From c86a38c318f12fb43d821220b0b6bedba98fab9b Mon Sep 17 00:00:00 2001 From: "Nathaniel J. Smith" Date: Fri, 14 Mar 2014 00:06:16 +0000 Subject: realized I wasn't so sure of my claim about in-place matrix multiplication, so couch it more cautiously --- doc/neps/return-of-revenge-of-matmul-pep.rst | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'doc') diff --git a/doc/neps/return-of-revenge-of-matmul-pep.rst b/doc/neps/return-of-revenge-of-matmul-pep.rst index 4bcd07266..8c6d410ec 100644 --- a/doc/neps/return-of-revenge-of-matmul-pep.rst +++ b/doc/neps/return-of-revenge-of-matmul-pep.rst @@ -551,10 +551,10 @@ still included, though, for consistency: if we have an ``@`` that is analogous to ``*``, then it would be weird and surprising to *not* have an ``@@`` that is analogous to ``**``. Similarly, the in-place operators ``@=`` and ``@@=`` provide limited value -- it's more common -to write ``a = (b @ a)`` than it is to write ``a = (a @ b)``, and it -is not generally possible to implement in-place matrix multiplication -any more efficiently than by making a full copy of the matrix -- but -they are included for completeness and symmetry. +to write ``a = (b @ a)`` than it is to write ``a = (a @ b)``, and +in-place matrix operations still generally have to allocate +substantial temporary storage -- but they are included for +completeness and symmetry. Compatibility considerations -- cgit v1.2.1