Tensorflow-federated

Major Features and Improvements

* Support for multiple outputs and loss functions in `keras` models.
* Support for stateful broadcast and aggregation functions in federated
averaging and federated SGD APIs.
* `tff.utils.update_state` extended to handle more general `state` arguments.
* Addition of `tff.utils.federated_min` and `tff.utils.federated_max`.
* Shuffle `client_ids` in `create_tf_dataset_from_all_clients` by default to
aid optimization.

Breaking Changes

* Dependencies added to `requirements.txt`; in particular, `grpcio` and
`portpicker`.

Bug Fixes

* Removes dependency on `tf.data.experimental.NestedStructure`.

Thanks to our Contributors

This release contains contributions from many people at Google, as well as:

Dheeraj R Reddy, Squadrick.

Major Features and Improvements

* Removed source level TF dependencies and switched from `tensorflow` to
`tf-nightly` dependency.
* Add support for `attr` module in TFF type system.
* Introduced new `tff.framework` interface layer.
* New AST transformations and optimizations.
* Preserve Python container usage in `tff.tf_computation`.

Bug Fixes

* Updated TFF model to reflect Keras `tf.keras.model.weights` order.
* Keras model with multiple inputs. 416

Major Features and Improvements

* New `tff.simulation.TransformingClientData` API and associated infinite
EMNIST dataset (see http://tensorflow.org/federated/api_docs/python/tff for
details)

Breaking Change

* Normalized `func` to `fn` across the repository (rename some parameters and
functions)

Bug Fixes

* Wrapped Keras models can now be used with
`tff.learning.build_federated_evaluation`
* Keras models with non-trainable variables in intermediate layers (e.g.
`BatchNormalization`) can be assigned back to Keras models with
`tff.learning.ModelWeights.assign_weights_to`

Breaking Changes

* Rename `tff.learning.federated_average` to `tff.learning.federated_mean`.
* Rename 'func' arguments to 'fn' throughout the API.

Bug Fixes

* Assorted fixes to typos in documentation and setup scripts.

Major Features and Improvements

* Updated to use TensorFlow version 1.13.1.
* Implemented Federated SGD in `tff.learning.build_federated_sgd_process()`.

Breaking Changes

* `next()` function of `tff.utils.IteratedProcess`s returned by
`build_federated_*_process()` no longer unwraps single value tuples (always
returns a tuple).

Bug Fixes

* Modify setup.py to require TensorFlow 1.x and not upgrade to 2.0 alpha.
* Stop unpacking single value tuples in `next()` function of objects returned
by `build_federated_*_process()`.
* Clear cached Keras sessions when wrapping Keras models to avoid referencing
stale graphs.

* Initial public release.