set_method

sherpa.astro.ui.set_method(meth)

Set the optimization method.

The primary task of Sherpa is to fit a model M(p) to a set of observed data, where the vector p denotes the model parameters. An optimization method is one that is used to determine the vector of model parameter values, p0, for which the chosen fit statistic is minimized.

Parameters

meth (str) – The name of the method (case is not important). The list_methods function returns the list of supported values.

Raises

sherpa.utils.err.ArgumentErr – If the meth argument is not recognized.

See also

get_method_name

Return the name of the current optimization method.

list_methods

List the supported optimization methods.

set_stat

Set the fit statistic.

Notes

The available methods include:

levmar

The Levenberg-Marquardt method is an interface to the MINPACK subroutine lmdif to find the local minimum of nonlinear least squares functions of several variables by a modification of the Levenberg-Marquardt algorithm 1.

moncar

The implementation of the moncar method is based on 2.

neldermead

The implementation of the Nelder Mead Simplex direct search is based on 3.

simplex

This is another name for neldermead.

References

1

J.J. More, “The Levenberg Marquardt algorithm: implementation and theory,” in Lecture Notes in Mathematics 630: Numerical Analysis, G.A. Watson (Ed.), Springer-Verlag: Berlin, 1978, pp.105-116.

2

Storn, R. and Price, K. “Differential Evolution: A Simple and Efficient Adaptive Scheme for Global Optimization over Continuous Spaces.” J. Global Optimization 11, 341-359, 1997.

3

Jeffrey C. Lagarias, James A. Reeds, Margaret H. Wright, Paul E. Wright “Convergence Properties of the Nelder-Mead Simplex Algorithm in Low Dimensions”, SIAM Journal on Optimization,Vol. 9, No. 1 (1998), pages 112-147.

Examples

>>> set_method('neldermead')