[Numpy-discussion] Documentation Team meeting - Monday July 5

Hi all! Our next Documentation Team meeting will be on *Monday, July 5* at ***4PM UTC***. All are welcome - you don't need to already be a contributor to join. If you have questions or are curious about what we're doing, we'll be happy to meet you! If you wish to join on Zoom, use this link: ht

Re: [Numpy-discussion] is_triangular, is_diagonal, is_symmetric et al. in NumPy or SciPy linalg

Yes they go by the name of morally triangular matrices (quite a stupid name but in their defense I think it was an insider joke) this is also given in Tim Davis' book as an exercise via linked lists. The issue is that LAPACK doesn't support these permuted matrices. Hence we are left with two option

Re: [Numpy-discussion] is_triangular, is_diagonal, is_symmetric et al. in NumPy or SciPy linalg

If you're going to provide routines for structure determination it might be worth looking at algorithms that can identify more general or less obvious structure as well. SymPy's matrices module needs a lot of work and is improving a lot which will become noticeable over the next few releases but on

Re: [Numpy-discussion] is_triangular, is_diagonal, is_symmetric et al. in NumPy or SciPy linalg

Ah right. So two things, the original reason f9r this question is because I can't decide in https://github.com/scipy/scipy/pull/12824 whether others would also benefit from quick structure determination. I can keep it private function or we can put them some misc or lib folder so all can use. Say