[GSoC] Next priority implementations for CSymPy
79 views
Skip to first unread message
Sumith 1896
Mar 3, 2015, 11:32:21 AM3/3/15
to sy...@googlegroups.com
Hi,
I am Sumith, a GSoC applicant. I have been pretty active in the community in recent times. I am planning of implementing solvers for CSymPy because there I feel there is a possibility to speed up some stuff. After going through the previous work done in SymPy, mostly Harsh Gupta's, I understand that there will be dependence on a sets module and polynomial module currently not implemented/developed enough in CSymPy and hence, if selected, the initial few weeks will be spent on implementing them. I have gone through the reading phase, mostly old PR/mailing list discussions and Professor Fateman's papers.
As CSymPy is planned as a optional fast SymPy core, I'll take care of what has to be implemented and what not. Also that whatever is implemented will be fast and robust.
I have the following questions:
1) Are the above mentioned necessary/planned modules for CSymPy?
2) As CSymPy doesn't get many slots(I'm not sure of this year,correct me if I am wrong!), which other high priority implementations are planned for CSymPy that I could work on?
Any other feedback are welcome.
Sumith
Computer Science and Engineering Undergrad.
I am Sumith, a GSoC applicant. I have been pretty active in the community in recent times. I am planning of implementing solvers for CSymPy because there I feel there is a possibility to speed up some stuff. After going through the previous work done in SymPy, mostly Harsh Gupta's, I understand that there will be dependence on a sets module and polynomial module currently not implemented/developed enough in CSymPy and hence, if selected, the initial few weeks will be spent on implementing them. I have gone through the reading phase, mostly old PR/mailing list discussions and Professor Fateman's papers.
As CSymPy is planned as a optional fast SymPy core, I'll take care of what has to be implemented and what not. Also that whatever is implemented will be fast and robust.
I have the following questions:
1) Are the above mentioned necessary/planned modules for CSymPy?
2) As CSymPy doesn't get many slots(I'm not sure of this year,correct me if I am wrong!), which other high priority implementations are planned for CSymPy that I could work on?
Any other feedback are welcome.
Sumith
Computer Science and Engineering Undergrad.
Ondřej Čertík
Mar 3, 2015, 12:27:29 PM3/3/15
to sympy
Hi Sumith,
On Tue, Mar 3, 2015 at 3:59 AM, Sumith 1896 <sumit...@gmail.com> wrote:
> Hi,
> I am Sumith, a GSoC applicant. I have been pretty active in the
> community in recent times. I am planning of implementing solvers for CSymPy
> because there I feel there is a possibility to speed up some stuff. After
> going through the previous work done in SymPy, mostly Harsh Gupta's, I
> understand that there will be dependence on a sets module and polynomial
> module currently not implemented/developed enough in CSymPy and hence, if
> selected, the initial few weeks will be spent on implementing them. I have
> gone through the reading phase, mostly old PR/mailing list discussions and
> Professor Fateman's papers.
> As CSymPy is planned as a optional fast SymPy core, I'll take care of
> what has to be implemented and what not. Also that whatever is implemented
> will be fast and robust.
>
> I have the following questions:
> 1) Are the above mentioned necessary/planned modules for CSymPy?
The polynomials module definitely is planned. The solver module I
would first figure out in SymPy. We can implement subsets, like fast
Gaussian elimination (which we already have in CSymPy) and so on.
> 2) As CSymPy doesn't get many slots(I'm not sure of this year,correct me if
> I am wrong!), which other high priority implementations are planned for
> CSymPy that I could work on?
I expect similar amount of slots like in last years, the only
difference is that this year the applications must convince mentors
from other organizations as well.
So one high priority project is to figure out what needs to be done so
that Sage could (in principle) switch to use CSymPy instead of Pynac.
As far as I can see, it means we need to add polynomials, series and
pattern matching. If the application is good, this can be sponsored by
the Sage organization under GSoC.
As we talked about in the past, there is plenty of work right here. So
I would concentrate on good polynomials support, series and pattern
matching for this summer.
Ondrej
>
> Any other feedback are welcome.
>
> Sumith
> Computer Science and Engineering Undergrad.
>
> --
> You received this message because you are subscribed to the Google Groups
> "sympy" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sympy+un...@googlegroups.com.
> To post to this group, send email to sy...@googlegroups.com.
> Visit this group at http://groups.google.com/group/sympy.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sympy/225b5361-6afe-43ff-92e4-7d787f38d567%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
On Tue, Mar 3, 2015 at 3:59 AM, Sumith 1896 <sumit...@gmail.com> wrote:
> Hi,
> I am Sumith, a GSoC applicant. I have been pretty active in the
> community in recent times. I am planning of implementing solvers for CSymPy
> because there I feel there is a possibility to speed up some stuff. After
> going through the previous work done in SymPy, mostly Harsh Gupta's, I
> understand that there will be dependence on a sets module and polynomial
> module currently not implemented/developed enough in CSymPy and hence, if
> selected, the initial few weeks will be spent on implementing them. I have
> gone through the reading phase, mostly old PR/mailing list discussions and
> Professor Fateman's papers.
> As CSymPy is planned as a optional fast SymPy core, I'll take care of
> what has to be implemented and what not. Also that whatever is implemented
> will be fast and robust.
>
> I have the following questions:
> 1) Are the above mentioned necessary/planned modules for CSymPy?
would first figure out in SymPy. We can implement subsets, like fast
Gaussian elimination (which we already have in CSymPy) and so on.
> 2) As CSymPy doesn't get many slots(I'm not sure of this year,correct me if
> I am wrong!), which other high priority implementations are planned for
> CSymPy that I could work on?
difference is that this year the applications must convince mentors
from other organizations as well.
So one high priority project is to figure out what needs to be done so
that Sage could (in principle) switch to use CSymPy instead of Pynac.
As far as I can see, it means we need to add polynomials, series and
pattern matching. If the application is good, this can be sponsored by
the Sage organization under GSoC.
As we talked about in the past, there is plenty of work right here. So
I would concentrate on good polynomials support, series and pattern
matching for this summer.
Ondrej
>
> Any other feedback are welcome.
>
> Sumith
> Computer Science and Engineering Undergrad.
>
> You received this message because you are subscribed to the Google Groups
> "sympy" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sympy+un...@googlegroups.com.
> To post to this group, send email to sy...@googlegroups.com.
> Visit this group at http://groups.google.com/group/sympy.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sympy/225b5361-6afe-43ff-92e4-7d787f38d567%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
Ondřej Čertík
Mar 3, 2015, 12:32:06 PM3/3/15
to sympy
On Tue, Mar 3, 2015 at 10:27 AM, Ondřej Čertík <ondrej...@gmail.com> wrote:
> So one high priority project is to figure out what needs to be done so
> that Sage could (in principle) switch to use CSymPy instead of Pynac.
> As far as I can see, it means we need to add polynomials, series and
> pattern matching.
That's feature wise. Then the Python wrappers need to be improved so
> So one high priority project is to figure out what needs to be done so
> that Sage could (in principle) switch to use CSymPy instead of Pynac.
> As far as I can see, it means we need to add polynomials, series and
> pattern matching.
that it plays well with Sage, and is able to use Sage objects as part
of our expressions, using:
https://github.com/sympy/csympy/blob/7585503c149e1e8c766cca3a7a21cce300ea74dd/src/functions.h#L588
it's already done for SymPy:
https://github.com/sympy/csympy/blob/7585503c149e1e8c766cca3a7a21cce300ea74dd/csympy/lib/csympy_wrapper.pyx#L452
So it would just need to get working for Sage as well.
Ondrej
Sumith 1896
Mar 3, 2015, 12:58:37 PM3/3/15
to sy...@googlegroups.com
Sure, I would like to work on the next priority of CSymPy more than anything else. I'll start from the discussion on polynomials that happened before. Also, since there will multiple people working we can keep discussions active on gitter/PRs.
--
You received this message because you are subscribed to the Google Groups "sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sympy+un...@googlegroups.com.
To post to this group, send email to sy...@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CADDwiVCgVe014zQ4_8vzR0CALKj%3Dp_SYT-vaDBKYg8f6fwVSHg%40mail.gmail.com.
Ondřej Čertík
Mar 3, 2015, 1:22:52 PM3/3/15
to sympy
On Tue, Mar 3, 2015 at 10:58 AM, Sumith 1896 <sumit...@gmail.com> wrote:
> Sure, I would like to work on the next priority of CSymPy more than anything
> else. I'll start from the discussion on polynomials that happened before.
> Also, since there will multiple people working we can keep discussions
> active on gitter/PRs.
Yes, thanks. Let me just repeat that it is ok for more students to
> Sure, I would like to work on the next priority of CSymPy more than anything
> else. I'll start from the discussion on polynomials that happened before.
> Also, since there will multiple people working we can keep discussions
> active on gitter/PRs.
apply for the same thing. Don't worry about others, just concentrate
on what you think is the best given your abilities. If we have
multiple applications/people that we would like to accept, then after
they are accepted, we'll amend the work, so that it is not
overlapping.
Ondrej
Sumith 1896
Mar 3, 2015, 1:24:23 PM3/3/15
to sy...@googlegroups.com
Sure, thanks.
To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CADDwiVCnWLu44pEH0PkJ73prZYPu9FiKj9FTWNVJuLKuFM3kFw%40mail.gmail.com.
Reply all
Reply to author
Forward
0 new messages