[SciPy-Dev] parameters of fitting data

James Phillips zunzun@zunzun....
Thu Nov 24 10:54:19 CST 2011


The ODR code I gave does not fit data, rather it only calculates the
desired statistics for previously fitted data.  So you can separately fit
data and calculate parameter statistics using that code.

     James

On Thu, Nov 24, 2011 at 10:27 AM, Krzysztof Berniak <
krzysztof.berniak@gmail.com> wrote:

> Hi,
>
> Thanks for your answer. Your idea works, but not im my example.
> When I did example on this page:
>
> http://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.curve_fit.htmleverythink works done. I use sqrt(np.diag(cov)) and I get the standard
> errors.
> But when I use this in my code
>
> func = lambda z,a,b: a*np.exp(z*b)
> popt, pcov = optimize.curve_fit(func, z,f, maxfev = 1000)
> errors = sqrt(np.diag(pcov))
>
> It displays:
> RuntimeError: Optimal parameters not found: Number of calls to function
> has reached maxfev = 1000.
> When I increase value of maxfev I have the same problem.
>
> regards and please help,
> Cristopher
>
>
> _______________________________________________
> SciPy-Dev mailing list
> SciPy-Dev@scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.scipy.org/pipermail/scipy-dev/attachments/20111124/a59c21a1/attachment.html 


More information about the SciPy-Dev mailing list