Module: tf.data

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 an iterator of a tf.data.Dataset.

class IteratorSpec: Type specification for tf.data.Iterator.

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: Creates a Dataset comprising lines from one or more text files.

class ThreadingOptions: Represents options for dataset threading.

AUTOTUNE -1
INFINITE_CARDINALITY -1
UNKNOWN_CARDINALITY -2