Package org.ddogleg.optimization

Interface OptimizationDerivative<State>

public interface OptimizationDerivative<State>
Interface for computing the gradient of a set of functions given a set of model parameters. Before computeDerivative(Object, double[][]) is called, the model must be set using setModel(double[]).

  • Method Summary

    Modifier and Type
    Method
    Description
    boolean
    computeDerivative(State state, double[][] gradient)
    Computes the gradient for each function with respect to model parameters.
    void
    setModel(double[] model)
    Specifies the current model parameters around which the gradient is computed.

  • Method Details

    • setModel

      void setModel(double[] model)
      Specifies the current model parameters around which the gradient is computed.
      Parameters:
      model - Model parameters.

    • computeDerivative

      boolean computeDerivative(State state, double[][] gradient)

      Computes the gradient for each function with respect to model parameters. The derivative is a 2D array. The first axis is for each function and the second for each model parameter:

      derivative[i][j] = ∂ fi / ∂ pj

      Parameters:
      state - State of the system being examined.
      gradient - Gradient with respect to the current model parameters,
      Returns:
      true if successful or false if it failed.