[SciPy-user] leastsq not converging with tutorial example

Christian Kristukat ckkart at hoc.net
Wed Oct 26 15:54:12 CDT 2005


Hi Kael,

On Wed, 26 Oct 2005, Kael Fischer wrote:
> Here 2 scripts that I hope the mailer doesn't mangle.  Using either
> (one is a sine wave; the other is a double exponential), I get back
> the initial parameters and the "The cosine of the angle between
> func(x) and any column of the\n  Jacobian is at most 0.000000 in
> absolute value" status message.  Anybody know what is going on.

I tested your first example on my system and it works. But I also had some
runs where the best fit was not found. In any case the output parameters
were different from the initial guess. My guess is that your initial guess
is really bad. Try with some values closer to the exact model.

Christian



More information about the SciPy-user mailing list