On Mar 14, 2011, at 6:50 AM, Alexander Riess wrote:
One thing that would be helpful would be if someone tried Alexander's
test problem in fmin_slsqp from SciPy, which also uses the SLSQP
code,
to see if a similar problem (albeit probably with different symptoms)
occurs there.
I tried to solve the test-problem min -s*x on [-1.0,1.0] with SciPy.
The
summary, program and results are listed bellow.
If 1.0 <= s <= 100 everything works fine: Optimization terminated
successfully. If s > 1.0E2 the result is: Positive directional
derivative for
linesearch. If s > 1.0E5 the calculated value of xmin is wrong: [0.0].
Thanks for checking on this; this confirms that the problem is in the
original SLSQP, and not in some mistake I made in porting it.
Steven
_______________________________________________
NLopt-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/nlopt-discuss
