tff.framework.local_executor_factory

See Stable See Nightly

Constructs an executor factory to execute computations locally.

tff.framework.local_executor_factory(
    num_clients=None, max_fanout=100, clients_per_thread=1, server_tf_device=None,
    client_tf_devices=tuple(), reference_resolving_clients=True
) -> tff.framework.ExecutorFactory

Note: The tff.federated_secure_sum() intrinsic is not implemented by this executor.

Args
num_clients	The number of clients. If specified, the executor factory function returned by local_executor_factory will be configured to have exactly num_clients clients. If unspecified (None), then the function returned will attempt to infer cardinalities of all placements for which it is passed values.
max_fanout	The maximum fanout at any point in the aggregation hierarchy. If num_clients > max_fanout, the constructed executor stack will consist of multiple levels of aggregators. The height of the stack will be on the order of log(num_clients) / log(max_fanout).
clients_per_thread	Integer number of clients for each of TFF's threads to run in sequence. Increasing clients_per_thread therefore reduces the concurrency of the TFF runtime, which can be useful if client work is very lightweight or models are very large and multiple copies cannot fit in memory.
server_tf_device	A tf.config.LogicalDevice to place server and other computation without explicit TFF placement.
client_tf_devices	List/tuple of tf.config.LogicalDevice to place clients for simulation. Possibly accelerators returned by tf.config.list_logical_devices().
reference_resolving_clients	Boolean indicating whether executors representing clients must be able to handle unplaced TFF lambdas.

Returns
An instance of executor_factory.ExecutorFactory encapsulating the executor construction logic specified above.

Raises
ValueError	If the number of clients is specified and not one or larger.