#4466: fix det in linbox case to fail if proof=False isn't also set
----------------------------+-----------------------------------------------
Reporter: was | Owner: was
Type: defect | Status: new
Priority: minor | Milestone: sage-3.2
Component: linear algebra | Keywords:
----------------------------+-----------------------------------------------
See below:
{{{
On Fri, Nov 7, 2008 at 4:26 PM, Clement Pernet <[EMAIL PROTECTED]>
wrote:
> Hi all,
>
> After checking, the default algorithm called in LinBox for det over Z
> with dense matrices is the naive Chinese Remainder Theorem, with an
> early termination (if det stabilizes after 4 iterations in a row, then
> you stop picking random primes).
>
> So yes, it is probabilistic, Monte Carlo, and definitely not the fastest
> way to go (on my box, Sage's implementation, using IML's p-adic solver,
> is always faster).
>
> There is a p-adic implementation, with several additional tricks that
> Ana Urbanska and J-G Dumas have done, which was supposedely the fastest
> (btw I like the docstring in sage!!!).
>
> So far, I remember, removing it as a default option since it had some
> leaks, or bugs, and I was waiting for them to polish up the prototypical
> code. Another line in the v2.0 TODO list, I guess!
>
>
>
> William Stein a écrit :
>> Hi Clement,
>>
>> I tried this in Sage:
>>
>> time d = random_matrix(ZZ,n).det(proof=True)
>> time d = random_matrix(ZZ,n).det(algorithm='linbox')
>>
>> for various n (say around 200 - 500, etc.,) then linbox is dramatically
faster
>> than Sage. However, if I do
>>
>> time d = random_matrix(ZZ,n).det(proof=False)
>>
>> then Sage is similar in speed to linbox (actually about twice as fast
on OS X,
>> and close on sage.math).
>>
>> This strongly suggests to me that the following code in Linbox is
actually
>> *not* proving its result. Could you clarify?
>>
>> void linbox_integer_dense_det(mpz_t ans, mpz_t** matrix, size_t nrows,
>> size_t ncols) {
>> commentator.setMaxDetailLevel(0);
>> commentator.setMaxDepth (0);
>>
>> DenseMatrix<Integers> A(new_matrix_integers(matrix, nrows, ncols));
>> Integers::Element d;
>> det(d, A);
>> mpz_set(ans, spy.get_mpz(d));
>> }
>>
>>
>> William
>>
>>
>
>
--
William Stein
Associate Professor of Mathematics
University of Washington
http://wstein.org
}}}
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/4466>
Sage <http://sagemath.org/>
Sage - Open Source Mathematical Software: Building the Car Instead of
Reinventing the Wheel
--~--~---------~--~----~------------~-------~--~----~
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
-~----------~----~----~----~------~----~------~--~---
