tf.compat.v1.data.make_initializable_iterator

Creates an iterator for elements of dataset.

tf.compat.v1.data.make_initializable_iterator(
    dataset, shared_name=None
)

Migrate to TF2

This is a legacy API for consuming dataset elements and should only be used during transition from TF 1 to TF 2. Note that using this API should be a transient state of your code base as there are in general no guarantees about the interoperability of TF 1 and TF 2 code.

In TF 2 datasets are Python iterables which means you can consume their elements using for elem in dataset: ... or by explicitly creating iterator via iterator = iter(dataset) and fetching its elements via values = next(iterator).

Description

dataset = ...
iterator = tf.compat.v1.data.make_initializable_iterator(dataset)
# ...
sess.run(iterator.initializer)

dataset A tf.data.Dataset.
shared_name (Optional.) If non-empty, the returned iterator will be shared under the given name across multiple sessions that share the same devices (e.g. when using a remote server).

A tf.data.Iterator for elements of dataset.

RuntimeError If eager execution is enabled.