tf.raw_ops.ParallelMapDataset

Creates a dataset that applies f to the outputs of input_dataset.

tf.raw_ops.ParallelMapDataset(
    input_dataset,
    other_arguments,
    num_parallel_calls,
    f,
    output_types,
    output_shapes,
    use_inter_op_parallelism=True,
    sloppy=False,
    preserve_cardinality=False,
    metadata='',
    name=None
)

Unlike a "MapDataset", which applies f sequentially, this dataset invokes up to num_parallel_calls copies of f in parallel.

input_datasetA Tensor of type variant.
other_argumentsA list of Tensor objects.
num_parallel_callsA Tensor of type int32. The number of concurrent invocations of f that process elements from input_dataset in parallel.
fA function decorated with @Defun.
output_typesA list of tf.DTypes that has length >= 1.
output_shapesA list of shapes (each a tf.TensorShape or list of ints) that has length >= 1.
use_inter_op_parallelismAn optional bool. Defaults to True.
sloppyAn optional bool. Defaults to False.
preserve_cardinalityAn optional bool. Defaults to False.
metadataAn optional string. Defaults to "".
nameA name for the operation (optional).

A Tensor of type variant.