Problem with the posted example using lsqcurvefit?

Question

Ana Cadavid 2015-5-29

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/221028-problem-with-the-posted-example-using-lsqcurvefit

评论： Star Strider 2015-5-29

I have tried exactly the example given in http://www.mathworks.com/help/optim/ug/lsqcurvefit.html

I get the following error

Trial>> [x,resnorm] = lsqcurvefit(@myfun,x0,xdata,ydata);

Local minimum possible.

lsqcurvefit stopped because the final change in the sum of squares relative to its initial value is less than the default value of the function tolerance.

criteria details

Optimization stopped because the relative sum of squares (r) is changing by less than options.TolFun = 1.000000e-06.

Optimization Metric Options relative change r = 9.32e-08 TolFun = 1e-06 (default)

If I try to change the TolFun in the options I start a chain requesting more iterations and more evaluations that do not stop..

Any idea what is failing? In the web it states that it converges in 27 iterations.

Thank you

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Star Strider 2015-5-29

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/221028-problem-with-the-posted-example-using-lsqcurvefit#answer_180954

That’s a Warning, not an Error. I suspect it throws that Warning if the residual norm is higher than some predetermined threshold when the function converges. In other circumstances, it would mean that experimenting with different initial parameter estimates would be appropriate.