strategy

ServerApp strategies.

Classes

Bulyan([fraction_train, fraction_evaluate, ...])

Bulyan 策略。

DifferentialPrivacyClientSideAdaptiveClipping(...)

Strategy wrapper for central DP with client-side adaptive clipping.

DifferentialPrivacyClientSideFixedClipping(...)

Strategy wrapper for central DP with client-side fixed clipping.

DifferentialPrivacyServerSideAdaptiveClipping(...)

Message-based central DP with server-side adaptive clipping.

DifferentialPrivacyServerSideFixedClipping(...)

Strategy wrapper for central DP with server-side fixed clipping.

FedAdagrad(*[, fraction_train, ...])

FedAdagrad 策略 - 使用 Adagrad 进行自适应联合优化。

FedAdam(*[, fraction_train, ...])

FedAdam - 使用 Adam 进行自适应联合优化。

FedAvg([fraction_train, fraction_evaluate, ...])

联邦平均策略。

FedAvgM([fraction_train, fraction_evaluate, ...])

联邦平均动量策略。

FedMedian([fraction_train, ...])

Federated Median (FedMedian) strategy.

FedProx([fraction_train, fraction_evaluate, ...])

联邦优化策略。

FedTrimmedAvg([fraction_train, ...])

带修剪均值的联邦平均法[Dong Yin 等，2021]。

FedXgbBagging([fraction_train, ...])

Configurable FedXgbBagging strategy implementation.

FedXgbCyclic([fraction_train, ...])

Configurable FedXgbCyclic strategy implementation.

FedYogi(*[, fraction_train, ...])

FedYogi [Reddi 等人，2020] 策略。

Krum([fraction_train, fraction_evaluate, ...])

Krum [Blanchard et al., 2017] strategy.

MultiKrum([fraction_train, ...])

MultiKrum [Blanchard et al., 2017] strategy.

QFedAvg(client_learning_rate[, q, ...])

Q-FedAvg strategy.

Result(arrays, train_metrics_clientapp, ...)

Data class carrying records generated during the execution of a strategy.

Strategy()

服务器策略实现的抽象基类。