Watch talks from the 2019 TensorFlow Dev Summit Watch now

Module: tff.learning.framework

Defined in learning/framework/

The public API for contributors who develop federated learning algorithms.


class ClientDeltaFn: Represents a client computation that produces an update to a model.

class ClientOutput: ClientOutput(weights_delta, weights_delta_weight, model_output, optimizer_output)

class EnhancedModel: A wrapper around a Model that adds sanity checking and metadata helpers.

class EnhancedTrainableModel

class ModelWeights: A container for the trainable and non-trainable variables of a Model.


build_model_delta_optimizer_process(...): Constructs tff.utils.IterativeProcess for Federated Averaging or SGD.