View source on GitHub |
Represents an iterator of a tf.data.Dataset
.
tf.data.Iterator
is the primary mechanism for enumerating elements of a tf.data.Dataset
. It supports the Python Iterator protocol, which means it can be iterated over using a for-loop:
dataset = tf.data.Dataset.range(2) for element in dataset: print(element) tf.Tensor(0, shape=(), dtype=int64) tf.Tensor(1, shape=(), dtype=int64)
or by fetching individual elements explicitly via get_next()
:
dataset = tf.data.Dataset.range(2) iterator = iter(dataset) print(iterator.get_next()) tf.Tensor(0, shape=(), dtype=int64) print(iterator.get_next()) tf.Tensor(1, shape=(), dtype=int64)
In addition, non-raising iteration is supported via get_next_as_optional()
, which returns the next element (if available) wrapped in a tf.experimental.Optional
.
dataset = tf.data.Dataset.from_tensors(42) iterator = iter(dataset) optional = iterator.get_next_as_optional() print(optional.has_value()) tf.Tensor(True, shape=(), dtype=bool) optional = iterator.get_next_as_optional() print(optional.has_value()) tf.Tensor(False, shape=(), dtype=bool)
Attributes | |
---|---|
element_spec | The type specification of an element of this iterator. dataset = tf.data.Dataset.from_tensors(42) iterator = iter(dataset) iterator.element_spec tf.TensorSpec(shape=(), dtype=tf.int32, name=None) |
get_next
@abc.abstractmethod get_next()
Returns a nested structure of tf.Tensor
s containing the next element.
dataset = tf.data.Dataset.from_tensors(42) iterator = iter(dataset) print(iterator.get_next()) tf.Tensor(42, shape=(), dtype=int32)
Returns | |
---|---|
A nested structure of tf.Tensor objects. |
Raises | |
---|---|
tf.errors.OutOfRangeError : If the end of the iterator has been reached. |
get_next_as_optional
@abc.abstractmethod get_next_as_optional()
Returns a tf.experimental.Optional
which contains the next element.
If the iterator has reached the end of the sequence, the returned tf.experimental.Optional
will have no value.
dataset = tf.data.Dataset.from_tensors(42) iterator = iter(dataset) optional = iterator.get_next_as_optional() print(optional.has_value()) tf.Tensor(True, shape=(), dtype=bool) print(optional.get_value()) tf.Tensor(42, shape=(), dtype=int32) optional = iterator.get_next_as_optional() print(optional.has_value()) tf.Tensor(False, shape=(), dtype=bool)
Returns | |
---|---|
A tf.experimental.Optional object representing the next element. |
__iter__
__iter__()
© 2020 The TensorFlow Authors. All rights reserved.
Licensed under the Creative Commons Attribution License 3.0.
Code samples licensed under the Apache 2.0 License.
https://www.tensorflow.org/versions/r2.4/api_docs/python/tf/data/Iterator