LostTech.TensorFlow : API Documentation

Type Iterator

Namespace tensorflow.data

Parent PythonObjectContainer

Interfaces Trackable, IIterator

Represents the state of iterating through a `Dataset`.

Methods

Properties

Public instance methods

object make_initializer(Dataset dataset, string name)

Returns a tf.Operation that initializes this iterator on `dataset`.
Parameters
Dataset dataset
A `Dataset` with compatible structure to this iterator.
string name
(Optional.) A name for the created operation.
Returns
object
A tf.Operation that can be run to initialize this iterator on the given `dataset`.

object make_initializer(BatchDataset dataset, string name)

Returns a tf.Operation that initializes this iterator on `dataset`.
Parameters
BatchDataset dataset
A `Dataset` with compatible structure to this iterator.
string name
(Optional.) A name for the created operation.
Returns
object
A tf.Operation that can be run to initialize this iterator on the given `dataset`.

object make_initializer(DatasetV1Adapter dataset, string name)

Returns a tf.Operation that initializes this iterator on `dataset`.
Parameters
DatasetV1Adapter dataset
A `Dataset` with compatible structure to this iterator.
string name
(Optional.) A name for the created operation.
Returns
object
A tf.Operation that can be run to initialize this iterator on the given `dataset`.

object make_initializer(RepeatDataset dataset, string name)

Returns a tf.Operation that initializes this iterator on `dataset`.
Parameters
RepeatDataset dataset
A `Dataset` with compatible structure to this iterator.
string name
(Optional.) A name for the created operation.
Returns
object
A tf.Operation that can be run to initialize this iterator on the given `dataset`.

object make_initializer_dyn(object dataset, object name)

Returns a tf.Operation that initializes this iterator on `dataset`.
Parameters
object dataset
A `Dataset` with compatible structure to this iterator.
object name
(Optional.) A name for the created operation.
Returns
object
A tf.Operation that can be run to initialize this iterator on the given `dataset`.

Tensor string_handle(string name)

Returns a string-valued tf.Tensor that represents this iterator.
Parameters
string name
(Optional.) A name for the created operation.
Returns
Tensor
A scalar tf.Tensor of type tf.string.

object string_handle_dyn(object name)

Returns a string-valued tf.Tensor that represents this iterator.
Parameters
object name
(Optional.) A name for the created operation.
Returns
object
A scalar tf.Tensor of type tf.string.

Public static methods

Iterator from_string_handle(IGraphNodeBase string_handle, DType output_types, IEnumerable<object> output_shapes, object output_classes)

Creates a new, uninitialized `Iterator` based on the given handle.

This method allows you to define a "feedable" iterator where you can choose between concrete iterators by feeding a value in a tf.Session.run call. In that case, `string_handle` would be a `tf.compat.v1.placeholder`, and you would feed it with the value of tf.data.Iterator.string_handle in each step.

For example, if you had two iterators that marked the current position in a training dataset and a test dataset, you could choose which to use in each step as follows:
Parameters
IGraphNodeBase string_handle
A scalar tf.Tensor of type tf.string that evaluates to a handle produced by the `Iterator.string_handle()` method.
DType output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
IEnumerable<object> 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.
object 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
Iterator
An `Iterator`.
Show Example 
train_iterator = tf.data.Dataset(...).make_one_shot_iterator()
            train_iterator_handle = sess.run(train_iterator.string_handle()) 
 test_iterator = tf.data.Dataset(...).make_one_shot_iterator()
test_iterator_handle = sess.run(test_iterator.string_handle()) 
 handle = tf.compat.v1.placeholder(tf.string, shape=[])
iterator = tf.data.Iterator.from_string_handle(
    handle, train_iterator.output_types) 
 next_element = iterator.get_next()
loss = f(next_element) 
 train_loss = sess.run(loss, feed_dict={handle: train_iterator_handle})
test_loss = sess.run(loss, feed_dict={handle: test_iterator_handle})

Iterator from_string_handle(IGraphNodeBase string_handle, DType output_types, TensorShape output_shapes, object output_classes)

Creates a new, uninitialized `Iterator` based on the given handle.

This method allows you to define a "feedable" iterator where you can choose between concrete iterators by feeding a value in a tf.Session.run call. In that case, `string_handle` would be a `tf.compat.v1.placeholder`, and you would feed it with the value of tf.data.Iterator.string_handle in each step.

For example, if you had two iterators that marked the current position in a training dataset and a test dataset, you could choose which to use in each step as follows:
Parameters
IGraphNodeBase string_handle
A scalar tf.Tensor of type tf.string that evaluates to a handle produced by the `Iterator.string_handle()` method.
DType output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
TensorShape 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.
object 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
Iterator
An `Iterator`.
Show Example 
train_iterator = tf.data.Dataset(...).make_one_shot_iterator()
            train_iterator_handle = sess.run(train_iterator.string_handle()) 
 test_iterator = tf.data.Dataset(...).make_one_shot_iterator()
test_iterator_handle = sess.run(test_iterator.string_handle()) 
 handle = tf.compat.v1.placeholder(tf.string, shape=[])
iterator = tf.data.Iterator.from_string_handle(
    handle, train_iterator.output_types) 
 next_element = iterator.get_next()
loss = f(next_element) 
 train_loss = sess.run(loss, feed_dict={handle: train_iterator_handle})
test_loss = sess.run(loss, feed_dict={handle: test_iterator_handle})

Iterator from_string_handle(IEnumerable<object> string_handle, DType output_types, TensorShape output_shapes, object output_classes)

Creates a new, uninitialized `Iterator` based on the given handle.

This method allows you to define a "feedable" iterator where you can choose between concrete iterators by feeding a value in a tf.Session.run call. In that case, `string_handle` would be a `tf.compat.v1.placeholder`, and you would feed it with the value of tf.data.Iterator.string_handle in each step.

For example, if you had two iterators that marked the current position in a training dataset and a test dataset, you could choose which to use in each step as follows:
Parameters
IEnumerable<object> string_handle
A scalar tf.Tensor of type tf.string that evaluates to a handle produced by the `Iterator.string_handle()` method.
DType output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
TensorShape 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.
object 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
Iterator
An `Iterator`.
Show Example 
train_iterator = tf.data.Dataset(...).make_one_shot_iterator()
            train_iterator_handle = sess.run(train_iterator.string_handle()) 
 test_iterator = tf.data.Dataset(...).make_one_shot_iterator()
test_iterator_handle = sess.run(test_iterator.string_handle()) 
 handle = tf.compat.v1.placeholder(tf.string, shape=[])
iterator = tf.data.Iterator.from_string_handle(
    handle, train_iterator.output_types) 
 next_element = iterator.get_next()
loss = f(next_element) 
 train_loss = sess.run(loss, feed_dict={handle: train_iterator_handle})
test_loss = sess.run(loss, feed_dict={handle: test_iterator_handle})

Iterator from_string_handle(IEnumerable<object> string_handle, DType output_types, IEnumerable<object> output_shapes, object output_classes)

Creates a new, uninitialized `Iterator` based on the given handle.

This method allows you to define a "feedable" iterator where you can choose between concrete iterators by feeding a value in a tf.Session.run call. In that case, `string_handle` would be a `tf.compat.v1.placeholder`, and you would feed it with the value of tf.data.Iterator.string_handle in each step.

For example, if you had two iterators that marked the current position in a training dataset and a test dataset, you could choose which to use in each step as follows:
Parameters
IEnumerable<object> string_handle
A scalar tf.Tensor of type tf.string that evaluates to a handle produced by the `Iterator.string_handle()` method.
DType output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
IEnumerable<object> 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.
object 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
Iterator
An `Iterator`.
Show Example 
train_iterator = tf.data.Dataset(...).make_one_shot_iterator()
            train_iterator_handle = sess.run(train_iterator.string_handle()) 
 test_iterator = tf.data.Dataset(...).make_one_shot_iterator()
test_iterator_handle = sess.run(test_iterator.string_handle()) 
 handle = tf.compat.v1.placeholder(tf.string, shape=[])
iterator = tf.data.Iterator.from_string_handle(
    handle, train_iterator.output_types) 
 next_element = iterator.get_next()
loss = f(next_element) 
 train_loss = sess.run(loss, feed_dict={handle: train_iterator_handle})
test_loss = sess.run(loss, feed_dict={handle: test_iterator_handle})

object from_string_handle_dyn(object string_handle, object output_types, object output_shapes, object output_classes)

Creates a new, uninitialized `Iterator` based on the given handle.

This method allows you to define a "feedable" iterator where you can choose between concrete iterators by feeding a value in a tf.Session.run call. In that case, `string_handle` would be a `tf.compat.v1.placeholder`, and you would feed it with the value of tf.data.Iterator.string_handle in each step.

For example, if you had two iterators that marked the current position in a training dataset and a test dataset, you could choose which to use in each step as follows:
Parameters
object string_handle
A scalar tf.Tensor of type tf.string that evaluates to a handle produced by the `Iterator.string_handle()` method.
object output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
object 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.
object 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
object
An `Iterator`.
Show Example 
train_iterator = tf.data.Dataset(...).make_one_shot_iterator()
            train_iterator_handle = sess.run(train_iterator.string_handle()) 
 test_iterator = tf.data.Dataset(...).make_one_shot_iterator()
test_iterator_handle = sess.run(test_iterator.string_handle()) 
 handle = tf.compat.v1.placeholder(tf.string, shape=[])
iterator = tf.data.Iterator.from_string_handle(
    handle, train_iterator.output_types) 
 next_element = iterator.get_next()
loss = f(next_element) 
 train_loss = sess.run(loss, feed_dict={handle: train_iterator_handle})
test_loss = sess.run(loss, feed_dict={handle: test_iterator_handle})

Iterator from_structure(ValueTuple<DType, object, object> output_types, IEnumerable<object> output_shapes, string shared_name, object output_classes)

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
Parameters
ValueTuple<DType, object, object> output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
IEnumerable<object> 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.
string 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).
object 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
Iterator
An `Iterator`.
Show Example 
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

Iterator from_structure(DType output_types, ValueTuple<IEnumerable<object>, object, object> output_shapes, string shared_name, object output_classes)

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
Parameters
DType output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
ValueTuple<IEnumerable<object>, object, object> 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.
string 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).
object 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
Iterator
An `Iterator`.
Show Example 
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

Iterator from_structure(DType output_types, IEnumerable<object> output_shapes, string shared_name, object output_classes)

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
Parameters
DType output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
IEnumerable<object> 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.
string 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).
object 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
Iterator
An `Iterator`.
Show Example 
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

Iterator from_structure(ValueTuple<DType, object, object> output_types, TensorShape output_shapes, string shared_name, object output_classes)

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
Parameters
ValueTuple<DType, object, object> output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
TensorShape 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.
string 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).
object 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
Iterator
An `Iterator`.
Show Example 
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

Iterator from_structure(ValueTuple<DType, object, object> output_types, ValueTuple<IEnumerable<object>, object, object> output_shapes, string shared_name, object output_classes)

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
Parameters
ValueTuple<DType, object, object> output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
ValueTuple<IEnumerable<object>, object, object> 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.
string 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).
object 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
Iterator
An `Iterator`.
Show Example 
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

Iterator from_structure(DType output_types, TensorShape output_shapes, string shared_name, object output_classes)

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
Parameters
DType output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
TensorShape 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.
string 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).
object 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
Iterator
An `Iterator`.
Show Example 
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

object from_structure_dyn(object output_types, object output_shapes, object shared_name, object output_classes)

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
Parameters
object output_types
A nested structure of tf.DType objects corresponding to each component of an element of this dataset.
object 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.
object 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).
object 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
object
An `Iterator`.
Show Example 
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

Public properties

object element_spec get;

The type specification of an element of this iterator.

object element_spec_dyn get;

The type specification of an element of this iterator.

object initializer get;

A tf.Operation that should be run to initialize this iterator.

object initializer_dyn get;

A tf.Operation that should be run to initialize this iterator.

object output_classes get;

Returns the class of each component of an element of this iterator. (deprecated)

Warning: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Use `tf.compat.v1.data.get_output_classes(iterator)`.

The expected values are tf.Tensor and tf.SparseTensor.

object output_classes_dyn get;

Returns the class of each component of an element of this iterator. (deprecated)

Warning: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Use `tf.compat.v1.data.get_output_classes(iterator)`.

The expected values are tf.Tensor and tf.SparseTensor.

object output_shapes get;

Returns the shape of each component of an element of this iterator. (deprecated)

Warning: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Use `tf.compat.v1.data.get_output_shapes(iterator)`.

object output_shapes_dyn get;

Returns the shape of each component of an element of this iterator. (deprecated)

Warning: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Use `tf.compat.v1.data.get_output_shapes(iterator)`.

object output_types get;

Returns the type of each component of an element of this iterator. (deprecated)

Warning: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Use `tf.compat.v1.data.get_output_types(iterator)`.

object output_types_dyn get;

Returns the type of each component of an element of this iterator. (deprecated)

Warning: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Use `tf.compat.v1.data.get_output_types(iterator)`.

object PythonObject get;