Class ElasticNetRegularizer

Combines L1 and L2 regularization into a single elastic net regularizer.

public class ElasticNetRegularizer : CompositeRegularizer, IRegularizer

Inherited Members: CompositeRegularizer.ReportPenaltyInLoss

CompositeRegularizer.ApplyGradient<T>(T, T)

CompositeRegularizer.ReportPenalty<T>(T)

CompositeRegularizer.UpdateParams<T>(T)

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

Remarks

Uses separate L1 and L2 regularizers (L1Regularizer and L2Regularizer) and applies both penalties during training.

Combines L1 and L2 regularization into a single elastic net regularizer.

public ElasticNetRegularizer(float l1, float l2, bool reportPenaltyInLoss = false)

l1 float: L1 regularization coefficient.
l2 float: L2 regularization coefficient.
reportPenaltyInLoss bool: Indicates whether the regularization penalty is included in the reported loss value.

Uses separate L1 and L2 regularizers (L1Regularizer and L2Regularizer) and applies both penalties during training.

public float L1 { get; }

public float L2 { get; }

Returns a string that represents the current object.

public override string ToString()