Scikeras

What's Changed
* Fixed Some Typos in `README.md` and Documentation by RoyiAvital in https://github.com/adriangb/scikeras/pull/269
* Add sklearn mixin inheritance by go95 in https://github.com/adriangb/scikeras/pull/267
* chore: run nightly tf in CI and fix compatibility with TF 2.9.0 by adriangb in https://github.com/adriangb/scikeras/pull/271

New Contributors
* RoyiAvital made their first contribution in https://github.com/adriangb/scikeras/pull/269
* go95 made their first contribution in https://github.com/adriangb/scikeras/pull/267

**Full Changelog**: https://github.com/adriangb/scikeras/compare/v0.6.1...v0.7.0

What's Changed
* doc: document that TensorFlow Datasets are not supported as inputs by adriangb in https://github.com/adriangb/scikeras/pull/266


**Full Changelog**: https://github.com/adriangb/scikeras/compare/v0.6.0...v0.6.1

What's Changed
* MAINT: add optional packages for tensorflow and tensorflow-cpu; update documentation by metasyn in https://github.com/adriangb/scikeras/pull/262

New Contributors
* metasyn made their first contribution in https://github.com/adriangb/scikeras/pull/262

**Full Changelog**: https://github.com/adriangb/scikeras/compare/v0.5.1...v0.6.0

What's Changed
* BREAKING CHANGE: bump TensorFlow dependency to 2.7.0 by adriangb in https://github.com/adriangb/scikeras/pull/252
* feat: add context manager for setting TensorFlow's random generator by adriangb in https://github.com/adriangb/scikeras/pull/254
* doc: add note on KerasRegressor.score() return value to migration guide by adriangb in https://github.com/adriangb/scikeras/pull/255


**Full Changelog**: https://github.com/adriangb/scikeras/compare/v0.4.1...v0.5.1

Fixes:

* Add missing dependency on `packaging` (247). Thank you sim-san for reporting and fixing!
* Fixed an issue causing routed parameters (`fit__class_weight`) to be overridden by non-routed parameters, including default values (`class_weight=None`) (244). Thank you 8549 for reporting the issue.
* Fix handling of `class_weight="balanced"` (245)

Features:

* Support routed parameters to callbacks (233)

Changes:

* Official support for extra keyword arguments (`**kwargs`) in `fit()`, `predict()` and `predict_proba()` that are routed directly to Keras (238)
* Remove hard dependency on `tensorflow`, allowing users to choose `tensorflow-cpu` or other distributions in environments where they can't control `pip` arguments (237)