On Nov 27, 11:38 am, Simon King <[EMAIL PROTECTED]> wrote:
> Dear Michael,
Hi Simon,
> On Nov 27, 8:26 pm, mabshoff <[EMAIL PROTECTED]
>
> dortmund.de> wrote:
> > Yes.
>
> Finally I understood something. By the way, it seems to me that this
> detail is not described in the documentation (of Sage :)
Did someone just volunteer to update/write the doctesting section of
the Sage development manual? :)
> > > What can I do to change it?
>
> > You doctesting example.
>
> Sniff :(
>
> But a minute ago, it seemed to me I have a solution.
> Instead of
> ... loops, best of ...: ... per loop
> I now write
> ... loops, best of ...: ... per loop
> with one additional blank in the beginning. Then the doc test passes!
Interesting trick. But with Sage 3.2 or higher you can actually get
the performance data back from timeit - see
http://groups.google.com/group/sage-devel/browse_thread/thread/b91c51672ae0f475
about the Integral testing code which uses that technique.
> > Re the timeit example: I would think it is pointless that you run that
> > code since you won't get any results to compare anyway. Once we have
> > the code that compares different doctest timeit runs you should use
> > that. Since that code doesn't exist no one would be sad if you even
> > wrote it :)
>
> I think I might provide the timing *with* the figures -- I compare two
> methods, and the figures show that one of them is clearly better. But
> I write
> sage: timeit('blah') # somehow random, since it is machine dependent
> so that the output doesn't count for the test, but is visible to the
> user.
Ok. See my comment above about actually being able to compare the
results in relative terms.
> Cheers,
> Simon
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
