serialization or iterator checkpointing.
    *   Changing `tf.data.experimental.save` to store the type specification of
        the dataset elements. This avoids the need for explicitly specifying the
        `element_spec` argument of `tf.data.experimental.load` when loading the
        previously saved dataset.

                            minimum TLS version or supported cipher suites.


                            A set of common keys MAY be defined by the API in the future. To avoid
                            any ambiguity, implementation-specific definitions MUST use
                            domain-prefixed names, such as `example.com/my-custom-option`.

          },
          "selector": {
            "description": "selector is the label query over pods that should match the replicas count. This is same as the label selector but in the string format to avoid introspection by clients. The string will be in the same format as the query-param syntax. More info about label selectors: https://kubernetes.io/docs/concepts/overview/working-with-objects/labels/",
            "type": "string"

    Arg<TF_I32OrI64Tensor, [{List of N integers containing the slice size for each
dimension. Each value must be strictly greater than zero, and start + size
must be less than or equal to the size of the dimension to avoid
implementation defined behavior.}]>:$start_indices,
    TF_I32OrI64Tensor:$size_indices
  );

  let results = (outs
    TF_Tensor:$output
  );