order, the violet lying on the inside of the Fringe next the Shadow, the
red on the outside farthest from the Shadow, and the blue, green, and
yellow, in the middle. And, in like manner, the middlemost Fringes of
all the Colours lying in order, and being contiguous, make another broad
Fringe composed of all the Colours; and the outmost Fringes of all the
Colours lying in order, and being contiguous, make a third broad Fringe

*   `tf.data`

    *   `tf.data.Dataset.zip` now supports Python-style zipping, i.e. `Dataset.zip(a, b, c)`.

- Enable feature Gate DefaultPodTopologySpread to use PodTopologySpread plugin to do defaultspreading. In doing so, legacy DefaultPodTopologySpread plugin is disabled. ([#91793](https://github.com/kubernetes/kubernetes/pull/91793), [@alculquicondor](https://github.com/alculquicondor)) [SIG Scheduling]

padded on the left with ones. So this gives the same result as the previous
example:

>>> x = tf.constant([1, 2, 3])   # Shape (3,)
>>> y = tf.broadcast_to(x, [2, 3])


When doing broadcasted operations such as multiplying a tensor
by a scalar, broadcasting (usually) confers some time or space
benefit, as the broadcasted tensor is never materialized.