As far as I know, everything that has been implemented is already in the codebase, with the exception of https://github.com/sympy/sympy/pull/563. I've CC'd Jeremias (the student who worked on it last year) and Mateusz (his mentor), to see if they have any more comments. As far as I know, Mario's suggestion is valid.
If Mateusz is free, he'll likely be the one to mentor such a project. Otherwise, I don't know. Mario, would you be willing to help out with such a project, if not directly mentor it. I think your knowledge of Groebner bases is probably the strongest. We can help out where your knowledge of the codebase is lacking. Otherwise, I'll do it, though I'll have to update my own knowledge of Groebner bases and these algorithms (which would not be a bad thing). As for the linear algebra, I'm not sure it's something you could just do really quickly, though feel free to look into it. What we need to do is make things faster, and one way would be to restructure things in Matrix so that they are similar to the way they are in Poly. Otherwise, from my understanding, the algorithms that use linear algebra are not really faster, as they kind of assume that the linear algebra will be fast(er). Aaron Meurer On Sat, Mar 10, 2012 at 2:08 PM, Sergiu Ivanov <[email protected]> wrote: > Hello, > > I'd like to hear a definitive word on the status of Groebner bases in > SymPy. The ideas page says (just like it did a couple weeks ago) that > there was a project on this topic last year, and then it invites the > student to contact the developers, which I hereby do :-) > > In the source tree, I can see the file sympy/polys/groebnertools.py, > which contains the implementation of Buchberger and F5B algorithms. > I've seen some suggestions for further improvement on the list ( > https://groups.google.com/forum/?fromgroups#!topic/sympy/EI47kQM6S8c > ), however neither of them has made it into the official ideas page. > > I'd be interested in working on (improving) Groebner bases, so it > would be great to settle on a task which I could start preparing for. > I would also be very glad to hear from the people who are willing to > mentor such a task. > > The ideas page says that Groebner bases in SymPy would benefit from > improvements to linear algebra. I absolutely don't mind (moreover, > I'd like it) to include a couple of such improvements into my eventual > proposal. Moreover, I could actually try to make a couple such > improvements right away, as a part of satisfying the patch > requirement. > > A slight disclaimer is that I'm still not completely through my > Groebner bases course at the uni, so I'm by far not an expert in this > domain. I do believe though that my passion for and (something that I > consider to be) general knowledge of algebra, and, particularly of > modules and rings, will be sufficient to eventually grasp any > Groebner-bases-related idea. > > Sergiu > > -- > You received this message because you are subscribed to the Google Groups > "sympy" 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/sympy?hl=en. > -- You received this message because you are subscribed to the Google Groups "sympy" 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/sympy?hl=en.
