On Nov 4, 2:49 pm, Simon King <[EMAIL PROTECTED]> wrote:
> Dear John,
Hi Simon,
>
> On Nov 4, 1:45 pm, "John Cremona" <[EMAIL PROTECTED]> wrote:
>
> > Is it recomputing a Grobner basis for the new ideal? That could be slow.
>
> No, it is simply
>
> > > singular.eval( I.name()+'[%d]' = '%(sz)+p.name())
>
> where sz is ncols(I)+1, and p is a polynomial.
>
> Of course, that line of code is part of a bigger computation. So, why
> do i think that this line is critical? When i observed that the whole
> computation got stuck, i inserted various "print" commands into my
> code - and it turned out that the print command in front of the above
> line was executed, but the print command after the line wasn't. So, i
> conclude that it is surprizingly difficult to put <p> into the
> position <sz> of <I>, while all other commands (including groebner
> basis computation) were easy.
>
In & output via the pexpect interface is quadratic, so it might as
well be a problem if you send a lot of data to and from Singular. It
would be interesting to see if a standalone Singular would behave
differently (I assume it does). In the past when problems like this
have popped up the solution was to write the input or output to a file
and then parse it from that, achieving more or less linear complexity.
I am not sure if this helps in this case, but it might be worth a try.
Another possibility is to get this particular operation into the
libSingular interface, but here also I am not sure how much work this
would be. Hopefully malb can/will enlighten you if this is possible
and/or how much work it would be.
> Probably it'd be the best to make a singular log and send you the log
> file, right?
Maybe, it would be useful to see timestamps from both sides of the
computation to isolate communication overhead for pexpect.
>
> Yours
> Simon
Cheers,
Michael
--~--~---------~--~----~------------~-------~--~----~
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-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~----------~----~----~----~------~----~------~--~---
