tf.raw_ops.ParallelFilterDataset

Creates a dataset containing elements of input_dataset matching predicate.

tf.raw_ops.ParallelFilterDataset(
    input_dataset,
    other_arguments,
    num_parallel_calls,
    predicate,
    output_types,
    output_shapes,
    deterministic='default',
    metadata='',
    name=None
)

The predicate function must return a scalar boolean and accept the following arguments:

  • One tensor for each component of an element of input_dataset.
  • One tensor for each value in other_arguments.

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

input_dataset A Tensor of type variant.
other_arguments A list of Tensor objects. A list of tensors, typically values that were captured when building a closure for predicate.
num_parallel_calls A Tensor of type int64. The number of concurrent invocations of predicate that process elements from input_dataset in parallel.
predicate A function decorated with @Defun. A function returning a scalar boolean.
output_types A list of tf.DTypes that has length >= 1.
output_shapes A list of shapes (each a tf.TensorShape or list of ints) that has length >= 1.
deterministic An optional string. Defaults to "default". A string indicating the op-level determinism to use. Deterministic controls whether the interleave is allowed to return elements out of order if the next element to be returned isn't available, but a later element is. Options are "true", "false", and "default". "default" indicates that determinism should be decided by the experimental_deterministic parameter of tf.data.Options.
metadata An optional string. Defaults to "".
name A name for the operation (optional).

A Tensor of type variant.