fl4health.servers.instance_level_dp_server module

class InstanceLevelDpServer(client_manager, fl_config, noise_multiplier, batch_size, num_server_rounds, strategy, local_epochs=None, local_steps=None, checkpoint_and_state_module=None, reporters=None, delta=None, on_init_parameters_config_fn=None, server_name=None, accept_failures=True)[source]

Bases: FlServer

__init__(client_manager, fl_config, noise_multiplier, batch_size, num_server_rounds, strategy, local_epochs=None, local_steps=None, checkpoint_and_state_module=None, reporters=None, delta=None, on_init_parameters_config_fn=None, server_name=None, accept_failures=True)[source]

Server to be used in case of Instance Level Differential Privacy with Federated Averaging. Modified the fit function to poll clients for sample counts prior to the first round of FL.

Parameters:

  • client_manager (ClientManager) – Determines the mechanism by which clients are sampled by the server, if they are to be sampled at all.

  • fl_config (Config) – This should be the configuration that was used to setup the federated training. In most cases it should be the “source of truth” for how FL training/evaluation should proceed. For example, the config used to produce the on_fit_config_fn and on_evaluate_config_fn for the strategy. NOTE: This config is DISTINCT from the Flwr server config, which is extremely minimal.

  • noise_multiplier (float) – The amount of Gaussian noise to be added to the per sample gradient during DP-SGD.

  • batch_size (int) – The batch size to be used in training on the client-side. Used in privacy accounting.

  • num_server_rounds (int) – The number of server rounds to be done in FL training. Used in privacy accounting

  • strategy (BasicFedAvg) – The aggregation strategy to be used by the server to handle client updates and other information potentially sent by the participating clients. this must be an OpacusBasicFedAvg strategy to ensure proper treatment of the model in the Opacus framework

  • local_epochs (int | None, optional) – Number of local epochs to be performed on the client-side. This is used in privacy accounting. One of local_epochs or local_steps should be defined, but not both. Defaults to None.

  • local_steps (int | None, optional) – Number of local steps to be performed on the client-side. This is used in privacy accounting. One of local_epochs or local_steps should be defined, but not both. Defaults to None.

  • checkpoint_and_state_module (OpacusServerCheckpointAndStateModule | None, optional) – This module is used to handle both model checkpointing and state checkpointing. The former is aimed at saving model artifacts to be used or evaluated after training. The latter is used to preserve training state (including models) such that if FL training is interrupted, the process may be restarted. If no module is provided, no checkpointing or state preservation will happen. Defaults to None.

  • reporters (Sequence[BaseReporter] | None, optional) – A sequence of FL4Health reporters which the client should send data to.

  • delta (float | None, optional) – The delta value for epsilon-delta DP accounting. If None it defaults to being 1/total_samples in the FL run. Defaults to None.

  • on_init_parameters_config_fn (Callable[[int], dict[str, Scalar]] | None, optional) – Function used to configure how one asks a client to provide parameters from which to initialize all other clients by providing a Config dictionary. If this is none, then a blank config is sent with the parameter request (which is default behavior for flower servers). Defaults to None.

  • server_name (str | None, optional) – An optional string name to uniquely identify server. This name is also used as part of any state checkpointing done by the server. Defaults to None.

  • accept_failures (bool, optional) – Determines whether the server should accept failures during training or evaluation from clients or not. If set to False, this will cause the server to shutdown all clients and throw an exception. Defaults to True.

fit(num_rounds, timeout)[source]

Run federated averaging for a number of rounds.

Parameters:

  • num_rounds (int) – Number of server rounds to run.

  • timeout (float | None) – The amount of time in seconds that the server will wait for results from the clients selected to participate in federated training.

Returns:

The first element of the tuple is a history object containing the full

set of FL training results, including things like aggregated loss and metrics. Tuple also includes elapsed time in seconds for round.

Return type:

tuple[History, float]

setup_privacy_accountant(sample_counts)[source]

Sets up FL Accountant and computes privacy loss based on class attributes and retrieved sample counts.

Parameters:

sample_counts (list[int]) – These should be the total number of training examples fetched from all clients during the sample polling process.

Return type:

None