>From the documentation at https://www.mathworks.com/help/matlab/ref/mldivide.html, it appears to me that MATLAB gives a warning: "Warning: Matrix is close to singular or badly scaled. Results may be inaccurate." That seems to me to be a better default behavior than what sage is doing now, but I think it would make sense to let users choose to turn off the warning (and maybe also let them upgrade it to an error).
On Friday, May 22, 2020 at 9:20:00 PM UTC-6, Michael Orlitzky wrote: > > On 5/22/20 6:40 PM, AlexGhitza wrote: > > I would also argue that, despite the validity of the arguments regarding > > inexact rings, this is a change in behavior that would have benefited > > from a deprecation warning for a short while. > > We were pretty careful not to break anything in the sage library. For > typical inexact rings, the risk of breakage was minimal, because not > much worked to begin with. What we overlooked was that not all inexact > rings are typical, and that in cases like RealBallField and exact-SR, > the change could turn a right answer to a wrong one *in a situation that > wasn't ridiculous to begin with.* None of those cases were doctested, so > they were easy to overlook. > > > > Another user-friendlier way of doing this (making the check argument > > irrelevant in the inexact case) would have been to display a warning > > when the user asks for check=True in the inexact case, rather than > > simply ignoring it. > > The base class defaults to check=True, and that can't be changed, so you > would get a warning on every inexact solve in that case. Moreover for > the "usual" inexact rings, things were unilaterally improved and not > deserving of a warning. > > It will probably take a bit of domain-specific knowledge to make RBF/CBF > work as Nils expects, but reverting SR back to the old behavior when all > of its elements are exact shouldn't be hard if it doesn't cause any new > unforeseen problems. Let's just fix it instead of throwing warnings. > -- 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 view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/e0969a6d-fc9d-457d-be3a-5413a1e23b4c%40googlegroups.com.
