[tf.data] Adding support for `num_parallel_calls` to `tf.data.Dataset.interleave`.

Unlike the `tf.data.contrib.parallel_interleave` whose parallelism is tied to the `cycle_length` argument, the newly introduced `num_parallel_calls` argument of `tf.data.Dataset.interleave` is decoupled from the `cycle_length` argument and identifies the degree of parallelism to use for fetching output elements.

PiperOrigin-RevId: 211886816

1 files changed, 13 insertions, 0 deletions

diff --git a/tensorflow/core/api_def/base_api/api_def_ParallelInterleaveDatasetV2.pbtxt b/tensorflow/core/api_def/base_api/api_def_ParallelInterleaveDatasetV2.pbtxt
new file mode 100644
index 0000000000..27bc4013c3
--- /dev/null
+++ b/tensorflow/core/api_def/base_api/api_def_ParallelInterleaveDatasetV2.pbtxt
@@ -0,0 +1,13 @@
+op {
+  graph_op_name: "ParallelInterleaveDatasetV2"
+  visibility: HIDDEN
+  attr {
+    name: "f"
+    description: <<END
+A function mapping elements of `input_dataset`, concatenated with
+`other_arguments`, to a Dataset variant that contains elements matching
+`output_types` and `output_shapes`.
+END
+  }
+  summary: "Creates a dataset that applies `f` to the outputs of `input_dataset`."
+}