Uses of Interface
org.nd4j.linalg.learning.regularization.Regularization

Packages that use Regularization

Package	Description
org.nd4j.autodiff.samediff
org.nd4j.linalg.learning.regularization

Uses of Regularization in org.nd4j.autodiff.samediff

Methods in org.nd4j.autodiff.samediff with parameters of type Regularization

Modifier and Type	Method and Description
TrainingConfig.Builder	TrainingConfig.Builder.addRegularization(Regularization... regularizations) Add regularization to all trainable parameters in the network
TrainingConfig.Builder	TrainingConfig.Builder.regularization(Regularization... regularization) Set the regularization for all trainable parameters in the network.

Method parameters in org.nd4j.autodiff.samediff with type arguments of type Regularization

Modifier and Type	Method and Description
TrainingConfig.Builder	TrainingConfig.Builder.regularization(List<Regularization> regularization) Set the regularization for all trainable parameters in the network.

Constructor parameters in org.nd4j.autodiff.samediff with type arguments of type Regularization

Constructor and Description
TrainingConfig(IUpdater updater, List<Regularization> regularization, boolean minimize, List<String> dataSetFeatureMapping, List<String> dataSetLabelMapping, List<String> dataSetFeatureMaskMapping, List<String> dataSetLabelMaskMapping, List<String> lossVariables) Create a training configuration suitable for training both single input/output and multi input/output networks. See also the TrainingConfig.Builder for creating a TrainingConfig
TrainingConfig(IUpdater updater, List<Regularization> regularization, boolean minimize, List<String> dataSetFeatureMapping, List<String> dataSetLabelMapping, List<String> dataSetFeatureMaskMapping, List<String> dataSetLabelMaskMapping, List<String> lossVariables, Map<String,List<IEvaluation>> trainEvaluations, Map<String,Integer> trainEvaluationLabels, Map<String,List<IEvaluation>> validationEvaluations, Map<String,Integer> validationEvaluationLabels)
TrainingConfig(IUpdater updater, List<Regularization> regularization, String dataSetFeatureMapping, String dataSetLabelMapping) Create a training configuration suitable for training a single input, single output network. See also the TrainingConfig.Builder for creating a TrainingConfig

Uses of Regularization in org.nd4j.linalg.learning.regularization

Classes in org.nd4j.linalg.learning.regularization that implement Regularization

Modifier and Type	Class and Description
class	L1Regularization L1 regularization: Implements updating as follows: L = loss + l1 * sum_i abs(w[i]) {@code w[i] -= updater(gradient[i] + l1 * sign(w[i])) - where sign(w[i]) is +/- 1 Note that L1 regularization is applied before the updater (Adam/Nesterov/etc) is applied.
class	L2Regularization L2 regularization: very similar to WeightDecay, but is applied before the updater is applied, not after.
class	WeightDecay WeightDecay regularization: Updater is not applied to the regularization term gradients, and (optionally) applies the learning rate.

Methods in org.nd4j.linalg.learning.regularization that return Regularization

Modifier and Type	Method and Description
Regularization	L2Regularization.clone()
Regularization	L1Regularization.clone()
Regularization	WeightDecay.clone()
Regularization	Regularization.clone()