def __init__(self, dataset):
"""Creates a new iterator over the given dataset.
For example:
```python
dataset = tf.contrib.data.Dataset.range(4)
for x in Iterator(dataset):
print(x)
```
Args:
dataset: A `tf.contrib.data.Dataset` object.
Raises:
RuntimeError: When invoked without eager execution enabled.
"""
if not context.in_eager_mode():
raise RuntimeError(
"{} objects only make sense when eager execution is enabled".format(
type(self)))
ds_variant = dataset._as_variant_tensor() # pylint: disable=protected-access
self._output_types = dataset.output_types
self._flat_output_types = nest.flatten(dataset.output_types)
self._flat_output_shapes = nest.flatten(dataset.output_shapes)
self._resource = gen_dataset_ops.iterator(
container="",
shared_name=_iterator_shared_name(),
output_types=self._flat_output_types,
output_shapes=self._flat_output_shapes)
gen_dataset_ops.make_iterator(ds_variant, self._resource)
def __init__(self, dataset):
"""Creates a new iterator over the given dataset.
For example:
```python
dataset = tf.contrib.data.Dataset.range(4)
for x in Iterator(dataset):
print(x)
```
Args:
dataset: A `tf.contrib.data.Dataset` object.
Raises:
RuntimeError: When invoked without eager execution enabled.
"""
if not context.in_eager_mode():
raise RuntimeError(
"{} objects only make sense when eager execution is enabled".
format(type(self)))
ds_variant = dataset.make_dataset_resource()
self._output_types = dataset.output_types
self._flat_output_types = nest.flatten(dataset.output_types)
self._flat_output_shapes = nest.flatten(dataset.output_shapes)
self._resource = gen_dataset_ops.iterator(
container="",
shared_name=_iterator_shared_name(),
output_types=self._flat_output_types,
output_shapes=self._flat_output_shapes)
gen_dataset_ops.make_iterator(ds_variant, self._resource)
def __init__(self, dataset):
"""Creates a new iterator over the given dataset.
For example:
```python
dataset = tf.data.Dataset.range(4)
for x in Iterator(dataset):
print(x)
```
Tensors produced will be placed on the device on which this iterator object
was created.
Args:
dataset: A `tf.data.Dataset` object.
Raises:
RuntimeError: When invoked without eager execution enabled.
"""
if not context.executing_eagerly():
raise RuntimeError(
"{} objects can only be used when eager execution is enabled, use "
"tf.data.Dataset.make_initializable_iterator or "
"tf.data.Dataset.make_one_shot_iterator for graph construction"
.format(type(self)))
with ops.device("/device:CPU:0"):
ds_variant = dataset._as_variant_tensor() # pylint: disable=protected-access
self._output_classes = dataset.output_classes
self._output_types = dataset.output_types
self._output_shapes = dataset.output_shapes
self._flat_output_types = nest.flatten(
sparse.as_dense_types(self._output_types,
self._output_classes))
self._flat_output_shapes = nest.flatten(
sparse.as_dense_shapes(self._output_shapes,
self._output_classes))
self._resource = gen_dataset_ops.iterator(
shared_name="",
container=_generate_shared_name("eageriterator"),
output_types=self._flat_output_types,
output_shapes=self._flat_output_shapes)
gen_dataset_ops.make_iterator(ds_variant, self._resource)
# Delete the resource when this object is deleted
self._resource_deleter = resource_variable_ops.EagerResourceDeleter(
handle=self._resource, handle_device="/device:CPU:0")
self._device = context.context().device_name
def __init__(self, dataset):
"""Creates a new iterator over the given dataset.
For example:
```python
dataset = tf.data.Dataset.range(4)
for x in Iterator(dataset):
print(x)
```
Tensors produced will be placed on the device on which this iterator object
was created.
Args:
dataset: A `tf.data.Dataset` object.
Raises:
RuntimeError: When invoked without eager execution enabled.
"""
if not context.executing_eagerly():
raise RuntimeError(
"{} objects can only be used when eager execution is enabled, use "
"tf.data.Dataset.make_initializable_iterator or "
"tf.data.Dataset.make_one_shot_iterator for graph construction".
format(type(self)))
with ops.device("/device:CPU:0"):
ds_variant = dataset._as_variant_tensor() # pylint: disable=protected-access
self._output_classes = dataset.output_classes
self._output_types = dataset.output_types
self._output_shapes = dataset.output_shapes
self._flat_output_types = nest.flatten(
sparse.as_dense_types(self._output_types, self._output_classes))
self._flat_output_shapes = nest.flatten(
sparse.as_dense_shapes(self._output_shapes, self._output_classes))
self._resource = gen_dataset_ops.iterator(
shared_name="",
container=_generate_shared_name("eageriterator"),
output_types=self._flat_output_types,
output_shapes=self._flat_output_shapes)
gen_dataset_ops.make_iterator(ds_variant, self._resource)
# Delete the resource when this object is deleted
self._resource_deleter = resource_variable_ops.EagerResourceDeleter(
handle=self._resource, handle_device="/device:CPU:0")
self._device = context.context().device_name
def from_structure(output_types,
output_shapes=None,
shared_name=None,
output_classes=None):
"""Creates a new, uninitialized `Iterator` with the given structure.
This iterator-constructing method can be used to create an iterator that
is reusable with many different datasets.
The returned iterator is not bound to a particular dataset, and it has
no `initializer`. To initialize the iterator, run the operation returned by
`Iterator.make_initializer(dataset)`.
The following is an example
```python
iterator = Iterator.from_structure(tf.int64, tf.TensorShape([]))
dataset_range = Dataset.range(10)
range_initializer = iterator.make_initializer(dataset_range)
dataset_evens = dataset_range.filter(lambda x: x % 2 == 0)
evens_initializer = iterator.make_initializer(dataset_evens)
# Define a model based on the iterator; in this example, the model_fn
# is expected to take scalar tf.int64 Tensors as input (see
# the definition of 'iterator' above).
prediction, loss = model_fn(iterator.get_next())
# Train for `num_epochs`, where for each epoch, we first iterate over
# dataset_range, and then iterate over dataset_evens.
for _ in range(num_epochs):
# Initialize the iterator to `dataset_range`
sess.run(range_initializer)
while True:
try:
pred, loss_val = sess.run([prediction, loss])
except tf.errors.OutOfRangeError:
break
# Initialize the iterator to `dataset_evens`
sess.run(evens_initializer)
while True:
try:
pred, loss_val = sess.run([prediction, loss])
except tf.errors.OutOfRangeError:
break
```
Args:
output_types: A nested structure of `tf.DType` objects corresponding to
each component of an element of this dataset.
output_shapes: (Optional.) A nested structure of `tf.TensorShape` objects
corresponding to each component of an element of this dataset. If
omitted, each component will have an unconstrainted shape.
shared_name: (Optional.) If non-empty, this iterator will be shared under
the given name across multiple sessions that share the same devices
(e.g. when using a remote server).
output_classes: (Optional.) A nested structure of Python `type` objects
corresponding to each component of an element of this iterator. If
omitted, each component is assumed to be of type `tf.Tensor`.
Returns:
An `Iterator`.
Raises:
TypeError: If the structures of `output_shapes` and `output_types` are
not the same.
"""
output_types = nest.map_structure(dtypes.as_dtype, output_types)
if output_shapes is None:
output_shapes = nest.map_structure(
lambda _: tensor_shape.TensorShape(None), output_types)
else:
output_shapes = nest.map_structure_up_to(output_types,
tensor_shape.as_shape,
output_shapes)
if output_classes is None:
output_classes = nest.map_structure(lambda _: ops.Tensor,
output_types)
nest.assert_same_structure(output_types, output_shapes)
output_structure = structure_lib.convert_legacy_structure(
output_types, output_shapes, output_classes)
if shared_name is None:
shared_name = ""
# pylint: disable=protected-access
if compat.forward_compatible(2018, 8, 3):
if _device_stack_is_empty():
with ops.device("/cpu:0"):
iterator_resource = gen_dataset_ops.iterator_v2(
container="",
shared_name=shared_name,
output_types=output_structure._flat_types,
output_shapes=output_structure._flat_shapes)
else:
iterator_resource = gen_dataset_ops.iterator_v2(
container="",
shared_name=shared_name,
output_types=output_structure._flat_types,
output_shapes=output_structure._flat_shapes)
else:
iterator_resource = gen_dataset_ops.iterator(
container="",
shared_name=shared_name,
output_types=output_structure._flat_types,
output_shapes=output_structure._flat_shapes)
# pylint: enable=protected-access
return Iterator(iterator_resource, None, output_types, output_shapes,
output_classes)
def __init__(self, dataset):
"""Creates a new iterator over the given dataset.
For example:
```python
dataset = tf.data.Dataset.range(4)
for x in Iterator(dataset):
print(x)
```
Tensors produced will be placed on the device on which this iterator object
was created.
Args:
dataset: A `tf.data.Dataset` object.
Raises:
RuntimeError: When invoked without eager execution enabled.
"""
if not context.in_eager_mode():
raise RuntimeError(
"{} objects can only be used when eager execution is enabled, use "
"tf.data.Dataset.make_iterator or "
"tf.data.Dataset.make_one_shot_iterator for graph construction"
.format(type(self)))
with ops.device("/device:CPU:0"):
ds_variant = dataset._as_variant_tensor() # pylint: disable=protected-access
self._output_types = dataset.output_types
self._output_shapes = dataset.output_shapes
self._flat_output_types = nest.flatten(dataset.output_types)
self._flat_output_shapes = nest.flatten(dataset.output_shapes)
self._resource = gen_dataset_ops.iterator(
container="",
shared_name=_generate_shared_name("eager_iterator"),
output_types=self._flat_output_types,
output_shapes=self._flat_output_shapes)
gen_dataset_ops.make_iterator(ds_variant, self._resource)
# Delete the resource when this object is deleted
self._resource_deleter = resource_variable_ops.EagerResourceDeleter(
handle=self._resource, handle_device="/device:CPU:0")
self._device = context.context().device_name
self._buffer_resource_handle = None
if not context.context().device_spec.device_type:
is_remote_device = False
else:
is_remote_device = context.context(
).device_spec.device_type != "CPU"
if is_remote_device:
with ops.device("/device:CPU:0"):
iter_string_handle = gen_dataset_ops.iterator_to_string_handle(
self._resource)
@function.Defun(dtypes.string)
def remote_fn(h):
remote_iterator = iterator_ops.Iterator.from_string_handle(
h, self._output_types, self._output_shapes)
return remote_iterator.get_next()
remote_fn.add_to_graph(None)
target = constant_op.constant("/device:CPU:0")
with ops.device(self._device):
self._buffer_resource_handle = prefetching_ops.function_buffering_resource(
string_arg=iter_string_handle,
f=remote_fn,
target_device=target,
buffer_size=10,
thread_pool_size=1,
container="",
shared_name=_generate_shared_name(
"function_buffer_resource"))
self._buffer_resource_deleter = resource_variable_ops.EagerResourceDeleter(
handle=self._buffer_resource_handle,
handle_device=self._device)
def from_structure(output_types,
output_shapes=None,
shared_name=None,
output_classes=None):
"""Creates a new, uninitialized `Iterator` with the given structure.
This iterator-constructing method can be used to create an iterator that
is reusable with many different datasets.
The returned iterator is not bound to a particular dataset, and it has
no `initializer`. To initialize the iterator, run the operation returned by
`Iterator.make_initializer(dataset)`.
The following is an example
```python
iterator = Iterator.from_structure(tf.int64, tf.TensorShape([]))
dataset_range = Dataset.range(10)
range_initializer = iterator.make_initializer(dataset_range)
dataset_evens = dataset_range.filter(lambda x: x % 2 == 0)
evens_initializer = iterator.make_initializer(dataset_evens)
# Define a model based on the iterator; in this example, the model_fn
# is expected to take scalar tf.int64 Tensors as input (see
# the definition of 'iterator' above).
prediction, loss = model_fn(iterator.get_next())
# Train for `num_epochs`, where for each epoch, we first iterate over
# dataset_range, and then iterate over dataset_evens.
for _ in range(num_epochs):
# Initialize the iterator to `dataset_range`
sess.run(range_initializer)
while True:
try:
pred, loss_val = sess.run([prediction, loss])
except tf.errors.OutOfRangeError:
break
# Initialize the iterator to `dataset_evens`
sess.run(evens_initializer)
while True:
try:
pred, loss_val = sess.run([prediction, loss])
except tf.errors.OutOfRangeError:
break
```
Args:
output_types: A nested structure of `tf.DType` objects corresponding to
each component of an element of this dataset.
output_shapes: (Optional.) A nested structure of `tf.TensorShape` objects
corresponding to each component of an element of this dataset. If
omitted, each component will have an unconstrainted shape.
shared_name: (Optional.) If non-empty, this iterator will be shared under
the given name across multiple sessions that share the same devices
(e.g. when using a remote server).
output_classes: (Optional.) A nested structure of Python `type` objects
corresponding to each component of an element of this iterator. If
omitted, each component is assumed to be of type `tf.Tensor`.
Returns:
An `Iterator`.
Raises:
TypeError: If the structures of `output_shapes` and `output_types` are
not the same.
"""
output_types = nest.map_structure(dtypes.as_dtype, output_types)
if output_shapes is None:
output_shapes = nest.map_structure(
lambda _: tensor_shape.TensorShape(None), output_types)
else:
output_shapes = nest.map_structure_up_to(
output_types, tensor_shape.as_shape, output_shapes)
if output_classes is None:
output_classes = nest.map_structure(lambda _: ops.Tensor, output_types)
nest.assert_same_structure(output_types, output_shapes)
if shared_name is None:
shared_name = ""
iterator_resource = gen_dataset_ops.iterator(
container="",
shared_name=shared_name,
output_types=nest.flatten(
sparse.as_dense_types(output_types, output_classes)),
output_shapes=nest.flatten(
sparse.as_dense_shapes(output_shapes, output_classes)))
return Iterator(iterator_resource, None, output_types, output_shapes,
output_classes)
def __init__(self, dataset):
"""Creates a new iterator over the given dataset.
For example:
```python
dataset = tf.data.Dataset.range(4)
for x in Iterator(dataset):
print(x)
```
Tensors produced will be placed on the device on which this iterator object
was created.
Args:
dataset: A `tf.data.Dataset` object.
Raises:
RuntimeError: When invoked without eager execution enabled.
"""
if not context.in_eager_mode():
raise RuntimeError(
"{} objects can only be used when eager execution is enabled, use "
"tf.data.Dataset.make_iterator or "
"tf.data.Dataset.make_one_shot_iterator for graph construction".
format(type(self)))
with ops.device("/device:CPU:0"):
ds_variant = dataset._as_variant_tensor() # pylint: disable=protected-access
self._output_types = dataset.output_types
self._output_shapes = dataset.output_shapes
self._flat_output_types = nest.flatten(dataset.output_types)
self._flat_output_shapes = nest.flatten(dataset.output_shapes)
self._resource = gen_dataset_ops.iterator(
container="",
shared_name=_generate_shared_name("eager_iterator"),
output_types=self._flat_output_types,
output_shapes=self._flat_output_shapes)
gen_dataset_ops.make_iterator(ds_variant, self._resource)
# Delete the resource when this object is deleted
self._resource_deleter = resource_variable_ops.EagerResourceDeleter(
handle=self._resource, handle_device="/device:CPU:0")
self._device = context.context().device_name
self._buffer_resource_handle = None
if not context.context().device_spec.device_type:
is_remote_device = False
else:
is_remote_device = context.context().device_spec.device_type != "CPU"
if is_remote_device:
with ops.device("/device:CPU:0"):
iter_string_handle = gen_dataset_ops.iterator_to_string_handle(
self._resource)
@function.Defun(dtypes.string)
def remote_fn(h):
remote_iterator = iterator_ops.Iterator.from_string_handle(
h, self._output_types, self._output_shapes)
return remote_iterator.get_next()
remote_fn.add_to_graph(None)
target = constant_op.constant("/device:CPU:0")
with ops.device(self._device):
self._buffer_resource_handle = prefetching_ops.function_buffering_resource(
string_arg=iter_string_handle,
f=remote_fn,
target_device=target,
buffer_size=10,
thread_pool_size=1,
container="",
shared_name=_generate_shared_name("function_buffer_resource"))
self._buffer_resource_deleter = resource_variable_ops.EagerResourceDeleter(
handle=self._buffer_resource_handle, handle_device=self._device)
