ai.djl.training

Interface TrainingConfig

All Known Implementing Classes:

DefaultTrainingConfig

public interface TrainingConfig

An interface that is responsible for holding the configuration required by Trainer.

A Trainer requires an Initializer to initialize the parameters of the model, an Optimizer to compute gradients and update the parameters according to a Loss function. It also needs to know the Evaluators that need to be computed during training. A TrainingConfig instance that is passed to the Trainer will provide this information, and thus facilitate the training process.

Method Summary

Modifier and Type	Method and Description
DataManager	getDataManager() Gets the DataManager that computes data and labels from the output of dataset.
Device[]	getDevices() Gets the Device that are available for computation.
java.util.List<Evaluator>	getEvaluators() Returns the list of Evaluators that should be computed during training.
Initializer	getInitializer() Gets the Initializer to initialize the parameters of the model.
Loss	getLossFunction() Gets the Loss function to compute the loss against.
Optimizer	getOptimizer() Gets the Optimizer to use during training.
java.util.List<TrainingListener>	getTrainingListeners() Returns the list of TrainingListeners that should be used during training.

Method Detail

getDevices

Device[] getDevices()

Gets the Device that are available for computation.

This is necessary for a Trainer as it needs to know what kind of device it is running on, and how many devices it is running on.

Returns:

an array of Device

getInitializer

Initializer getInitializer()

Gets the Initializer to initialize the parameters of the model.

Returns:

an Initializer

getOptimizer

Optimizer getOptimizer()

Gets the Optimizer to use during training.

Returns:

an Optimizer

getLossFunction

Loss getLossFunction()

Gets the Loss function to compute the loss against.

Returns:

a Loss function

getDataManager

DataManager getDataManager()

Gets the DataManager that computes data and labels from the output of dataset.

Returns:

a DataManager

getEvaluators

java.util.List<Evaluator> getEvaluators()

Returns the list of Evaluators that should be computed during training.

Returns:

a list of Evaluators

getTrainingListeners

java.util.List<TrainingListener> getTrainingListeners()

Returns the list of TrainingListeners that should be used during training.

Returns:

a list of TrainingListeners