*   Add delegate support for `QUANTIZED_16BIT_LSTM`.
    *   Converts hardswish subgraphs into atomic ops.
*   Add support for defaulting the value of `cycle_length` argument of
    `tf.data.Dataset.interleave` to the number of schedulable CPU cores.
*   `parallel_for`: Add converter for `MatrixDiag`.
*   Add `narrow_range` attribute to `QuantizeAndDequantizeV2` and V3.
*   Added new op: `tf.strings.unsorted_segment_join`.

  let summary = "BatchToSpace for N-D tensors of type T.";

  let description = [{
This operation reshapes the "batch" dimension 0 into `M + 1` dimensions of shape
`block_shape + [batch]`, interleaves these blocks back into the grid defined by
the spatial dimensions `[1, ..., M]`, to obtain a result with the same rank as
the input.  The spatial dimensions of this intermediate result are then