> > This question seems like it's more relevant to the scipy mailing list, > since scipy has scipy.sparse. > My point was that the current situation (i.e., people reinventing the wheel) was a by-product of the fact that there's nothing as standardized as numpy's ndarray or Python's buffer interface, if only for the simple reason that most people only ever work with dense matrices.
(scipy.sparse uses several different data structures that are all limited to two axes, and have no common low-level interface). I'll be happy to ask on the scipy list, if that's where the discussion should be. Yannick
_______________________________________________ NumPy-Discussion mailing list [email protected] http://mail.scipy.org/mailman/listinfo/numpy-discussion
