#17561: Speed up vectors
----------------------------------+------------------------
Reporter: jdemeyer | Owner:
Type: enhancement | Status: new
Priority: minor | Milestone: sage-6.5
Component: performance | Resolution:
Keywords: | Merged in:
Authors: Jeroen Demeyer | Reviewers:
Report Upstream: N/A | Work issues:
Branch: | Commit:
Dependencies: | Stopgaps:
----------------------------------+------------------------
Description changed by jdemeyer:
Old description:
> We should declare the types `_entries` in `free_module_element.pxd`. This
> leads to a significant speed-up in the sparse case, and a minor speedup
> in the dense case:
>
> Before:
> {{{
> sage: v = vector(RR, range(10000)); timeit('v._add_(v)', repeat=10,
> number=1000)
> 1000 loops, best of 10: 1.85 ms per loop
> sage: v = vector(RR, range(10000), sparse=True); timeit('v._add_(v)',
> repeat=10, number=1000)
> 1000 loops, best of 10: 3.49 ms per loop
> }}}
>
> After:
> {{{
> sage: v = vector(RR, range(10000)); timeit('v._add_(v)', repeat=10,
> number=1000)
> 1000 loops, best of 10: 1.82 ms per loop
> sage: v = vector(RR, range(10000), sparse=True); timeit('v._add_(v)',
> repeat=10, number=1000)
> 1000 loops, best of 10: 2.85 ms per loop
> }}}
New description:
1. Currently, every call to `vector()` computes a basis, only to ask for
the ring over which the basis is defined (see #11751). We avoid this by
introducing a new method `basis_ring()` for free modules.
Before:
{{{
sage: L = [RR(x) for x in range(10000)]
sage: timeit('vector(RR, L)')
5 loops, best of 3: 77.4 ms per loop
}}}
After:
{{{
sage: L = [RR(x) for x in range(10000)]
sage: timeit('vector(RR, L)')
125 loops, best of 3: 1.82 ms per loop
}}}
------
2. We should declare the types `_entries` in `free_module_element.pxd`.
This leads to a significant speed-up in the sparse case, and a minor
speedup in the dense case:
Before:
{{{
sage: v = vector(RR, range(10000)); timeit('v._add_(v)', repeat=10,
number=1000)
1000 loops, best of 10: 1.85 ms per loop
sage: v = vector(RR, range(10000), sparse=True); timeit('v._add_(v)',
repeat=10, number=1000)
1000 loops, best of 10: 3.49 ms per loop
}}}
After:
{{{
sage: v = vector(RR, range(10000)); timeit('v._add_(v)', repeat=10,
number=1000)
1000 loops, best of 10: 1.82 ms per loop
sage: v = vector(RR, range(10000), sparse=True); timeit('v._add_(v)',
repeat=10, number=1000)
1000 loops, best of 10: 2.85 ms per loop
}}}
--
--
Ticket URL: <http://trac.sagemath.org/ticket/17561#comment:2>
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.
