CLASS

SyftPlan

Contents

• Methods
  • execute(trainingData:validationData:clientConfig:)
  • generateDiffData()

public class SyftPlan

Holds the training script to be used for training your data and generating diffs

Methods

execute(trainingData:validationData:clientConfig:)

@discardableResult public func execute<T, U>(trainingData: TrainingData<T>, validationData: ValidationData<U>, clientConfig: FederatedClientConfig) -> Float

Executes the model received from PyGrid on your training and validation data. Loop through your entire training data and call this method to update the model parameters received from PyGrid.

• Parameters:
  • trainingData: tensor data used for training
  • validationData: tensor data used for validation
  • clientConfig: contains training parameters (batch size and learning rate)

Parameters

Name	Description
trainingData	tensor data used for training
validationData	tensor data used for validation
clientConfig	contains training parameters (batch size and learning rate)

generateDiffData()

public func generateDiffData() throws -> Data

Calculates difference between the original model parameters received from PyGrid and the updated model parameters generated from running this plan on your training and validation data. This diff data will be passed to the model report closure to send it to PyGrid.