Package org.ddogleg.optimization

Interface UnconstrainedLeastSquaresBase<S extends DMatrix,FunctionJacobian>

All Superinterfaces:

IterativeOptimization, Serializable, VerbosePrint

All Known Subinterfaces:

UnconstrainedLeastSquares<S>, UnconstrainedLeastSquaresSchur<S>

All Known Implementing Classes:

UnconLeastSqLevenbergMarquardt_F64, UnconLeastSqLevenbergMarquardtSchur_F64, UnconLeastSqTrustRegion_F64, UnconLeastSqTrustRegionSchur_F64

public interface UnconstrainedLeastSquaresBase<S extends DMatrix,FunctionJacobian>
extends IterativeOptimization

Common base for implementations of UnconstrainedLeastSquares that differ in the JacobianFunction

Method Summary

Modifier and Type	Method	Description
double	getFunctionValue()	Returns the value of the objective function being evaluated at the current parameters value.
Class<S>	getJacobianType()	Returns the Class used to store Jacobian matrices
double[]	getParameters()	After each iteration this function can be called to get the current best set of parameters.
void	initialize(double[] initial, double ftol, double gtol)	Specify the initial set of parameters from which to start from.
void	setFunction(FunctionNtoM function, FunctionJacobian jacobian)	Specifies a set of functions and their Jacobian.
void	setLoss(LossFunction loss, LossFunctionGradient lossGradient)	Specifies a specialized loss function, typically to improve robustness to outliers.
void	setPostUpdate(AdjustArray_F64 adjuster)	After a step has been computed and applied this function is called with the new candidate state passed in.

Methods inherited from interface org.ddogleg.optimization.IterativeOptimization

isConverged, isUpdated, iterate, setVerbose

Methods inherited from interface org.ddogleg.struct.VerbosePrint

setVerbose

Method Details

setFunction

void setFunction(FunctionNtoM function,
 @Nullable
 FunctionJacobian jacobian)

Specifies a set of functions and their Jacobian. See class description for documentation on output data format.

Parameters:

function - Computes the output of M functions f_i(x) which take in N fit parameters as input.

jacobian - Computes the Jacobian of the M functions. If null a numerical Jacobian will be used.

setLoss

void setLoss(LossFunction loss,
 LossFunctionGradient lossGradient)

Specifies a specialized loss function, typically to improve robustness to outliers. Squared error is the default.

setPostUpdate

void setPostUpdate(AdjustArray_F64 adjuster)

After a step has been computed and applied this function is called with the new candidate state passed in. It provides the user the opportunity to adjust the data in anyway which will not change the score. For example, if a vector is scale invariant you can ensure it always has a norm of 1.

Parameters:

adjuster - The new user provided adjuster

initialize

void initialize(double[] initial,
 double ftol,
 double gtol)

Specify the initial set of parameters from which to start from. Call after setFunction(org.ddogleg.optimization.functions.FunctionNtoM, FunctionJacobian) has been called.

Parameters:

initial - Initial parameters or guess with N elements..

ftol - Relative threshold for change in function value between iterations. 0 ≤ ftol ≤ 1. Try 1e-12

gtol - Absolute threshold for convergence based on the gradient's norm. 0 disables test. 0 ≤ gtol. Try 1e-12

getParameters

double[] getParameters()

After each iteration this function can be called to get the current best set of parameters.

getFunctionValue

double getFunctionValue()

Returns the value of the objective function being evaluated at the current parameters value. If not supported then an exception is thrown.

Returns:

Objective function's value.

getJacobianType

Class<S> getJacobianType()

Returns the Class used to store Jacobian matrices