tfdatasets (version 2.17.0)

make-iterator: Creates an iterator for enumerating the elements of this dataset.

Description

Creates an iterator for enumerating the elements of this dataset.

Usage

make_iterator_one_shot(dataset)
make_iterator_initializable(dataset, shared_name = NULL)
make_iterator_from_structure(
  output_types,
  output_shapes = NULL,
  shared_name = NULL
)
make_iterator_from_string_handle(
  string_handle,
  output_types,
  output_shapes = NULL
)

Value

An Iterator over the elements of this dataset.

Arguments

dataset

A 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).

output_types

A nested structure of tf$DType objects corresponding to each component of an element of this iterator.

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_handle

A scalar tensor of type string that evaluates to a handle produced by the iterator_string_handle() method.

Initialization

For make_iterator_one_shot(), the returned iterator will be initialized automatically. A "one-shot" iterator does not currently support re-initialization.

For make_iterator_initializable(), the returned iterator will be in an uninitialized state, and you must run the object returned from iterator_initializer() before using it.

For make_iterator_from_structure(), 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().

See Also

Other iterator functions: iterator_get_next(), iterator_initializer(), iterator_make_initializer(), iterator_string_handle()