Hi Nathann, On 2014-11-28, Nathann Cohen <nathann.co...@gmail.com> wrote: > sage: g = posets.BooleanLattice(5).hasse_diagram().transitive_closure(); > sage: g = g.cartesian_product(g) > sage: %time Poset(g) > CPU times: user 284 ms, sys: 32 ms, total: 316 ms > Wall time: 278 ms > Finite poset containing 1024 elements > sage: %time Poset(g) > CPU times: user 1.63 s, sys: 44 ms, total: 1.68 s > Wall time: 1.61 s > Finite poset containing 1024 elements > > Even without my optimisations you see this effect, and I am pretty sure > that the huge time penalty here is because Poset inherits from > UniqueRepresentation. Profiling shows that most of the time is spent > testing equality of posets, which is part of the UniqueRepresentation > initialization.
You mean the comparison of g with itself takes so much time? Hard to believe, but it really seems to be the case: sage: g = posets.BooleanLattice(5).hasse_diagram().transitive_closure() sage: g = g.cartesian_product(g) sage: %timeit g==g 1 loops, best of 3: 10.3 s per loop How can that be? Isn't generally the test by "==" first checking whether the two arguments are identical, before calling `.__eq__`? If python does not do it automatically, then at least in your example it would be good to make g.__eq__ use the "self is other" test, before doing anything expensive. Probably the problem would also be solved by implementing rich comparison via __richcmp__ (not sure though). In any case, if there is a reason to have unique representation, then ideally the defining data should be easy to compare. Evidently, a dictionary lookup has to be involved for any kind of unique represenation. Best regards, Simon -- 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.
