I think you should ask either sage-support@googlegroups.com, or at ask.sagemath.org. Is this on your own machine or on fermat ? If the latter I can dig around to see if the file already exists and copy it.
John On 14 March 2014 10:48, Moore, Warren <warren.mo...@warwick.ac.uk> wrote: > Random Sage (Notebook) question... I've stupidly ran a rather long program > without saving to file, just relying on the full_output.txt file that the > Sage Notebook generates once a cell has finished completely. It might be > another week and I'd rather not wait for it to finish if I can avoid, but > don't want to lose all the curves it's found so far. > > Do you know of any way for me to get a hold of the complete output so far > without waiting for the cell to finish evaluating? When I've interrupted Sage > in the past I only seem to be able to access the truncated data, not all the > data that was output before I terminated / interrupted the cell. > > I know you sent me a link to somewhere where I could ask Sage questions a > while ago, but haven't been able to find it - that might be a more > appropriate place to ask whenever you get the chance to pass on the link to > the forum / Google Group. > > Thanks! > > Warren > >> On 14 Mar 2014, at 17:30, "John Cremona" <john.crem...@gmail.com> wrote: >> >> Sounds good. Feel free to use hilbert too. >> >> John >> >>> On 14 March 2014 09:46, Moore, Warren <warren.mo...@warwick.ac.uk> wrote: >>> It found the curve I was missing in 10-15 minutes with 'Effort' set to 400 >>> and without specifying any primes/Hecke eigenvalues: >>> >>> y^2 - x*y + (-2*i + 2)*y = x^3 + (2*i - 2)*x^2 + (50*i + 72)*x + (366*i - >>> 250) >>> (a global minimal model is y^2 + x*y + (i+1)*y = x^3 + (-i+1)*x^2 + >>> (52*i+71)*x + (345*i-126)) >>> >>> I wasn't able to find this with egrosNF, or by any other method I had at my >>> disposal. If it's always that quick, then this could be very useful for >>> filling in the gaps...! I don't want to flood Fermat since it only has 24 >>> cores to run on, but Hilbert looks like it has the same version of Magma? I >>> have been using a lot of it for running egrosNF constantly over the last >>> week or so, but maybe I'd be better off computing with this instead. Hmm... >>> >>> I'd be very interested to find out exactly what it's doing to find these >>> curves, that is if the method isn't too complicated for me to understand >>> and you manage to get some more info from the author! >>> ________________________________________ >>> From: Moore, Warren >>> Sent: 14 March 2014 16:21 >>> To: John Cremona >>> Subject: RE: Using Magma >>> >>> Oh wow... Okay... Well it would have felt like a bit of a cheat to just use >>> that sort of function anyway! Not that I'm expecting it to be a sort of >>> 'magic' solution to finding every elliptic curve. >>> >>> I'm giving it a quick go on Fermat now with the first missing conductor >>> over Q(sqrt(-1)) and 'Effort' set to 400 to see what happens, as that seems >>> to be the first value that tries all techniques according to the docs. I >>> found nothing at 'Effort' set to 1. But like you said, if it happens to >>> find some of the missing curves that I'm struggling to find, then that can >>> only be a good thing! >>> >>> I had a really quick glance at the >>> /usr/local/magma/package/Geometry/CrvEll/ec_search.m file on Fermat, and I >>> recognised some of the methods (some were taken from yours and Mark >>> Lingham's good reduction outside S paper), and it says that passing in >>> primes and Hecke eigenvalues would speed it up, which I may try later. I >>> don't how much time I'll have now, but I'll see whether I can get some good >>> use of it. >>> >>> Warren >>> ________________________________________ >>> From: John Cremona <john.crem...@gmail.com> >>> Sent: 14 March 2014 15:47 >>> To: Moore, Warren >>> Subject: Using Magma >>> >>> You'll hate me for this but: I only learned yesterday that the >>> current Magma (as on fermat) has an EllipticCurveSearch function which >>> I am told works well (I do not know quite what it does). See >>> http://magma.maths.usyd.edu.au/magma/handbook/text/1401#15780 for >>> info. >>> >>> The people who found curves to match modular forms over a complex >>> cubic field said that it worked very well for them. I just tried it >>> with (7+4*i) and it did not find the curve but I did not have a large >>> value for their "effort" parameter. >>> >>> I certainly would have told you about this earlier if I have known. >>> But if it helps find more curves, so much the better, even if it is a >>> "black box" with no clue as to method (though I am now about to write >>> to the person who wrote it). >>> >>> John -- You received this message because you are subscribed to the Google Groups "sage-support" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
