Python tensorflow.python.framework.tensor_shape.unknown_shape() Examples

def _testStackWhileSwap(self, use_gpu):
    with self.test_session(use_gpu=use_gpu):
      n = tf.constant(0)
      h = gen_data_flow_ops._stack(tf.float32, stack_name="foo")

      def c(x):
        return tf.less(x, 10)
      def b(x):
        with tf.control_dependencies([x]):
          a = tf.constant(np.ones(2000), dtype=tf.float32)
          v = gen_data_flow_ops._stack_push(h, a, swap_memory=True)
        with tf.control_dependencies([v]):
          return tf.add(x, 1)
      r = tf.while_loop(c, b, [n])

      v = tf.constant(np.zeros(2000), dtype=tf.float32)
      def c1(x, y):
        return tf.greater(x, 0)
      def b1(x, y):
        nx = tf.sub(x, 1)
        ny = y + gen_data_flow_ops._stack_pop(h, tf.float32)
        return [nx, ny]
      rx, ry = tf.while_loop(c1, b1, [r, v],
                             [r.get_shape(), tensor_shape.unknown_shape()])
      self.assertAllClose(np.ones(2000) * 10.0, ry.eval()) 

def fix_image_flip_shape(image, result):
  """Set the shape to 3 dimensional if we don't know anything else.

  Args:
    image: original image size
    result: flipped or transformed image

  Returns:
    An image whose shape is at least None,None,None.
  """

  image_shape = image.get_shape()
  if image_shape == tensor_shape.unknown_shape():
    result.set_shape([None, None, None])
  else:
    result.set_shape(image_shape)
  return result 

def testSplitShape(self):
    with self.test_session():
      ta = tensor_array_ops.TensorArray(
          dtype=tf.float32, tensor_array_name="foo",
          size=0, dynamic_size=True, infer_shape=True)
      value = tf.constant([[1.0, -1.0], [2.0, -2.0], [3.0, -3.0]])
      w0 = ta.split(value, [1, 1, 1])
      r0 = w0.read(0)
      self.assertAllEqual((1, 2), r0.get_shape())

      ta1 = tensor_array_ops.TensorArray(
          dtype=tf.float32, tensor_array_name="foo1",
          size=0, dynamic_size=True, infer_shape=True)
      w0 = ta1.split(value, [1, 2])
      r0 = w0.read(0)
      self.assertAllEqual(r0.get_shape(), tensor_shape.unknown_shape()) 

def _TileGradShape(op):
  """Shape function for the TileGrad op."""
  multiples_shape = op.inputs[1].get_shape().with_rank(1)
  input_shape = op.inputs[0].get_shape().with_rank(multiples_shape[0])
  # NOTE(mrry): Represent `multiples` as a `TensorShape` because (i)
  # it is a vector of non-negative integers, and (ii) doing so allows
  # us to handle partially-known multiples.
  multiples = tensor_util.constant_value_as_shape(op.inputs[1]).with_rank(
      input_shape.ndims)
  if multiples.ndims is None:
    return [tensor_shape.unknown_shape()]
  else:
    output_dims = []
    for dim, multiple in zip(input_shape.dims, multiples.dims):
      output_dims.append(dim // multiple)
    return [tensor_shape.TensorShape(output_dims)] 

def testWhile_5(self):
    with self.test_session():

      def compute(i, c, o):
        c = tf.slice(x, tf.expand_dims(i, 0), [1])
        o = tf.concat(0, [o, c])
        i = tf.add(i, 1)
        return [i, c, o]

      i = tf.convert_to_tensor(0)
      c = tf.convert_to_tensor([0])
      o = tf.convert_to_tensor([0])
      x = tf.convert_to_tensor([1, 2, 3, 4, 5, 6])
      s = tf.size(x)
      r = tf.while_loop(
          lambda i, c, o: tf.less(i, s), compute, [i, c, o],
          [i.get_shape(), c.get_shape(), tensor_shape.unknown_shape()])
      result = r[2].eval()
    self.assertTrue(check_op_order(i.graph))
    self.assertAllEqual(np.array([0, 1, 2, 3, 4, 5, 6]), result) 

def __init__(self, dtype, shape, accumulator_ref):
    """Creates a new ConditionalAccumulator.

    Args:
      dtype: Datatype of the accumulated gradients.
      shape: Shape of the accumulated gradients.
      accumulator_ref: A handle to the conditional accumulator, created by sub-
        classes
    """
    self._dtype = dtype
    if shape is not None:
      self._shape = tensor_shape.TensorShape(shape)
    else:
      self._shape = tensor_shape.unknown_shape()
    self._accumulator_ref = accumulator_ref
    self._name = self._accumulator_ref.op.name.split("/")[-1] 

def _DynamicPartitionShape(op):
  """Shape function for data_flow_ops.dynamic_partition."""
  data_shape = op.inputs[0].get_shape()
  partitions_shape = op.inputs[1].get_shape()
  # If we don't know the rank of partitions, we don't know anything
  mid = partitions_shape.ndims
  if mid is None:
    result_shape = tensor_shape.unknown_shape()
  else:
    # data_shape must start with partitions_shape
    partitions_shape.assert_is_compatible_with(data_shape[:mid])
    # The partition shape is dynamic in the 0th dimension, and matches
    # data_shape in the remaining dimensions.
    result_shape = tensor_shape.TensorShape([None]).concatenate(
        data_shape[mid:])
  return [result_shape] * op.get_attr("num_partitions") 

def __init__(self, op, value_index, dtype):
    """Creates a new `Tensor`.

    Args:
      op: An `Operation`. `Operation` that computes this tensor.
      value_index: An `int`. Index of the operation's endpoint that produces
        this tensor.
      dtype: A `DType`. Type of elements stored in this tensor.

    Raises:
      TypeError: If the op is not an `Operation`.
    """
    if not isinstance(op, Operation):
      raise TypeError("op needs to be an Operation: %s" % op)
    self._op = op
    self._value_index = value_index
    self._dtype = dtypes.as_dtype(dtype)
    self._shape = tensor_shape.unknown_shape()
    # List of operations that use this Tensor as input.  We maintain this list
    # to easily navigate a computation graph.
    self._consumers = []

    # Attributes used for C++ shape inference. Not inspected, only forwarded.
    self._handle_shape = tensor_shape_pb2.TensorShapeProto()
    self._handle_dtype = types_pb2.DT_INVALID 

def __init__(self, op, value_index, dtype):
    """Creates a new `Tensor`.

    Args:
      op: An `Operation`. `Operation` that computes this tensor.
      value_index: An `int`. Index of the operation's endpoint that produces
        this tensor.
      dtype: A `DType`. Type of elements stored in this tensor.

    Raises:
      TypeError: If the op is not an `Operation`.
    """
    if not isinstance(op, Operation):
      raise TypeError("op needs to be an Operation: %s" % op)
    self._op = op
    self._value_index = value_index
    self._dtype = dtypes.as_dtype(dtype)
    self._shape = tensor_shape.unknown_shape()
    # List of operations that use this Tensor as input.  We maintain this list
    # to easily navigate a computation graph.
    self._consumers = []

    # Attributes used for C++ shape inference. Not inspected, only forwarded.
    self._handle_shape = tensor_shape_pb2.TensorShapeProto()
    self._handle_dtype = types_pb2.DT_INVALID 

def testWhileFuncBasic(self):
    @function.Defun(tf.float32)
    def func(x):
      return tf.square(tf.square(x))

    with self.test_session():
      x = tf.constant(2.0, tf.float32)
      r = tf.while_loop(
          lambda i, v: i < 2,
          lambda i, v: [i + 1, func(v)],
          [tf.constant(0), x],
          [tensor_shape.unknown_shape(), tensor_shape.unknown_shape()])
      self.assertEqual(r[1].eval(), 65536.0)

      r = tf.gradients(r, x)[0]
      self.assertEqual(r.eval(), 524288.0)
      self.assertEqual(len([op for op in x.graph.get_operations()
                            if op.type == "Stack"]),
                       1) 

def fix_image_flip_shape(image, result):
  """Set the shape to 3 dimensional if we don't know anything else.

  Args:
    image: original image size
    result: flipped or transformed image

  Returns:
    An image whose shape is at least None,None,None.
  """

  image_shape = image.get_shape()
  if image_shape == tensor_shape.unknown_shape():
    result.set_shape([None, None, None])
  else:
    result.set_shape(image_shape)
  return result 

def fix_image_flip_shape(image, result):
    """Set the shape to 3 dimensional if we don't know anything else.
    Args:
      image: original image size
      result: flipped or transformed image
    Returns:
      An image whose shape is at least None,None,None.
    """
    image_shape = image.get_shape()
    if image_shape == tensor_shape.unknown_shape():
        result.set_shape([None, None, None])
    else:
        result.set_shape(image_shape)
    return result


# =========================================================================== #
# Image + BBoxes methods: cropping, resizing, flipping, ...
# =========================================================================== # 

def fix_image_flip_shape(image, result):
    """Set the shape to 3 dimensional if we don't know anything else.
    Args:
      image: original image size
      result: flipped or transformed image
    Returns:
      An image whose shape is at least None,None,None.
    """
    image_shape = image.get_shape()
    if image_shape == tensor_shape.unknown_shape():
        result.set_shape([None, None, None])
    else:
        result.set_shape(image_shape)
    return result


# =========================================================================== #
# Image + BBoxes methods: cropping, resizing, flipping, ...
# =========================================================================== # 

def _TileGradShape(op):
  """Shape function for the TileGrad op."""
  multiples_shape = op.inputs[1].get_shape().with_rank(1)
  input_shape = op.inputs[0].get_shape().with_rank(multiples_shape[0])
  # NOTE(mrry): Represent `multiples` as a `TensorShape` because (i)
  # it is a vector of non-negative integers, and (ii) doing so allows
  # us to handle partially-known multiples.
  multiples = tensor_util.constant_value_as_shape(op.inputs[1]).with_rank(
      input_shape.ndims)
  if multiples.ndims is None:
    return [tensor_shape.unknown_shape()]
  else:
    output_dims = []
    for dim, multiple in zip(input_shape.dims, multiples.dims):
      output_dims.append(dim // multiple)
    return [tensor_shape.TensorShape(output_dims)] 

def fix_image_flip_shape(image, result):
    """Set the shape to 3 dimensional if we don't know anything else.
    Args:
      image: original image size
      result: flipped or transformed image
    Returns:
      An image whose shape is at least None,None,None.
    """
    image_shape = image.get_shape()
    if image_shape == tensor_shape.unknown_shape():
        result.set_shape([None, None, None])
    else:
        result.set_shape(image_shape)
    return result


# =========================================================================== #
# Image + BBoxes methods: cropping, resizing, flipping, ...
# =========================================================================== # 

def fix_image_flip_shape(image, result):
    """Set the shape to 3 dimensional if we don't know anything else.
    Args:
      image: original image size
      result: flipped or transformed image
    Returns:
      An image whose shape is at least None,None,None.
    """
    image_shape = image.get_shape()
    if image_shape == tensor_shape.unknown_shape():
        result.set_shape([None, None, None])
    else:
        result.set_shape(image_shape)
    return result


# =========================================================================== #
# Image + BBoxes methods: cropping, resizing, flipping, ...
# =========================================================================== # 

def __init__(self, dtype, shape, accumulator_ref):
    """Creates a new ConditionalAccumulator.

    Args:
      dtype: Datatype of the accumulated gradients.
      shape: Shape of the accumulated gradients.
      accumulator_ref: A handle to the conditional accumulator, created by sub-
        classes
    """
    self._dtype = dtype
    if shape is not None:
      self._shape = tensor_shape.TensorShape(shape)
    else:
      self._shape = tensor_shape.unknown_shape()
    self._accumulator_ref = accumulator_ref
    self._name = self._accumulator_ref.op.name.split("/")[-1] 

def _TileGradShape(op):
  """Shape function for the TileGrad op."""
  multiples_shape = op.inputs[1].get_shape().with_rank(1)
  input_shape = op.inputs[0].get_shape().with_rank(multiples_shape[0])
  # NOTE(mrry): Represent `multiples` as a `TensorShape` because (i)
  # it is a vector of non-negative integers, and (ii) doing so allows
  # us to handle partially-known multiples.
  multiples = tensor_util.constant_value_as_shape(op.inputs[1]).with_rank(
      input_shape.ndims)
  if multiples.ndims is None:
    return [tensor_shape.unknown_shape()]
  else:
    output_dims = []
    for dim, multiple in zip(input_shape.dims, multiples.dims):
      output_dims.append(dim // multiple)
    return [tensor_shape.TensorShape(output_dims)] 

def __init__(self, dtypes, shapes, names, queue_ref):
    """Constructs a queue object from a queue reference.

    The two optional lists, `shapes` and `names`, must be of the same length
    as `dtypes` if provided.  The values at a given index `i` indicate the
    shape and name to use for the corresponding queue component in `dtypes`.

    Args:
      dtypes:  A list of types.  The length of dtypes must equal the number
        of tensors in each element.
      shapes: Constraints on the shapes of tensors in an element:
        A list of shape tuples or None. This list is the same length
        as dtypes.  If the shape of any tensors in the element are constrained,
        all must be; shapes can be None if the shapes should not be constrained.
      names: Optional list of names.  If provided, the `enqueue()` and
        `dequeue()` methods will use dictionaries with these names as keys.
        Must be None or a list or tuple of the same length as `dtypes`.
      queue_ref: The queue reference, i.e. the output of the queue op.

    Raises:
      ValueError: If one of the arguments is invalid.
    """
    self._dtypes = dtypes
    if shapes is not None:
      if len(shapes) != len(dtypes):
        raise ValueError("Queue shapes must have the same length as dtypes")
      self._shapes = [tensor_shape.TensorShape(s) for s in shapes]
    else:
      self._shapes = [tensor_shape.unknown_shape() for _ in self._dtypes]
    if names is not None:
      if len(names) != len(dtypes):
        raise ValueError("Queue names must have the same length as dtypes")
      self._names = names
    else:
      self._names = None
    self._queue_ref = queue_ref
    self._name = self._queue_ref.op.name.split("/")[-1] 

def _TileShape(op):
  """Shape function for the Tile op.

  This op has two inputs:

  * input: A rank-N tensor.
  * multiples: A length-N vector, in which the i^th element contains
    the factor by which `input` will be tiled in the i^th dimension.

  It has one output, which has the same rank as input, and additional
  elements according to the values in multiples

  Args:
    op: A Tile Operation.

  Returns:
    A single-element list containing the shape of the output.
  """
  multiples_shape = op.inputs[1].get_shape().with_rank(1)
  input_shape = op.inputs[0].get_shape().with_rank(multiples_shape[0].value)
  # NOTE(mrry): Represent `multiples` as a `TensorShape` because (i)
  # it is a vector of non-negative integers, and (ii) doing so allows
  # us to handle partially-known multiples.
  multiples = tensor_util.constant_value_as_shape(op.inputs[1]).with_rank(
      input_shape.ndims)
  if multiples.ndims is None:
    return [tensor_shape.unknown_shape()]
  else:
    output_dims = []
    for dim, multiple in zip(input_shape.dims, multiples.dims):
      output_dims.append(dim * multiple)
    return [tensor_shape.TensorShape(output_dims)] 

def _SliceShape(op):
  """Shape function for array_ops.slice."""
  input_shape = op.inputs[0].get_shape()
  begin_shape = op.inputs[1].get_shape().with_rank(1)
  sizes_shape = op.inputs[2].get_shape().with_rank(1)
  ndims = begin_shape.merge_with(sizes_shape)[0].value
  if ndims is not None:
    input_shape.assert_has_rank(ndims)
  # NOTE(mrry): Use `constant_value_as_shape()` to handle
  # partially-known values.
  begin_value = tensor_util.constant_value_as_shape(
      op.inputs[1]).with_rank(ndims)
  # NOTE(mrry): We can't use `constant_value_as_shape()` for `sizes`
  # because it might contain -1, which can't be represented as a
  # `TensorShape`.
  sizes_value = tensor_util.constant_value(op.inputs[2])
  if sizes_value is not None:
    returned_dims = []
    for i, (slice_size, begin_dim) in enumerate(zip(sizes_value.ravel(),
                                                    begin_value.dims)):
      if slice_size != -1:
        returned_dims.append(slice_size)
      else:
        returned_dims.append(input_shape[i] - begin_dim)
    return [tensor_shape.TensorShape(returned_dims)]
  else:
    if input_shape.ndims is not None:
      return [tensor_shape.unknown_shape(ndims=input_shape.ndims)]
    elif ndims is not None:
      return [tensor_shape.unknown_shape(ndims=ndims)]
    else:
      return [tensor_shape.unknown_shape()] 

def get_shape(self):
    return tensor_shape.unknown_shape() 

def _reverse_seq(input_seq, lengths):
  """Reverse a list of Tensors up to specified lengths.

  Args:
    input_seq: Sequence of seq_len tensors of dimension (batch_size, n_features)
               or nested tuples of tensors.
    lengths:   A `Tensor` of dimension batch_size, containing lengths for each
               sequence in the batch. If "None" is specified, simply reverses
               the list.

  Returns:
    time-reversed sequence
  """
  if lengths is None:
    return list(reversed(input_seq))

  flat_input_seq = tuple(nest.flatten(input_) for input_ in input_seq)

  flat_results = [[] for _ in range(len(input_seq))]
  for sequence in zip(*flat_input_seq):
    input_shape = tensor_shape.unknown_shape(
        ndims=sequence[0].get_shape().ndims)
    for input_ in sequence:
      input_shape.merge_with(input_.get_shape())
      input_.set_shape(input_shape)

    # Join into (time, batch_size, depth)
    s_joined = array_ops.stack(sequence)

    # Reverse along dimension 0
    s_reversed = array_ops.reverse_sequence(s_joined, lengths, 0, 1)
    # Split again into list
    result = array_ops.unstack(s_reversed)
    for r, flat_result in zip(result, flat_results):
      r.set_shape(input_shape)
      flat_result.append(r)

  results = [nest.pack_sequence_as(structure=input_, flat_sequence=flat_result)
             for input_, flat_result in zip(input_seq, flat_results)]
  return results 

def _AccumulatorShape(inputs):
  shape = tensor_shape.unknown_shape()
  for i in inputs:
    if isinstance(i, ops.Tensor):
      shape = shape.merge_with(i.get_shape())
  return shape 

def _DynamicStitchShape(op):
  """Shape function for data_flow_ops.dynamic_stitch."""
  num_partitions = op.get_attr("N")
  indices_shapes = [t.get_shape() for t in op.inputs[0:num_partitions]]
  data_shapes = [t.get_shape() for t in op.inputs[num_partitions:]]
  output_shape = tensor_shape.unknown_shape()
  extra_shape = tensor_shape.TensorShape(None)
  for indices_shape, data_shape in zip(indices_shapes, data_shapes):
    indices_ndims = indices_shape.ndims
    if indices_ndims is not None:
      # Assert that data_shape starts with indices_shape
      indices_shape.merge_with(data_shape[:indices_ndims])
      # The rest belongs to output
      extra_shape = extra_shape.merge_with(data_shape[indices_ndims:])
  return [tensor_shape.TensorShape([None]).concatenate(extra_shape)] 

def testAsProto(self):
    self.assertTrue(tensor_shape.unknown_shape().as_proto().unknown_rank)
    self.assertFalse(
        tensor_shape.unknown_shape(ndims=3).as_proto().unknown_rank)
    self.assertFalse(
        tensor_shape.TensorShape([1, 2, 3]).as_proto().unknown_rank)
    self.assertFalse(
        tensor_shape.TensorShape([1, None, 3]).as_proto().unknown_rank) 

def _shape_common(s1, s2):
  """The greatest lower bound (ordered by specificity) TensorShape."""
  s1 = tensor_shape.TensorShape(s1)
  s2 = tensor_shape.TensorShape(s2)
  if s1.ndims is None or s2.ndims is None or s1.ndims != s2.ndims:
    return tensor_shape.unknown_shape()
  d = [
      d1 if d1 is not None and d1 == d2 else None
      for (d1, d2) in zip(s1.as_list(), s2.as_list())]
  return tensor_shape.TensorShape(d)


# pylint: disable=protected-access 

def variable_op(shape, dtype, name="Variable", set_shape=True, container="",
                shared_name=""):
  """Create a variable Operation.

  See also variables.Variable.

  Args:
    shape: The shape of the tensor managed by this variable
    dtype: The underlying type of the tensor values.
    name: optional name to use for the variable op.
    set_shape: If True, set the shape property of the returned Tensor to
      the shape argument.
    container: An optional string. Defaults to "".
      If non-empty, this variable is placed in the given container.
      Otherwise, a default container is used.
    shared_name: An optional string. Defaults to "".
      If non-empty, this variable is named in the given bucket
      with this shared_name. Otherwise, the node name is used instead.

  Returns:
    A variable tensor.
  """
  if not set_shape:
    shape = tensor_shape.unknown_shape()
  ret = gen_state_ops._variable(shape=shape, dtype=dtype, name=name,
                                container=container, shared_name=shared_name)
  # TODO(mrry): Move this to where it is used, so we can get rid of this op
  #   wrapper?
  if set_shape:
    ret.set_shape(shape)
  return ret


# NOTE(mrry): Shapes are conditionally set in the Python wrapper. 

def testAssignNoShapeNoValidateShape(self):
    with self.test_session():
      value = self._NewShapelessTensor()
      var = state_ops.variable_op([1, 2], tf.float32, set_shape=False)
      self.assertEqual(tensor_shape.unknown_shape(), var.get_shape())
      self.assertEqual(tensor_shape.unknown_shape(),
                       tf.assign(var, value, validate_shape=False).get_shape()) 

def testAssignNoShape(self):
    with self.test_session():
      value = self._NewShapelessTensor()
      var = state_ops.variable_op([1, 2], tf.float32, set_shape=False)
      self.assertEqual(tensor_shape.unknown_shape(), var.get_shape())
      self.assertEqual(tensor_shape.unknown_shape(),
                       tf.assign(var, value).get_shape())