Nathann Cohen <nathann.co...@gmail.com> writes: > Hello everybody !!! > > I recently asked a question here : I have a set of points in R^n (or > C^n, or any vectorial space for the matter..), to which is associated > a set of values. Said differently, I have a function whose values I > only know at several points. I then would like, given a degree d, to > find a polynomial function of maximum degree d that goes through all > these points. > > As Jason mentionned it in > http://groups.google.com/group/sage-support/browse_thread/thread/fa4a2936571de92f/a04de428e88abd77#a04de428e88abd77, > it seems to be the Chebyshev approximation defined in > http://mpmath.googlecode.com/svn/tags/0.13/doc/build/calculus/approximation.html. > > Well, this approximation actually only works for dimension 1, while I > needed it in higher dimension. I wrote this functions, which is > basically the inversion of a matrix once everything is defined, and > thought it could be good to have it in Sage. Sadly, I know next to > nothing in this field of mathematics, and I am not sure my code is > clever/optimal. Besides, what is the best option ? Include it in > MPMaths ( fron which Jason found the approximation, or directly into > Sage ? )
We have a *very* efficient (well, I believe so, at least) version of (generalized) one-dimensional rational interpolation in FriCAS, and it's one of the very short term goals to adapt it for several dimensions. Hm, if your values are in the ground field, it should be there already, actually. In case of interest, I would check. Help appreciated, Martin --~--~---------~--~----~------------~-------~--~----~ 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: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
