[SciPy-User] Why optimize.minimize returns worse solution?
Wed Apr 10 04:50:46 CDT 2013
I'm having a weird problem with scipy.optimize.minimize (using CG
method). Sometimes the function returns a solution which is worse than
the initial solution (in terms of the provided cost function). I don't
understand this, because I thought that the method returns the best
solution it has found, and at least the initial solution is not worse
than itself so it can always return that instead of a worse solution..
My gradient computations are incorrect at the moment, so that might be a
reason for this problem. However, I still don't understand why the
method returns a worse solution, even if the gradient computation is
wrong. (And I'm not sure whether this is even the reason, or one of the
Thanks for any help,
More information about the SciPy-User