Hi Pablo
>In my university, we have a room with 24 computers and one nfs server
>serving the home folders for all of them. SAGE is installed in each of
>the computers individually. As the course progresses, we're running into
>severe performance problems when using SAGE in this setting. We have no
>
>
> The worsheets are very small objects; so I guess that the problem is
> somewhere else.
>
Maybe the .mozilla folder is responsible. Or maybe sage attempts to
browse the .sage directory and that causes nfs to transfer all the
files. My .sage folder is 150mb big.
> If your nfs server is a linux
Hi Nicolas,
We've been told non-US institutions are possible. Obviously, for
logistical reasons, and given the nature of our possible funding,
we'll probably favor places in the US that are close to one or more of
the main folks on the grant.
Please consider sending me a very short description o
From IRC:
ncohen this is getting really annoying.05:47
ncohen when I have an equation of the form 05:48
ncohen x + 2*y + 3 05:48
ncohen with var('x y') 05:48
ncohen how can I get the "2" which is a coefficient of y ? 05:48
ncohen I mean the real "2" 05:48
ncohen whe
Florent Hivert wrote:
> Hi there,
>
> I'll use the excuse that I'm now writing on a laptop in a train for not having
> searched if this as already been discussed...
>
> Is there a limitation somewhere (apart of course the available free time of
> the developers) which prevent us from improv
On Nov 26, 10:13 pm, Christopher Olah
wrote:
> That's awesome.
Thanks ;)
The 16x16 one looks ugly, but that's no wonder. But the svg one and
since I know they are maybe a bit larger, I also thought about some
additional tweaks like shading or offsetting the logo. Any ideas or
wishes?
H
--
T
> > Could you elaborate ? What's makes you skeptical ?
>
> Two things, mostly. The huge amount of code that wasn't being merged
> -- that appears to now be merged :) And the whole categories/generic
> code effort: while I support the ends, I'm worried that the system
> will become so slow
On 26-Nov-09, at 12:23 PM, Florent Hivert wrote:
>>> On Thu, Nov 26, 2009 at 08:30:53AM -0800, YannLC wrote:
Just a toy implementation as a very thin layer over dict (at
least it
should be fast)
>>>
>>> That's precisely what CombinatorialFreeModule elements are :-)
>>>
>>> Furthe
That's awesome.
You don't know how much time I wasted searching for something like
that recently. (OK, not that much, ~20 min). But these are nice.
Thanks. Now I have proper icons to set up on some workstations...
On Thu, Nov 26, 2009 at 1:38 PM, Harald Schilly
wrote:
> Hi, I've created some pro
> > On Thu, Nov 26, 2009 at 08:30:53AM -0800, YannLC wrote:
> >> Just a toy implementation as a very thin layer over dict (at least it
> >> should be fast)
> >
> > That's precisely what CombinatorialFreeModule elements are :-)
> >
> > Further optimizations to it are most welcome (For example, I am
On 26-Nov-09, at 10:18 AM, Nicolas M. Thiery wrote:
> On Thu, Nov 26, 2009 at 08:30:53AM -0800, YannLC wrote:
>> Just a toy implementation as a very thin layer over dict (at least it
>> should be fast)
>
> That's precisely what CombinatorialFreeModule elements are :-)
>
> Further optimizations to
On Nov 26, 9:31 am, Simon King wrote:
> InfinitePolynomialRing has an underlying *finite* polynomial ring,
> that changes whenever you need a variable that is not in the finite
> ring.
Hi, I have seen this thread has various aspects and I don't know the
details, but just reading this it reminds m
Hi, I've created some program icons.
http://sage.math.washington.edu/home/schilly/icon/
It might be a good idea to include them in Sage, so that it is easier
to create a launcher with a nice icon.
H
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this
On Thu, Nov 26, 2009 at 05:09:13PM +0100, Florent hivert wrote:
> > On Thu, Nov 26, 2009 at 06:54:43AM -0800, Nathann Cohen wrote:
> > > Actually, I use these polynomials to emulate what your
> > > CombinatorialFreeModule does on a much larger basis : everything that
> > > is hashable ;-)
> > >
>
On Thu, Nov 26, 2009 at 08:30:53AM -0800, YannLC wrote:
> Just a toy implementation as a very thin layer over dict (at least it
> should be fast)
That's precisely what CombinatorialFreeModule elements are :-)
Further optimizations to it are most welcome (For example, I am not
sure += is implement
Pablo Angulo wrote:
> Hello:
> In my university, we have a room with 24 computers and one nfs server
> serving the home folders for all of them. SAGE is installed in each of
> the computers individually. As the course progresses, we're running into
> severe performance problems when using SAGE
Pablo Angulo wrote:
> Hello:
> In my university, we have a room with 24 computers and one nfs server
> serving the home folders for all of them.
What sort of server? Is it by chance Solaris with ZFS file systems? If so, I
probably know the answer.
--
To post to this group, send an email to
Just a toy implementation as a very thin layer over dict (at least it
should be fast)
no doc
-
first see it in action:
sage: x=Test()
sage: p=x.zero_element()
sage: time for i in range(1): p+=x[i]
CPU times: user 0
> On Thu, Nov 26, 2009 at 06:54:43AM -0800, Nathann Cohen wrote:
> > Actually, I use these polynomials to emulate what your
> > CombinatorialFreeModule does on a much larger basis : everything that
> > is hashable ;-)
> >
> > I want to be able to index my variables with sets, with edges, with
> >
On Wed, Nov 25, 2009 at 10:54 PM, Nathann Cohen wrote:
> Hello !!
>
> I have some invitations for Wave left... If you are interested, leave
> your email here !
> Nathann
>
>
Same here. Ask for any if you want.
> --
> To post to this group, send an email to sage-devel@googlegroups.com
> To unsub
On Thu, Nov 26, 2009 at 06:54:43AM -0800, Nathann Cohen wrote:
> Actually, I use these polynomials to emulate what your
> CombinatorialFreeModule does on a much larger basis : everything that
> is hashable ;-)
>
> I want to be able to index my variables with sets, with edges, with
> nodes, with al
Here we use NFS and Sage: we have 3 computers. The first one receives
all the users'directories in a directory /ws. /ws is exported to the 2nd
and 3rd computer and we have no performance problems. But ok, this is
not the same configuration as yours where you have a lot of computers
and only one nfs
Actually, I use these polynomials to emulate what your
CombinatorialFreeModule does on a much larger basis : everything that
is hashable ;-)
I want to be able to index my variables with sets, with edges, with
nodes, with almost anything we can come up with in Sage...
Nathann
On Nov 26, 3:39 pm,
On Thu, Nov 26, 2009 at 05:29:47AM -0800, YannLC wrote:
> If you only want linear terms, you can also use an univariate
> polynomial ring
>
> just treat x^i as x_i.
>
> it's lightning fast and allow you to easily access coefficients.
Variant:
sage: F =CombinatorialFreeModule(QQ, NonNegativeInte
Then I think you found the very thing I needed... Thank you !!! :-)
I do not need 1 millions variables, but clearly I do not want the
computations to be too slow under 1.. If I have something like
1000 variables, I very often have 5000 or more functions to define and
work on, so I need the sym
On Nov 26, 2:17 pm, Simon King wrote:
[...]
> However, if I understood correctly, you have a *uni*variate polynomial
> ring, right? So, probably you can disregard what I just said, since
> univariate polynomial rings are different from multivariate (based on
> ntl? not sure...) .
Yep, as pointed
Hi Nathann!
On Nov 26, 2:06 pm, Nathann Cohen wrote:
> I could cache the results... But I still do not understand why just
> evaluating x^99 takes so much time !
I guess this is a limitation of Singular.
In Singular, exponents are restricted to 32767. Usually, multivariate
polynomial rings
because you use dense representation.
Try P.=PolynomialRing(QQ,sparse=True)
By the way, do you need QQ? RR or ZZ would probably be faster.
On Nov 26, 3:06 pm, Nathann Cohen wrote:
> I could cache the results... But I still do not understand why just
> evaluating x^99 takes so much time !
Y
On 25 nov, 15:54, Nathann Cohen wrote:
> Hello !!
>
> I have some invitations for Wave left... If you are interested, leave
> your email here !
> Nathann
I would like to try it. My email is mma...@unizar.es
Miguel Marco
--
To post to this group, send an email to sage-devel@googlegroups.com
T
I could cache the results... But I still do not understand why just
evaluating x^99 takes so much time !
On Nov 26, 2:54 pm, YannLC wrote:
> can you avoid sums for initialisation?
>
> sage: P.=PolynomialRing(QQ)
> sage: time p=P(dict([(i,1) for i in range()]))
> CPU times: user 0.07 s, sy
can you avoid sums for initialisation?
sage: P.=PolynomialRing(QQ)
sage: time p=P(dict([(i,1) for i in range()]))
CPU times: user 0.07 s, sys: 0.00 s, total: 0.07 s
Wall time: 0.07 s
On Nov 26, 2:43 pm, Nathann Cohen wrote:
> R = PolynomialRing(QQ, 'x')
> x = R.gen()
> sum([x^i for i in rang
R = PolynomialRing(QQ, 'x')
x = R.gen()
sum([x^i for i in range(2,)])
This is still very slow ( even if the values are larger ) :-/
Nathann
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel-unsubscr...@googleg
Oops, I misread your message... You're right !! ;-)
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL:
My problem is that I am dealing with linear functions having an
arbitrary large number of variables.
Nathann
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit th
If you only want linear terms, you can also use an univariate
polynomial ring
just treat x^i as x_i.
it's lightning fast and allow you to easily access coefficients.
On Nov 26, 2:10 pm, Nathann Cohen wrote:
> Hello !!!
>
> I am writing the patch to move from InfinitePolynomialRing to just "var
Hello !!!
I am writing the patch to move from InfinitePolynomialRing to just "var
()", and I am having several annoying problems :
To obtain the coefficients of each variable, I have to write
for v in expression.variables():
c = expression.coefficient(v)
And I wonder if this is not a quadrat
Hi there,
I'll use the excuse that I'm now writing on a laptop in a train for not having
searched if this as already been discussed...
Is there a limitation somewhere (apart of course the available free time of
the developers) which prevent us from improving the following ugly printing ?
s
On Sun, Nov 22, 2009 at 10:36:15PM -0800, Rob Beezer wrote:
> A small group is preparing a National Science Foundation education
> grant proposal to investigate how to make it easier for faculty to
> adopt and integrate mathematics software into undergraduate
> mathematics courses. Of course, we t
On Thu, 26 Nov 2009 02:51:17 -0800 (PST)
Nathann Cohen wrote:
> Hmm After thinking about it for a bit, is using var() really a
> good solution ? It is fast and everything, but I use my variables in
> functions that should not spoil the userspace with them ! When I
> define symbolic variab
Hi Martin!
As you have pointed out in the wrong thread, having a smaller ring
*has* advantages.
But the more I think about it, the more I find it stupid that I let
any element of an infinite polynomial "sparse" ring have its own
underlying finite polynomial ring. It should be better to have the
f
Hmm After thinking about it for a bit, is using var() really a
good solution ? It is fast and everything, but I use my variables in
functions that should not spoil the userspace with them ! When I
define symbolic variables, they are global and could even be accessed
in the userspace... Can
Hi Robert!
On Nov 26, 10:20 am, Robert Bradshaw
wrote:
[...]
> With over-allocation one might not even need the dense/sparse
> distinction--creating 1000 variables in a "sparse" manner would only
> need 10 reallocations. (There could still be the question of how
> expensive it is to do arit
Note that over allocating has a performance hit attached to it:
sage: P = PolynomialRing(QQ,500,'x')
sage: f = P.random_element()
sage: R = PolynomialRing(QQ,1000,'x')
sage: g = R(f)
sage: %timeit f*f
10 loops, best of 3: 18.2 µs per loop
sage: %timeit g*g
1 loops, best of 3: 32.3 µs per
Sorry, wrong thread.
Martin
--
name: Martin Albrecht
_pgp: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x8EF0DC99
_otr: 47F43D1A 5D68C36F 468BAEBA 640E8856 D7951CCF
_www: http://www.informatik.uni-bremen.de/~malb
_jab: martinralbre...@jabber.ccc.de
--
To post to this group, send an email
Note that over allocating has a performance hit attached to it:
sage: P = PolynomialRing(QQ,500,'x')
sage: f = P.random_element()
sage: R = PolynomialRing(QQ,1000,'x')
sage: g = R(f)
sage: %timeit f*f
10 loops, best of 3: 18.2 µs per loop
sage: %timeit g*g
1 loops, best of 3: 32.3 µs per
Hi Robert,
On Nov 26, 9:16 am, Robert Bradshaw
wrote:
> > On Nov 26, 8:41 am, Robert Bradshaw
> > wrote:
> > [...]
> >> Though all of the above look like errors to me, not that there is the
> >> special value NotImplemented that can be *returned* in certain cases
Oops, I did not auto-correct yo
On Nov 26, 2009, at 2:10 AM, Simon King wrote:
> Hi Robert!
>
> On Nov 26, 9:46 am, Robert Bradshaw
> wrote:
> [...]
I think this makes perfect sense...I'm actually surprised it's not
implemented that way already.
>>
>>> That's impossible.
>>
>> Over-allocating the number of generators
Hi Robert!
On Nov 26, 9:46 am, Robert Bradshaw
wrote:
[...]
> >> I think this makes perfect sense...I'm actually surprised it's not
> >> implemented that way already.
>
> > That's impossible.
>
> Over-allocating the number of generators ahead of time whenever you
> need more to achieve O(log(n)
I do not know in advance the number of variables needed.
It can be pre-computed, of course (and it would be equivalent to
actually running the whole algorithm), but we are definitely better
without this hindrance... Actually, I tried several things using var
("x_"+str(i)) and it is much better in m
Hi Nathann!
On Nov 26, 9:11 am, Nathann Cohen wrote:
> Ok, now I understand... ;-)
>
> The trouble is that obviously, I have no idea of how many variables I
> will need. I do no want to ask the user, as not having to say it is --
> really-- a relief !
I don't know exactly what you plan to do.
I
Hi Florent!
PS:
On Nov 26, 9:24 am, Florent Hivert
wrote:
> On Thu, Nov 26, 2009 at 01:16:09AM -0800, Simon King wrote:
> > > On Nov 26, 2009, at 12:35 AM, Florent Hivert wrote:
> > [...]
> > > I think this makes perfect sense...I'm actually surprised it's not
> > > implemented that way already.
Hi Florent!
On Nov 26, 9:24 am, Florent Hivert
wrote:
[...]
> I don't understand why what you say here is an answer to the following
> sentence of mine:
>
> Is there a problem in Symmetric Ideals if you have unused variables ?
You will always have an infinity of unused variables, of course: In
e
On Nov 26, 2009, at 1:16 AM, Simon King wrote:
> Hi Robert!
>
> On Nov 26, 8:43 am, Robert Bradshaw
> wrote:
>> On Nov 26, 2009, at 12:35 AM, Florent Hivert wrote:
>>> Though this could be improved by using a similar trick than
>>> doubling the size of a list when appending element, I'm not sure
Hello:
In my university, we have a room with 24 computers and one nfs server
serving the home folders for all of them. SAGE is installed in each of
the computers individually. As the course progresses, we're running into
severe performance problems when using SAGE in this setting. We have now
s
Hi David!
On Nov 26, 9:07 am, David Kohel wrote:
> Rather I would say that "sparse" should be the default:
>
> P. = InfinitePolynomialRing(QQ, implementation="sparse")
No. The main purpose of InfinitePolynomialRing is the computation of
symmetric Groebner bases, and simply it turned out in examp
Hi Simon,
On Thu, Nov 26, 2009 at 01:16:09AM -0800, Simon King wrote:
> > On Nov 26, 2009, at 12:35 AM, Florent Hivert wrote:
> [...]
> > I think this makes perfect sense...I'm actually surprised it's not
> > implemented that way already.
>
> That's impossible.
>
> The whole point of In
On Nov 26, 2009, at 1:06 AM, Simon King wrote:
> Hi Robert,
>
> On Nov 26, 8:41 am, Robert Bradshaw
> wrote:
> [...]
>> Though all of the above look like errors to me, not that there is the
>> special value NotImplemented that can be *returned* in certain cases
>>
>> http://docs.python.org/librar
Hi Robert!
On Nov 26, 8:43 am, Robert Bradshaw
wrote:
> On Nov 26, 2009, at 12:35 AM, Florent Hivert wrote:
[...]
> I think this makes perfect sense...I'm actually surprised it's not
> implemented that way already.
That's impossible.
The whole point of InfinitePolynomialRing is that you do *n
Ok, now I understand... ;-)
The trouble is that obviously, I have no idea of how many variables I
will need. I do no want to ask the user, as not having to say it is --
really-- a relief !
My other problem is that sometimes computation on symbolic variables
take a lot of time, and I think it come
Rather I would say that "sparse" should be the default:
P. = InfinitePolynomialRing(QQ, implementation="sparse")
Moreover, this syntax (and for gens, etc.) is inconsistent
with PolynomialRing. The syntax:
PolynomialRing(ring, integer, sparse=True)
would be a more coherent, where integer=Set(ZZ
Hi Robert,
On Nov 26, 8:41 am, Robert Bradshaw
wrote:
[...]
> Though all of the above look like errors to me, not that there is the
> special value NotImplemented that can be *returned* in certain cases
>
> http://docs.python.org/library/constants.html
But NotImplementedError is a Python objec
On Nov 26, 2009, at 12:35 AM, Florent Hivert wrote:
> Hi Nathann,
>
>> For Linear Programming, I need to create plenty of symbolic variables
>> which I use to represent linear functions To do it, I use the
>> class InfinitePolynomialRing which lets me create them easily ( and
>> it
>> is r
On Nov 26, 2009, at 12:21 AM, Florent Hivert wrote:
> Hi there,
>
> On Wed, Nov 25, 2009 at 09:38:23PM -0800, William Stein wrote:
>> On Wed, Nov 25, 2009 at 7:26 PM, John H Palmieri > > wrote:
>>> In ring.pyx, there is code like this:
>>>
>>>if proof:
>>>return NotImpleme
Hi Nathann,
> For Linear Programming, I need to create plenty of symbolic variables
> which I use to represent linear functions To do it, I use the
> class InfinitePolynomialRing which lets me create them easily ( and it
> is really needed, as my colleagues often have to create Linear
> Pro
Hi Nathann!
On Nov 26, 8:18 am, Nathann Cohen wrote:
[...]
> To understand my problem, just try this code :
>
> X. = InfinitePolynomialRing(RR)
> sum([x[i] for i in range(200)])
>
> Don't you think it is a bit long just to generate a sum ? I have to
> admit I do not know how this class is coded,
Hi there,
On Wed, Nov 25, 2009 at 09:38:23PM -0800, William Stein wrote:
> On Wed, Nov 25, 2009 at 7:26 PM, John H Palmieri
> wrote:
> > In ring.pyx, there is code like this:
> >
> > if proof:
> > return NotImplementedError
> > else:
> > return False
> >
Hello everybody
For Linear Programming, I need to create plenty of symbolic variables
which I use to represent linear functions To do it, I use the
class InfinitePolynomialRing which lets me create them easily ( and it
is really needed, as my colleagues often have to create Linear
Program
67 matches
Mail list logo
