fl4health.mixins.core_protocols module¶

class FlexibleClientProtocol(*args, **kwargs)[source]¶

Bases: FlexibleClientProtocolPreSetup, Protocol

A minimal protocol for BasicClient focused on methods.

compute_evaluation_loss(preds, features, target)[source]¶

Return type:: EvaluationLosses

compute_training_loss(preds, features, target)[source]¶

Return type:: TrainingLosses

criterion: _Loss¶

initialize_all_model_weights(parameters, config)[source]¶

Return type:: None

model: Module¶

optimizers: dict[str, Optimizer]¶

predict_with_model(model, input)[source]¶

Return type:: tuple[dict[str, Tensor], dict[str, Tensor]]

test_loader: DataLoader | None¶

train_loader: DataLoader¶

transform_gradients(losses)[source]¶

Return type:: None

transform_target(target)[source]¶

Return type:: Tensor | dict[str, Tensor]

update_before_train(current_server_round)[source]¶

Return type:: None

val_loader: DataLoader¶

validate(include_losses_in_metrics=False)[source]¶

Return type:: tuple[float, dict[str, Union[bool, bytes, float, int, str]]]

class FlexibleClientProtocolPreSetup(*args, **kwargs)[source]¶

Bases: NumPyClientMinimalProtocol, Protocol

A minimal protocol for BasicClient focused on methods.

compute_loss_and_additional_losses(preds, features, target)[source]¶

Return type:: tuple[Tensor, dict[str, Tensor] | None]

device: device¶

get_criterion(config)[source]¶

Return type:: _Loss

get_data_loaders(config)[source]¶

Return type:: tuple[DataLoader, ...]

get_model(config)[source]¶

Return type:: Module

get_optimizer(config)[source]¶

Return type:: Optimizer | dict[str, Optimizer]

initialized: bool¶

setup_client(config)[source]¶

Return type:: None

class NumPyClientMinimalProtocol(*args, **kwargs)[source]¶

Bases: Protocol

A minimal protocol for NumPyClient with just essential methods.

evaluate(parameters, config)[source]¶

Return type:: tuple[float, int, dict[str, Union[bool, bytes, float, int, str]]]

fit(parameters, config)[source]¶

Return type:: tuple[list[ndarray[Any, dtype[Any]]], int, dict[str, Union[bool, bytes, float, int, str]]]

get_parameters(config)[source]¶

Return type:: list[ndarray[Any, dtype[Any]]]

set_parameters(parameters, config, fitting_round)[source]¶

Return type:: None

update_after_train(local_steps, loss_dict, config)[source]¶

Return type:: None