Dear Tim,
On Tue, Jan 12, 2010 at 10:58:23PM +0800, Tim Joseph Dumol wrote:
> I've been wondering why sage.combinat.combinat.combinations still exists,
> when sage.combinat.combination.Combinations is much faster, being native
> Sage code, and thus not requiring GAP interfacing (also more reliable, I
> suppose). The huge speed differnce can be seen here:
>
> {{{
> sage: timeit('[Combinations(range(i), j).list() for (i, j) in
> CartesianProduct(range(15), range(15))]')
> 5 loops, best of 3: 449 ms per loop
>
> sage: timeit('[combinations(range(i), j) for (i, j) in
> CartesianProduct(range(15), range(15))]')
> 5 loops, best of 3: 4.17 s per loop
> }}}
>
> `combinations` is 10x slower than `Combinations`. If compatibility needs
> to be maintained, `combinations(mset,k)` can be defiend as an alias to
> `Combinations(mset,k).list`.
Thanks for the report!
Could you please open a track ticket and send us a patch, with an
alias as you suggest, and possibly a deprecation warning? Note: the
following (dubious) feature advertised in 'combinations?' might be
vaguely tricky to emulate:
sage: mset = ["d","a","v","i","d"]
sage: combinations(mset,3)
['add', 'adi', 'adv', 'aiv', 'ddi', 'ddv', 'div']
Thanks in advance!
Best,
Nicolas
--
Nicolas M. Thiéry "Isil" <nthi...@users.sf.net>
http://Nicolas.Thiery.name/
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
