[Numpy-discussion] lstsq functionality

Keith Goodman kwgoodman@gmail....
Wed Jul 21 18:44:03 CDT 2010


On Wed, Jul 21, 2010 at 4:35 PM, Skipper Seabold <jsseabold@gmail.com> wrote:
> On Tue, Jul 20, 2010 at 10:24 PM, Keith Goodman <kwgoodman@gmail.com> wrote:
>> Good point. Looks like we can get rid of 2 copies! I didn't get rid of
>> the second copy but I did cut things down just to see what the timing
>> was like. I also threw out the ability to handle complex numbers (only
>> saves an if iscomplex statement):
>
> Just for timing purposes though right?

Yes.

Can someone confirm that the copy in np.linalg.lstsq

bstar[:b.shape[0],:n_rhs] = b.copy()

is not needed? I'm assuming that ndarray.__setitem__ never gives a
view of the right-hand side.


More information about the NumPy-Discussion mailing list