* ✏️ Fix typo in `docs/ja/docs/tutorial/encoder.md`. PR [#13815](https://github.com/fastapi/fastapi/pull/13815) by [@ruzia](https://github.com/ruzia).
* ✏️ Fix typo in `docs/ja/docs/tutorial/handling-errors.md`. PR [#13814](https://github.com/fastapi/fastapi/pull/13814) by [@ruzia](https://github.com/ruzia).
* ✏️ Fix typo in `docs/ja/docs/tutorial/body-fields.md`. PR [#13802](https://github.com/fastapi/fastapi/pull/13802) by [@ruzia](https://github.com/ruzia).

    *   Raise `ValueError` if an integer is passed to the training APIs.
    *   Added fault-tolerance support for training Keras model via `model.fit()`
        with `MultiWorkerMirroredStrategy`, tutorial available.
    *   Custom Callback tutorial is now available.
    *   To train with `tf.distribute`, Keras API is recommended over estimator.
    *   `steps_per_epoch` and `steps` arguments are supported with numpy arrays.

You can find more information about how to restrict container system calls with seccomp in the new [documentation page on Kubernetes.io][seccomp-docs]

[seccomp-docs]: https://kubernetes.io/docs/tutorials/clusters/seccomp/


### Production images moved to community control

As of Kuberenetes v1.19, Kubernetes container images are stored on a community-controlled storage bucket,