[SciPy-Dev] PR #196: simplification of optimization wrappers
Wed Apr 18 21:09:53 CDT 2012
On Wed, Apr 18, 2012 at 9:27 PM, Denis Laxalde <email@example.com> wrote:
> firstname.lastname@example.org wrote:
>> What's the overhead of retall, especially fmin with a few thousand
> The retall parameter is replaced by the field 'return_all' in the
> options dictionary which, if True, will lead to an 'allvecs' field in
> info. So there's no extra overhead as it remains optional.
good, I don't see any problem then.
>> The rest of info looks all calculated as a byproduct, so
>> there shouldn't be much extra cost, is there?
> No, I don't think so.
I just saw that the options use a mutable keyword, dict.
Are we running into problems? It might be safer to set it to None
instead of an empty dict, given that an empty dict doesn't make a more
informative signature either.
(I saw a reminder on this on planet python today.
> SciPy-Dev mailing list
More information about the SciPy-Dev