def testTensorArrayUnpackWrongMajorSizeFails(self):
with self.test_session():
h = gen_data_flow_ops._tensor_array(
dtype=tf.float32, tensor_array_name="foo", size=3)
with self.assertRaisesOpError(
r"Input value must have first dimension "
r"equal to the array size \(2 vs. 3\)"):
gen_data_flow_ops._tensor_array_unpack(h, [1.0, 2.0]).run()
def unpack(self, value, name=None):
"""Pack the values of a `Tensor` in the TensorArray.
Args:
value: (N+1)-D. Tensor of type `dtype`. The Tensor to unpack.
name: A name for the operation (optional).
Returns:
A new TensorArray object with flow that ensures the unpack occurs.
Use this object all for subsequent operations.
Raises:
ValueError: if the shape inference fails.
"""
with ops.colocate_with(self._handle):
flow_out = gen_data_flow_ops._tensor_array_unpack(
handle=self._handle, value=value, flow_in=self._flow,
name=name)
ta = TensorArray(dtype=self._dtype, handle=self._handle)
ta._flow = flow_out
ta._infer_shape = self._infer_shape
ta._elem_shape = self._elem_shape
if ta._infer_shape:
val_shape = flow_out.op.inputs[1].get_shape()
elem_shape = tensor_shape.unknown_shape()
if val_shape.dims is not None:
elem_shape = tensor_shape.TensorShape(val_shape.dims[1:])
if ta._elem_shape:
if not elem_shape == ta._elem_shape[0]:
raise ValueError(
"Inconsistent shapes: saw %s but expected %s "
"(and infer_shape=True)" % (elem_shape, ta._elem_shape[0]))
else:
ta._elem_shape.append(elem_shape)
return ta
def unpack(self, value, name=None):
"""Pack the values of a `Tensor` in the TensorArray.
Args:
value: (N+1)-D. Tensor of type `dtype`. The Tensor to unpack.
name: A name for the operation (optional).
Returns:
A new TensorArray object with flow that ensures the unpack occurs.
Use this object all for subsequent operations.
Raises:
ValueError: if the shape inference fails.
"""
with ops.colocate_with(self._handle):
flow_out = gen_data_flow_ops._tensor_array_unpack(
handle=self._handle,
value=value,
flow_in=self._flow,
name=name)
ta = TensorArray(dtype=self._dtype, handle=self._handle)
ta._flow = flow_out
ta._infer_shape = self._infer_shape
ta._elem_shape = self._elem_shape
if ta._infer_shape:
val_shape = flow_out.op.inputs[1].get_shape()
elem_shape = tensor_shape.unknown_shape()
if val_shape.dims:
elem_shape = tensor_shape.TensorShape(val_shape.dims[1:])
if ta._elem_shape:
if not elem_shape == ta._elem_shape[0]:
raise ValueError("Shape inference failed.")
else:
ta._elem_shape.append(elem_shape)
return ta
def unpack(self, value, name=None):
"""Packs the values of a `Tensor` in the TensorArray."""
flow_out = gen_data_flow_ops._tensor_array_unpack(
handle=self._handle, value=value, flow_in=self._flow,
name=name)
ta = TensorArray(dtype=self._dtype, size=-1, handle=self._handle)
ta._flow = flow_out
return ta
def unpack(self, value, name=None):
"""Packs the values of a `Tensor` in the TensorArray."""
flow_out = gen_data_flow_ops._tensor_array_unpack(
handle=self._handle, value=value, flow_in=self._flow,
name=name)
ta = TensorArray(dtype=self._dtype, handle=self._handle)
ta._flow = flow_out
return ta
def unpack(self, value):
"""Packs the values of a `Tensor` in the TensorArray."""
flow_out = gen_data_flow_ops._tensor_array_unpack(
handle=self._handle,
value=value,
flow_in=self._flow,
gradient_add=self._gradient_add)
ta = TensorArray(dtype=self._dtype, size=-1, handle=self._handle)
ta._gradient_add = self._gradient_add
ta._flow = flow_out
return ta
def _testTensorArrayUnpackRead(self, tf_dtype, use_gpu):
dtype = tf_dtype.as_numpy_dtype()
with self.test_session(use_gpu=use_gpu) as sess:
h = gen_data_flow_ops._tensor_array(
dtype=tf_dtype, tensor_array_name="foo", size=3)
if tf_dtype == tf.string:
convert = lambda x: np.asarray(x).astype(np.str)
else:
convert = lambda x: np.asarray(x).astype(dtype)
# Unpack a vector into scalars
with tf.control_dependencies([
gen_data_flow_ops._tensor_array_unpack(
h, convert([1.0, 2.0, 3.0]))]):
r0 = gen_data_flow_ops._tensor_array_read(h, 0, tf_dtype)
r1 = gen_data_flow_ops._tensor_array_read(h, 1, tf_dtype)
r2 = gen_data_flow_ops._tensor_array_read(h, 2, tf_dtype)
d0, d1, d2 = sess.run([r0, r1, r2])
self.assertAllEqual(convert(1.0), d0)
self.assertAllEqual(convert(2.0), d1)
self.assertAllEqual(convert(3.0), d2)
# Unpack a matrix into vectors
with tf.control_dependencies([
gen_data_flow_ops._tensor_array_unpack(
h, convert([[1.0, 1.1], [2.0, 2.1], [3.0, 3.1]]))]):
r0 = gen_data_flow_ops._tensor_array_read(h, 0, tf_dtype)
r1 = gen_data_flow_ops._tensor_array_read(h, 1, tf_dtype)
r2 = gen_data_flow_ops._tensor_array_read(h, 2, tf_dtype)
d0, d1, d2 = sess.run([r0, r1, r2])
self.assertAllEqual(convert([1.0, 1.1]), d0)
self.assertAllEqual(convert([2.0, 2.1]), d1)
self.assertAllEqual(convert([3.0, 3.1]), d2)
def unpack(self, value, name=None):
"""Pack the values of a `Tensor` in the TensorArray.
Args:
value: (N+1)-D. Tensor of type `dtype`. The Tensor to unpack.
name: A name for the operation (optional).
Returns:
A new TensorArray object with flow that ensures the unpack occurs.
Use this object all for subsequent operations.
"""
with ops.colocate_with(self._handle):
flow_out = gen_data_flow_ops._tensor_array_unpack(
handle=self._handle, value=value, flow_in=self._flow, name=name
)
ta = TensorArray(dtype=self._dtype, handle=self._handle)
ta._flow = flow_out
return ta
def unpack(self, value, name=None):
"""Pack the values of a `Tensor` in the TensorArray.
Args:
value: (N+1)-D. Tensor of type `dtype`. The Tensor to unpack.
name: A name for the operation (optional).
Returns:
A new TensorArray object with flow that ensures the unpack occurs.
Use this object all for subsequent operations.
"""
with ops.colocate_with(self._handle):
flow_out = gen_data_flow_ops._tensor_array_unpack(
handle=self._handle,
value=value,
flow_in=self._flow,
name=name)
ta = TensorArray(dtype=self._dtype, handle=self._handle)
ta._flow = flow_out
return ta
def _legacy_unpack(self, value, name=None):
"""Pack the values of a `Tensor` in the TensorArray.
This is the legacy version of pack, kept for testing
backwards compatibility.
Args:
value: (N+1)-D. Tensor of type `dtype`. The Tensor to unpack.
name: A name for the operation (optional).
Returns:
A new TensorArray object with flow that ensures the unpack occurs.
Use this object all for subsequent operations.
Raises:
ValueError: if the shape inference fails.
"""
with ops.colocate_with(self._handle):
flow_out = gen_data_flow_ops._tensor_array_unpack(
handle=self._handle,
value=value,
flow_in=self._flow,
name=name)
ta = TensorArray(dtype=self._dtype, handle=self._handle)
ta._flow = flow_out
ta._infer_shape = self._infer_shape
ta._elem_shape = self._elem_shape
if ta._infer_shape:
val_shape = flow_out.op.inputs[1].get_shape()
elem_shape = tensor_shape.unknown_shape()
if val_shape.dims is not None:
elem_shape = tensor_shape.TensorShape(val_shape.dims[1:])
if ta._elem_shape:
if not elem_shape == ta._elem_shape[0]:
raise ValueError(
"Inconsistent shapes: saw %s but expected %s "
"(and infer_shape=True)" %
(elem_shape, ta._elem_shape[0]))
else:
ta._elem_shape.append(elem_shape)
return ta
