tf.data.Dataset API for input pipelines.
See Importing Data for an overview.
Modules
experimental module: Experimental API for building input pipelines.
Classes
class Dataset: Represents a potentially large set of elements.
class DatasetSpec: Type specification for tf.data.Dataset.
class FixedLengthRecordDataset: A Dataset of fixed-length records from one or more binary files.
class Iterator: Represents the state of iterating through a Dataset.
class NumpyIterator: Iterator over a dataset with elements converted to numpy.
class Options: Represents options for tf.data.Dataset.
class TFRecordDataset: A Dataset comprising records from one or more TFRecord files.
class TextLineDataset: A Dataset comprising lines from one or more text files.
class ThreadingOptions: Represents options for dataset threading.
Functions
get_output_classes(...): Returns the output classes for elements of the input dataset / iterator.
get_output_shapes(...): Returns the output shapes for elements of the input dataset / iterator.
get_output_types(...): Returns the output shapes for elements of the input dataset / iterator.
make_initializable_iterator(...): Creates an iterator for elements of dataset.
make_one_shot_iterator(...): Creates an iterator for elements of dataset.
Other Members | |
|---|---|
| AUTOTUNE |
-1
|
| INFINITE_CARDINALITY |
-1
|
| UNKNOWN_CARDINALITY |
-2
|