I'm a little confused by that - isn't that the same as solving a linear programming problem? Anyway, I think the answer is yes. cddlib includes a linear programming solver, but I have not used it directly. I have no idea how it compares to cvxopt in speed - I expect that cvxopt is faster for most things.
Perhaps Rekha's interest is in exact computations, which cvxopt can't do (is that right? I haven't used it much). My interface is currently very crude, and only converts back and forth from vertex to inequality descriptions. It could solve the above problem, of course, but in the most brute force way possible - computing every vertex and then it is easy to find the maximum. The upside of that is if you are interested in changing the functional for the same inequalities you could solve it once and for all. I currently don't have the time to really dive into using cddlib the way its supposed to be, but hopefully I will next spring. Right now I am just using the supplied example commands like testcdd1_gmp, rather than writing my own code in C. But in the medium term, the next year or two, I think most useful things in polymake can and should be moved to a Sage module using cddlib. (gfan uses cddlib for quite a bit, although I am not sure of exactly what.) Cheers, Marshall On Mar 31, 11:25 pm, "William Stein" <[EMAIL PROTECTED]> wrote: > On Mon, Mar 31, 2008 at 7:56 PM, Marshall Hampton <[EMAIL PROTECTED]> wrote: > > > Hello, > > > What do you want to do with polymake? I've been working on > > interfacing to cddlib, which does most of the heavy lifting for > > polymake for things I care about. Because my interface does > > everything I need, I've stopped installing the polymake package > > anymore. But if you describe what functionality you need and it > > overlaps with my interests I might be able to extend my stuff to do > > what you want. > > Marshall, > > Just out of curiosity can cddlib compute the maximum value of a linear > functional on a polytope described by linear inequalities? Rekha Thomas > asked me this a couple of days ago. > > Rekha -- if i mis-asked this question, could you respond with something > more precise? > > -- William --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-support@googlegroups.com 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://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
