On 26/04/15 21:21, Dima Pasechnik wrote:
On Saturday, 25 April 2015 10:26:48 UTC+1, vdelecroix wrote:Hello, I have COIN-OR installed on my computer and hence MILP chooses it as a default: {{{ sage: default_mip_solver() 'Coin' }}} (it would have been GLPK otherwise) It is annoying since as a consequence the following fails: ********************************************************************** File "src/doc/en/thematic_tutorials/linear_programming.rst", line 432, in doc.en.thematic_tutorials.linear_programming Failed example: p.solve() # rel tol 2e-12 Expected: 2.0 Got: 2.0000000000040004 Tolerance exceeded: 2.0 vs 2.0000000000040004, tolerance 2e-12 > 2e-12 ********************************************************************** In particular, all MILP related doctests are highly dependent on the machine installation!I would advocate the approach that all the MILP tests are explicit about the solver to be used. Indeed, it's not only COIN, but all the other solvers like CPLEX behave this way. Nathann, would you mind this?
I do not know for Nathann. But I would. Moreover, all MILP backends should agree (up to numerical noise). So it is nice that this code gets tested with different backends.
Note that #18297 solves the above issue. Vincent -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at http://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
