KSLCore/ksl.app.config.optimization/SolverSpec/RSpline

RSpline

@Serializable

@SerialName(value = "rSpline")

data class RSpline(val startingPoint: Map<String, Double>? = null, val maxIterations: Int, val randomRestart: RandomRestartSpec? = null, val streamNum: Int = 0, val name: String? = null, val initialNumReps: Int, val sampleSizeGrowthRate: Double, val maxNumReplications: Int) : SolverSpec(source)

R-SPLINE. Mirrors ksl.simopt.solvers.algorithms.RSplineSolver.

R-SPLINE drives replications per evaluation through a ksl.simopt.solvers.FixedGrowthRateReplicationSchedule rather than a fixed-count, so this variant has no replicationsPerEvaluation: Int field. The growth schedule is configured by initialNumReps, sampleSizeGrowthRate, and maxNumReplications.

Constructors

RSpline

constructor(startingPoint: Map<String, Double>? = null, maxIterations: Int, randomRestart: RandomRestartSpec? = null, streamNum: Int = 0, name: String? = null, initialNumReps: Int, sampleSizeGrowthRate: Double, maxNumReplications: Int)

Properties

initialNumReps

val initialNumReps: Int

maxIterations

open override val maxIterations: Int

Maximum number of main-loop iterations the solver is permitted to run; must be positive when validated.

maxNumReplications

val maxNumReplications: Int

name

open override val name: String?

Optional human-readable solver instance name.

randomRestart

open override val randomRestart: RandomRestartSpec?

Optional random-restart wrapper. When non-null the solver factory wraps the chosen algorithm in a ksl.simopt.solvers.algorithms.RandomRestartSolver.

sampleSizeGrowthRate

val sampleSizeGrowthRate: Double

startingPoint

open override val startingPoint: Map<String, Double>?

Optional starting point for the search, keyed by decision-variable name. When null, the solver chooses its own starting point.

streamNum

open override val streamNum: Int

Random-number stream number used to seed the solver's stochastic decisions. 0 means "next available stream".