the [RFC](https://github.com/tensorflow/community/pull/415) for more
        details regarding its design and properties.

*   `tf.data`:
    *   Graduated experimental APIs:
        * [`tf.data.Dataset.ragged_batch`](https://www.tensorflow.org/api_docs/python/tf/data/Dataset/#ragged_batch), which batches elements of `tf.data.Dataset`s into `tf.RaggedTensor`s.

    // TODO(jlevy): For optimal performance, use a binary search when
    // targets.size() < size() / log(size())
    // TODO(kevinb): see if we can share code with OrderedIterator after it
    // graduates from labs.
    if (targets instanceof Multiset) {
      targets = ((Multiset<?>) targets).elementSet();
    }
    if (!SortedIterables.hasSameComparator(comparator(), targets) || (targets.size() <= 1)) {

    // TODO(jlevy): For optimal performance, use a binary search when
    // targets.size() < size() / log(size())
    // TODO(kevinb): see if we can share code with OrderedIterator after it
    // graduates from labs.
    if (targets instanceof Multiset) {
      targets = ((Multiset<?>) targets).elementSet();
    }
    if (!SortedIterables.hasSameComparator(comparator(), targets) || (targets.size() <= 1)) {