public interface TrainingConfig
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 Evaluator
s 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.
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
Evaluator s 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
TrainingListener s that should be used during training. |
Device[] getDevices()
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.
Device
Initializer getInitializer()
Initializer
to initialize the parameters of the model.Initializer
Optimizer getOptimizer()
Optimizer
to use during training.Optimizer
Loss getLossFunction()
Loss
function to compute the loss against.Loss
functionDataManager getDataManager()
DataManager
that computes data and labels from the output of dataset.DataManager
java.util.List<Evaluator> getEvaluators()
Evaluator
s that should be computed during training.Evaluator
sjava.util.List<TrainingListener> getTrainingListeners()
TrainingListener
s that should be used during training.TrainingListener
s