TensorFlow 2.0 Beta is available Learn more

Module: tff.learning

The public API for model developers using federated learning algorithms.

Defined in python/learning/__init__.py.

Modules

framework module: The public API for contributors who develop federated learning algorithms.

Classes

class BatchOutput: A structure that holds the output of a tff.learning.Model.

class Model: Represents a model for use in TensorFlow Federated.

class TrainableModel: A Model with an additional method for (local) training.

Functions

assign_weights_to_keras_model(...): Assigns a nested structure of TFF weights to a Keras model.

build_federated_averaging_process(...): Builds the TFF computations for optimization using federated averaging.

build_federated_evaluation(...): Builds the TFF computation for federated evaluation of the given model.

build_federated_sgd_process(...): Builds the TFF computations for optimization using federated SGD.

from_compiled_keras_model(...): Builds a tff.learning.Model for an example mini batch.

from_keras_model(...): Builds a tff.learning.Model for an example mini batch.

state_with_new_model_weights(...): Returns a ServerState with updated model weights.