" and "Loss/avg".

  Args:
    logits: Logits from inference().
    labels: Labels from distorted_inputs or inputs(). 1-D tensor
            of shape [batch_size]

  Returns:
    Loss tensor of type float.
  """
  # Reshape the labels into a dense Tensor of
  # shape [batch_size, NUM_CLASSES].
  sparse_labels = tf.reshape(labels, [FLAGS.batch_size, 1])
  indices = tf.reshape(tf.range(0, FLAGS.batch_size), [FLAGS.batch_size, 1])
  concated = tf.concat(axis=1, values=[indices, sparse_labels])
  dense_labels = tf.sparse_to_dense(concated,
                                    [FLAGS.batch_size, NUM_CLASSES],
                                    1.0, 0.0)
  
  # Calculate the average cross entropy loss across the batch.
  cross_entropy = tf.nn.softmax_cross_entropy_with_logits(
    logits=logits, labels=dense_labels, name='cross_entropy_per_example')
  cross_entropy_mean = tf.reduce_mean(cross_entropy, name='cross_entropy')
  tf.add_to_collection('losses', cross_entropy_mean)

  # The total loss is defined as the cross entropy loss plus all of the weight
  # decay terms (L2 loss).
  return tf.add_n(tf.get_collection('losses'), name='total_loss') 

def indices_to_dense_vector(indices,
                            size,
                            indices_value=1.,
                            default_value=0,
                            dtype=tf.float32):
  """Creates dense vector with indices set to specific value and rest to zeros.

  This function exists because it is unclear if it is safe to use
    tf.sparse_to_dense(indices, [size], 1, validate_indices=False)
  with indices which are not ordered.
  This function accepts a dynamic size (e.g. tf.shape(tensor)[0])

  Args:
    indices: 1d Tensor with integer indices which are to be set to
        indices_values.
    size: scalar with size (integer) of output Tensor.
    indices_value: values of elements specified by indices in the output vector
    default_value: values of other elements in the output vector.
    dtype: data type.

  Returns:
    dense 1D Tensor of shape [size] with indices set to indices_values and the
        rest set to default_value.
  """
  size = tf.to_int32(size)
  zeros = tf.ones([size], dtype=dtype) * default_value
  values = tf.ones_like(indices, dtype=dtype) * indices_value

  return tf.dynamic_stitch([tf.range(size), tf.to_int32(indices)],
                           [zeros, values]) 

def pad_tensor_to_batch_size(tensor, batch_size):
  """Pads a Tensor along the batch dimension to the desired batch size."""
  if batch_size < 2:
    raise ValueError("Cannot pad along batch dimension with batch_size < 2.")

  ndims = len(tensor.shape)
  if ndims < 1:
    raise ValueError("Cannot pad a 0-dimensional Tensor")

  num_pad_examples = batch_size - tf.shape(tensor)[0]

  # paddings is a 2D Tensor with shape [ndims, 2]. Every element is zero except
  # for paddings[0][1], which is the number of values to add along the 0-th
  # dimension (the batch dimension) after the contents of the input tensor.
  paddings = tf.sparse_to_dense(
      sparse_indices=[[0, 1]],
      output_shape=[ndims, 2],
      sparse_values=num_pad_examples)

  padded_tensor = tf.pad(tensor, paddings, name=tensor.op.name + "/pad")

  # Set the new shape.
  output_shape = tensor.shape.as_list()
  output_shape[0] = batch_size
  padded_tensor.set_shape(output_shape)

  return padded_tensor 

def parse_example_batch(serialized):
  """Parses a batch of tf.Example protos.

  Args:
    serialized: A 1-D string Tensor; a batch of serialized tf.Example protos.
  Returns:
    encode: A SentenceBatch of encode sentences.
    decode_pre: A SentenceBatch of "previous" sentences to decode.
    decode_post: A SentenceBatch of "post" sentences to decode.
  """
  features = tf.parse_example(
      serialized,
      features={
          "encode": tf.VarLenFeature(dtype=tf.int64),
          "decode_pre": tf.VarLenFeature(dtype=tf.int64),
          "decode_post": tf.VarLenFeature(dtype=tf.int64),
      })

  def _sparse_to_batch(sparse):
    ids = tf.sparse_tensor_to_dense(sparse)  # Padding with zeroes.
    mask = tf.sparse_to_dense(sparse.indices, sparse.dense_shape,
                              tf.ones_like(sparse.values, dtype=tf.int32))
    return SentenceBatch(ids=ids, mask=mask)

  output_names = ("encode", "decode_pre", "decode_post")
  return tuple(_sparse_to_batch(features[x]) for x in output_names) 

def AddCrossEntropy(batch_size, n):
  """Adds a cross entropy cost function."""
  cross_entropies = []
  def _Pass():
    return tf.constant(0, dtype=tf.float32, shape=[1])

  for beam_id in range(batch_size):
    beam_gold_slot = tf.reshape(
        tf.strided_slice(n['gold_slot'], [beam_id], [beam_id + 1]), [1])
    def _ComputeCrossEntropy():
      """Adds ops to compute cross entropy of the gold path in a beam."""
      # Requires a cast so that UnsortedSegmentSum, in the gradient,
      # is happy with the type of its input 'segment_ids', which
      # must be int32.
      idx = tf.cast(
          tf.reshape(
              tf.where(tf.equal(n['beam_ids'], beam_id)), [-1]), tf.int32)
      beam_scores = tf.reshape(tf.gather(n['all_path_scores'], idx), [1, -1])
      num = tf.shape(idx)
      return tf.nn.softmax_cross_entropy_with_logits(
          labels=tf.expand_dims(
              tf.sparse_to_dense(beam_gold_slot, num, [1.], 0.), 0),
          logits=beam_scores)
    # The conditional here is needed to deal with the last few batches of the
    # corpus which can contain -1 in beam_gold_slot for empty batch slots.
    cross_entropies.append(cf.cond(
        beam_gold_slot[0] >= 0, _ComputeCrossEntropy, _Pass))
  return {'cross_entropy': tf.div(tf.add_n(cross_entropies), batch_size)} 

def _count_matrix_input(self, filenames, submatrix_rows, submatrix_cols):
    """Creates ops that read submatrix shards from disk."""
    random.shuffle(filenames)
    filename_queue = tf.train.string_input_producer(filenames)
    reader = tf.WholeFileReader()
    _, serialized_example = reader.read(filename_queue)
    features = tf.parse_single_example(
        serialized_example,
        features={
            'global_row': tf.FixedLenFeature([submatrix_rows], dtype=tf.int64),
            'global_col': tf.FixedLenFeature([submatrix_cols], dtype=tf.int64),
            'sparse_local_row': tf.VarLenFeature(dtype=tf.int64),
            'sparse_local_col': tf.VarLenFeature(dtype=tf.int64),
            'sparse_value': tf.VarLenFeature(dtype=tf.float32)
        })

    global_row = features['global_row']
    global_col = features['global_col']

    sparse_local_row = features['sparse_local_row'].values
    sparse_local_col = features['sparse_local_col'].values
    sparse_count = features['sparse_value'].values

    sparse_indices = tf.concat(
        axis=1, values=[tf.expand_dims(sparse_local_row, 1),
                        tf.expand_dims(sparse_local_col, 1)])

    count = tf.sparse_to_dense(sparse_indices, [submatrix_rows, submatrix_cols],
                               sparse_count)

    return global_row, global_col, count 

def as_one_hot(input_, n_indices):
    """Convert indices to one-hot."""
    shape = input_.get_shape().as_list()
    n_elem = numpy.prod(shape)
    indices = tf.range(n_elem)
    indices = tf.cast(indices, tf.int64)
    indices_input = tf.concat(axis=0, values=[indices, tf.reshape(input_, [-1])])
    indices_input = tf.reshape(indices_input, [2, -1])
    indices_input = tf.transpose(indices_input)
    res = tf.sparse_to_dense(
        indices_input, [n_elem, n_indices], 1., 0., name="flat_one_hot")
    res = tf.reshape(res, [elem for elem in shape] + [n_indices])

    return res 

def parse_example_batch(serialized):
  """Parses a batch of tf.Example protos.

  Args:
    serialized: A 1-D string Tensor; a batch of serialized tf.Example protos.
  Returns:
    encode: A SentenceBatch of encode sentences.
    decode_pre: A SentenceBatch of "previous" sentences to decode.
    decode_post: A SentenceBatch of "post" sentences to decode.
  """
  features = tf.parse_example(
      serialized,
      features={
          "encode": tf.VarLenFeature(dtype=tf.int64),
          "decode_pre": tf.VarLenFeature(dtype=tf.int64),
          "decode_post": tf.VarLenFeature(dtype=tf.int64),
      })

  def _sparse_to_batch(sparse):
    ids = tf.sparse_tensor_to_dense(sparse)  # Padding with zeroes.
    mask = tf.sparse_to_dense(sparse.indices, sparse.dense_shape,
                              tf.ones_like(sparse.values, dtype=tf.int32))
    return SentenceBatch(ids=ids, mask=mask)

  output_names = ("encode", "decode_pre", "decode_post")
  return tuple(_sparse_to_batch(features[x]) for x in output_names) 

def labels_to_dense(labels):
  # Reshape the labels into a dense Tensor of
  # shape [batch_size, NUM_CLASSES].
  sparse_labels = tf.reshape(labels, [FLAGS.batch_size, 1])
  indices = tf.reshape(tf.range(FLAGS.batch_size), [FLAGS.batch_size, 1])
  concated = tf.concat(1, [indices, sparse_labels])
  dense_labels = tf.sparse_to_dense(concated,
                                    [FLAGS.batch_size, NUM_CLASSES],
                                    1.0, 0.0)
  return dense_labels 

def indices_to_dense_vector(indices,
                            size,
                            indices_value=1.,
                            default_value=0,
                            dtype=tf.float32):
  """Creates dense vector with indices set to specific (the para "indices_value" ) and rest to zeros.

  This function exists because it is unclear if it is safe to use
    tf.sparse_to_dense(indices, [size], 1, validate_indices=False)
  with indices which are not ordered.
  This function accepts a dynamic size (e.g. tf.shape(tensor)[0])

  Args:
    indices: 1d Tensor with integer indices which are to be set to
        indices_values.
    size: scalar with size (integer) of output Tensor.
    indices_value: values of elements specified by indices in the output vector
    default_value: values of other elements in the output vector.
    dtype: data type.

  Returns:
    dense 1D Tensor of shape [size] with indices set to indices_values and the
        rest set to default_value.
  """
  size = tf.to_int32(size)
  zeros = tf.ones([size], dtype=dtype) * default_value
  values = tf.ones_like(indices, dtype=dtype) * indices_value

  return tf.dynamic_stitch([tf.range(size), tf.to_int32(indices)],
                           [zeros, values]) 

def _SparseToDense(sparse_indices, output_size, sparse_values,
                   default_value, validate_indices=True):
  return tf.sparse_to_dense(sparse_indices, output_size,
                            sparse_values,
                            default_value=default_value,
                            validate_indices=validate_indices)