NLFitLevenbergMarquardtFit(NLFitLMFunction, Double, Double, Double, Double, Double, Int32, Double, Double, Int32, Double, Int32, Int32, Int32, Double, Int32, Int32, Double, Double, Double, Double, Double, CancellationToken) Method |
Most general interface to the Levenberg-Marquardt algorithm.
Namespace: Altaxo.Calc.RegressionAssembly: AltaxoCore (in AltaxoCore.dll) Version: 4.8.3448.0 (4.8.3448.0)
Syntaxpublic static void LevenbergMarquardtFit(
NLFitLMFunction fcn,
double[] xvec,
double[] fvec,
double ftol,
double xtol,
double gtol,
int maxfev,
double epsfcn,
double[] diag,
int mode,
double factor,
int nprint,
ref int info,
ref int nfev,
double[] fjac,
int ldfjac,
int[] ipvt,
double[] qtf,
double[] wa1,
double[] wa2,
double[] wa3,
double[] wa4,
CancellationToken cancellationToken
)
Parameters
- fcn NLFitLMFunction
- The user-supplied function that provides values to minimize.
- xvec Double
- Parameter vector (input: initial estimate; output: final estimate).
- fvec Double
- Output array that receives the evaluated function values at the output xvec.
- ftol Double
- Tolerance on the relative reduction of the sum of squares.
- xtol Double
- Tolerance on the relative change in the solution.
- gtol Double
- Tolerance on the orthogonality between the function vector and the Jacobian columns.
- maxfev Int32
- Maximum number of function evaluations.
- epsfcn Double
-
Step-length control for the forward-difference Jacobian approximation. If less than machine precision, machine precision is assumed.
- diag Double
- Scaling factors for the variables (depending on mode).
- mode Int32
- Scaling mode for variables (1: internal scaling; 2: use diag).
- factor Double
- Initial step bound scaling factor.
- nprint Int32
- Controls calling fcn with iflag = 0 for printing iterates.
- info Int32
- Output status code.
- nfev Int32
- On output, receives the number of calls to fcn.
- fjac Double
- Jacobian array (m-by-n, column-major). On output contains the final QR information.
- ldfjac Int32
- Leading dimension of fjac.
- ipvt Int32
- Pivot indices defining permutation matrix.
- qtf Double
- The first n elements of (Qᵀ)f.
- wa1 Double
- Working array of length n.
- wa2 Double
- Working array of length n.
- wa3 Double
- Working array of length n.
- wa4 Double
- Working array of length m.
- cancellationToken CancellationToken
- Token to cancel the fit.
Remarks
This overload provides full control over the minimization process and auxiliary storage allocation.
Prefer one of the simpler interfaces above unless you need that level of control.
See Also