ActiveSetSequentialQuadraticProgrammingOptimizerComputeNextStep Method |
Computes the next optimization step that this parameter optimizer should take.
Namespace:
AGI.Foundation.NumericalMethods
Assembly:
AGI.Foundation.Core (in AGI.Foundation.Core.dll) Version: 24.3.420.0 (24.3.420.0)
Syntax public override MultivariableFunctionSolverStepResult<OptimizerMultivariableFunctionResults, OptimizerMultivariableFunctionDerivativeResults> ComputeNextStep(
double[] variableValues,
Matrix hessian,
ParameterOptimizerIterationResults previousIterationResults,
ITrackCalculationProgress progressTracker
)
Public Overrides Function ComputeNextStep (
variableValues As Double(),
hessian As Matrix,
previousIterationResults As ParameterOptimizerIterationResults,
progressTracker As ITrackCalculationProgress
) As MultivariableFunctionSolverStepResult(Of OptimizerMultivariableFunctionResults, OptimizerMultivariableFunctionDerivativeResults)
public:
virtual MultivariableFunctionSolverStepResult<OptimizerMultivariableFunctionResults^, OptimizerMultivariableFunctionDerivativeResults^>^ ComputeNextStep(
array<double>^ variableValues,
Matrix^ hessian,
ParameterOptimizerIterationResults^ previousIterationResults,
ITrackCalculationProgress^ progressTracker
) override
abstract ComputeNextStep :
variableValues : float[] *
hessian : Matrix *
previousIterationResults : ParameterOptimizerIterationResults *
progressTracker : ITrackCalculationProgress -> MultivariableFunctionSolverStepResult<OptimizerMultivariableFunctionResults, OptimizerMultivariableFunctionDerivativeResults>
override ComputeNextStep :
variableValues : float[] *
hessian : Matrix *
previousIterationResults : ParameterOptimizerIterationResults *
progressTracker : ITrackCalculationProgress -> MultivariableFunctionSolverStepResult<OptimizerMultivariableFunctionResults, OptimizerMultivariableFunctionDerivativeResults>
Parameters
- variableValues
- Type: SystemDouble
The current values of the variables. - hessian
- Type: AGI.Foundation.CoordinatesMatrix
An approximation of the second derivative
matrix of the Lagrangian. - previousIterationResults
- Type: AGI.Foundation.NumericalMethodsParameterOptimizerIterationResults
The results of the previous iteration
that are used to compute the next step. - progressTracker
- Type: AGI.FoundationITrackCalculationProgress
An optional progress tracker.
Return Value
Type:
MultivariableFunctionSolverStepResultOptimizerMultivariableFunctionResults,
OptimizerMultivariableFunctionDerivativeResultsThe step that the variables should take in this iteration.
See Also