kumulant/com.eignex.kumulant.stat.regression.glm/Penalty

Penalty

@Serializable

Regularisation knob shared across regression stats. The mathematical effect is always the same; the mechanics differ per host: a closed-form read()-time projection in UnivariateRegressionStat, lazy multiplicative scaling / truncated-gradient in StochasticRegressionStat, per-coordinate proximal step (L1) or gradient term (L2) in DiagonalRegressionStat. See each stat's KDoc.

Inheritors

None

Types

@Serializable

@SerialName(value = "L1")

data class L1(val lambda: Double) : Penalty

L1 / Lasso: drives sparsity in the fitted weights.

@Serializable

@SerialName(value = "L2")

data class L2(val lambda: Double) : Penalty

L2 / Ridge: shrinks the fitted weights toward zero.

None

@Serializable

@SerialName(value = "None")

data object None : Penalty

No regularisation.