On 2013-03-26, tvn <[email protected]> wrote: > ------=_Part_262_20933178.1364323802734 > Content-Type: text/plain; charset=ISO-8859-1 > > Yes -- it would be very convenient for the users. For example, applying > solve() on a a set of linear equations probably do some preprocessing to > turn them into more efficient format and send it to some specialize > solver. But all those steps happen behind the scene , the only thing I > give solve() is a list of equations ! > > I just wrote some functions for this at http://pastebin.com/FBUQ0jsr . It's > geared toward my project but can easily be modified for more general > purpose. Feel free to use them. > I already wrote that there is an ambiguity in the setup, for some variables in the set of inequalities might already be bound. Should one neglect this? Probably not.
Anyway, please feel free to join in the Sage development and submit an enhancement patch for MixedIntegerLinearProgram class, that is, adding an extra constructor. If not you, then who, if not now, then when? :) Dmitrii > > > > On Tuesday, March 26, 2013 12:17:41 PM UTC-6, john_perry_usm wrote: >> >> Dima & Nathann >> >> What he's facing is an annoyance. Would it be feasible (ha ha) to rewrite >> the creation of constraints so that variables are automatically created in >> the LP if they're not there? That is, rather than looking for a workaround >> for the user, should we see this as an opportunity for enhancement? >> >> john perry >> >> On Tuesday, March 26, 2013 12:54:48 AM UTC-5, tvn wrote: >>> >>> I am trying to play around with MixedIntegerLinearProgram but now sure >>> how to add constraints properly to it. My input is a list of relations , >>> e.g. csts = [l <= 15, u >= 10, l <= u] and I want to add these to the >>> solver. >>> >>> All the examples I've seen requires constructing these constraints >>> explicity (e.g. p = MixedIntegerLinearProgram(), l = p['l'], u = p['u'], >>> p.set_objective(l-u), p.add_constraint(l <= 15), p.add_constraint(u>=10)) >>> . I could analyze the input list of constraint to determine is >>> coeficients and variables etc and then recreate new variables and relations >>> but that seem too cumbersome. Is there an easier method to feed a given >>> list of relations as above directly to p.add_constraint() ? >>> >>> >>> >>> > -- 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 [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sage-support?hl=en. For more options, visit https://groups.google.com/groups/opt_out.
