[SciPy-user] leastsq not converging with tutorial example

Kael Fischer kael.fischer at gmail.com
Wed Oct 26 16:30:26 CDT 2005


Christrian:

I wish that were the answer but, alas, no matter what starting
parameters you use the problem is the same.  The staring parameters in
those examples are very reasonable, and they should converge, as you
saw when you ran them.

Note that on my machines, it do not do the things minimizers typically do.

In particular:

It does not converge to a local minimum.
Nor does it run until the iteration limt is reached.

It exits with the error I gave, and with ier = 4.  According to the
docs, this indicates non-convergence.

Has any one seen _this_ behaviour before?  I suspect that something is
wrong with a library or with the the python wrappers in the FreeBSD
context.

Rgds,
Kael



--
Kael Fischer, Ph.D
DeRisi Lab - Univ. Of California San Francisco
415-514-4320



More information about the SciPy-user mailing list