#15780: Increase Performance in Projective Morphism
-------------------------------------+-------------------------------------
Reporter: drose | Owner: drose
Type: enhancement | Status: needs_work
Priority: minor | Milestone: sage-6.2
Component: algebraic | Resolution:
geometry | Merged in:
Keywords: Projective, | Reviewers: Ben Hutz
Morphism | Work issues:
Authors: Dillon Rose | Commit:
Report Upstream: N/A | 6be7361b9186ede2ff230139c7972c9f38fc8f94
Branch: | Stopgaps:
u/bhutz/ticket/15780 |
Dependencies: #16051 |
-------------------------------------+-------------------------------------
Comment (by nbruin):
Replying to [comment:20 drose]:
> I used timeit on a random degree 3 map on P^3^ and it took 717
microseconds per loop. This is very fast and doesn't really warrant lazy
initialization.
Whether 717 ms is a lot depends on what you compare it to. I'm finding (on
6.0):
{{{
sage: P5=ProjectiveSpace(Rationals(),5)
sage: R=P2.coordinate_ring()
sage: mon=[R({tuple(a):1}) for a in
WeightedIntegerVectors(4,[1,1,1,1,1,1])]
sage: L=[sum(random_sublist(mon,0.1)) for j in [0..5]]
sage: %timeit P2.hom(L,P2)
10000 loops, best of 3: 63.3 us per loop
sage: %timeit [fast_callable(l) for l in L]
100 loops, best of 3: 1.94 ms per loop
}}}
as you see, creating the fast callables is *much* slower than creating the
map. If you increase the polynomials in L (by increasing the degree of the
monomials, for instance), the difference gets worse. I'd say that's enough
time penalty to hold off on it unless you actually need the
fast_callables.
--
Ticket URL: <http://trac.sagemath.org/ticket/15780#comment:22>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-trac.
For more options, visit https://groups.google.com/d/optout.
