*   `tf.data.Dataset.zip` now supports Python-style zipping, i.e. `Dataset.zip(a, b, c)`.
    * `tf.data.Dataset.shuffle` now supports `tf.data.UNKNOWN_CARDINALITY` When doing a "full shuffle" using  `dataset = dataset.shuffle(dataset.cardinality())`. But remember, a "full shuffle" will load the full dataset into memory so that it can be shuffled, so make sure to only use this with small datasets or datasets of small objects (like filenames).

*   `tf.math`

len(src) < BlockSize { panic("crypto/aes: input not full block") } if len(dst) < BlockSize { panic("crypto/aes: output not full block") } if alias.InexactOverlap(dst[:BlockSize], src[:BlockSize]) { panic("crypto/aes: invalid buffer overlap") } encryptBlock(c, dst, src) } func (c *Block) Decrypt(dst, src []byte) { // AES-ECB is not approved in FIPS 140-3 mode. fips140.RecordNonApproved() if len(src) < BlockSize { panic("crypto/aes: input not full block") } if len(dst) < BlockSize { panic("crypto/aes:...

* Add GitHub social preview images to git. PR [#752](https://github.com/tiangolo/fastapi/pull/752).
* Update PyPI "trove classifiers". PR [#751](https://github.com/tiangolo/fastapi/pull/751).
* Add full support for Python 3.8. Enable Python 3.8 in full in Travis. PR [749](https://github.com/tiangolo/fastapi/pull/749).
* Update "new issue" templates. PR [#749](https://github.com/tiangolo/fastapi/pull/749).