On Dec 17, 2009, at 5:58 PM, Christian Szegedy wrote:
> It is impossible to come up with any reasonable explanation for this
> kind of slowdown. Even if you do extremely stupid things like
> summing all permutations and simplifying the expression at the end,
> you
> can't get that slow.
No, but
On Thu, Dec 17, 2009 at 5:58 PM, Christian Szegedy
wrote:
> It is impossible to come up with any reasonable explanation for this
> kind of slowdown. Even if you do extremely stupid things like
> summing all permutations and simplifying the expression at the end, you
> can't get that slow.
>
> Addi
It is impossible to come up with any reasonable explanation for this
kind of slowdown. Even if you do extremely stupid things like
summing all permutations and simplifying the expression at the end, you
can't get that slow.
Additionally, you cansee that the inverse is computed readily. If you
loo
The speed could be do to the inefficiency of fraction field arithmetic
over the polynomial ring. Ideally, we should have fraction-free
gaussian elimination. Also, easily invertable/small determinant may
actually be worse--as it could be creating a lot of large intermediate
values with non-t
You evaluate it over ZZ[x1,...,xn] rather than GF(2)[x1,...,x4].
Anyways, it simply can't be *that* slow in any case: even: the
(theoretically ) maximum number of monoms that can be in any
expansion is less than a few thousands, so the upper limit
for a naively implemented Gaussian elimination is
On 17 dic, 11:48, "ma...@mendelu.cz" wrote:
> And another observation:
>
> maxima returns answer immediatelly (with a lag necessary to start
> maxima)
> m is the original matrix from x.py
>
> sage: m._maxima_().determinant().expand().sage()
> x0^2*x2^2*x3^2*x7^2 - 2*x0*x1*x2*x3*x4*x5*x6*x7 + x1^
And another observation:
maxima returns answer immediatelly (with a lag necessary to start
maxima)
m is the original matrix from x.py
sage: m._maxima_().determinant().expand().sage()
x0^2*x2^2*x3^2*x7^2 - 2*x0*x1*x2*x3*x4*x5*x6*x7 + x1^2*x4^2*x5^2*x6^2
Anyway, the answer is different from expec
perhaps problems expanding polynomials? even determinant of submatrix
(0,0,5,5) is suprisingly slow.
workaroud is to replace polynomials in your matrix by variables.
var('x0 x1 x2 x3 x4 x5 x6 x7 a1 a2 a3 a4 a5 a6 a7 b1 b2 b3 b4 b5 b6
b7')
m=matrix([[ 0, a1, a2, a3, a4, a5, a6, a7],
[b1,
