#11900: Serious regression caused by #9138
-----------------------------------------------------------------------------------+
Reporter: SimonKing
| Owner: tbd
Type: defect
| Status: needs_work
Priority: critical
| Milestone: sage-4.8
Component: performance
| Keywords: categories regression
Work_issues: finalize the patch with the different ideas that have been
explored | Upstream: N/A
Reviewer: Jeroen Demeyer, Nicolas M. Thiéry
| Author: Simon King
Merged:
| Dependencies: #9138 #11911
-----------------------------------------------------------------------------------+
Comment(by SimonKing):
For the record: With the latest patches, one gets some doctest errors, but
they seem managable:
{{{
sage -t devel/sage-
main/sage/rings/polynomial/polynomial_quotient_ring.py # 6 doctests failed
sage -t devel/sage-main/sage/categories/pushout.py # 1 doctests
failed
sage -t devel/sage-
main/sage/rings/polynomial/polynomial_quotient_ring_element.py # 6
doctests failed
sage -t devel/sage-main/sage/combinat/debruijn_sequence.pyx # 21
doctests failed
sage -t devel/sage-main/sage/misc/constant_function.pyx # 2
doctests failed
sage -t devel/sage-main/sage/categories/category_singleton.pyx #
1 doctests failed
}}}
Concerning timings, I find with sage-4.7.2+#9138+the stuff from here the
following results for the benchmarks that I posted five weeks ago. By
"vanilla sage", I mean sage-4.7.2.alpha2 unpatched (copying the timings
from above):
__Creation of Matrix spaces__
It is as with the previous patch versions, hence, faster than vanilla sage
sage (0.38 s cpu):
{{{
sage: def test():
....: for i in xrange(5000):
....: MS = MatrixSpace(QQ,i)
....:
sage: %time test()
CPU times: user 0.25 s, sys: 0.01 s, total: 0.26 s
Wall time: 0.26 s
}}}
__Creation of finite fields__
It is faster than with the previous patches. Even better, it is now faster
than vanilla sage (0.60 s cpu), which has not been the case with the
previous patches:
{{{
sage: def test():
....: for p in prime_range(10000):
....: P = GF(p)
....:
sage: %time test()
CPU times: user 0.53 s, sys: 0.00 s, total: 0.53 s
Wall time: 0.53 s
}}}
__Creation of polynomial rings__
It is slower than vanilla sage (2.41 s cpu), but a bit faster than what I
had with the patches 5 weeks ago (4.32 s CPU):
{{{
sage: def test():
....: for p in prime_range(10000):
....: P = GF(p)['x']
....: P = GF(p)['x','y']
....:
sage: %time test()
CPU times: user 3.92 s, sys: 0.05 s, total: 3.97 s
Wall time: 3.98 s
}}}
'''__The two critical elliptic curve tests__'''
Recall that the following two were most critical and were the reason for
creating this ticket:
{{{
sage: %time L = EllipticCurve('960d1').prove_BSD()
CPU times: user 3.96 s, sys: 0.08 s, total: 4.04 s
Wall time: 4.25 s
}}}
That's about as fast as vanilla sage (CPU time went up a bit, wall time
went down a bit)
{{{
sage: def test():
....: E = EllipticCurve('389a')
....: for p in prime_range(10000):
....: if p != 389:
....: G = E.change_ring(GF(p)).abelian_group()
....:
sage: %time test()
CPU times: user 16.53 s, sys: 0.14 s, total: 16.67 s
Wall time: 16.72 s
}}}
Again, that can very well compete with vanilla sage (16.57 s cpu, 16.77 s
wall).
'''__TODO__'''
Instead of `P.is_ring()`, define `_Rings = Rings()` and test `P in
_Rings`. Nicolas, please tell whether I forgot further things todo on this
ticket.
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/11900#comment:114>
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 post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
