View source on GitHub
|
Libraries for using Federated Learning algorithms.
Modules
framework module: Libraries for developing Federated Learning algorithms.
Classes
class BatchOutput: A structure that holds the output of a tff.learning.Model.
class ClientFedAvg: Client TensorFlow logic for Federated Averaging.
class ClientWeighting: Enum for built-in methods for weighing clients.
class Model: Represents a model for use in TensorFlow Federated.
class ModelWeights: A container for the trainable and non-trainable variables of a Model.
Functions
build_federated_averaging_process(...): Builds an iterative process that performs 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.
build_personalization_eval(...): Builds the TFF computation for evaluating personalization strategies.
compression_aggregator(...): Creates aggregator with compression and adaptive zeroing and clipping.
dp_aggregator(...): Creates aggregator with adaptive zeroing and differential privacy.
federated_aggregate_keras_metric(...): Aggregates variables a keras metric placed at CLIENTS to SERVER.
from_keras_model(...): Builds a tff.learning.Model from a tf.keras.Model.
robust_aggregator(...): Creates aggregator for mean with adaptive zeroing and clipping.
secure_aggregator(...): Creates secure aggregator with adaptive zeroing and clipping.
state_with_new_model_weights(...): Returns a ServerState with updated model weights.