[SciPy-User] vdot vs. dot

[Nico Schlömer]

> No idea, but it's probably always been like this.

Oh well, I'd consider this a bug in the design of vdot(), but probably
not something we could fix that easily.
Maybe a new function (cdot()?) could take care of this.

In any case, the documentation needs updating here.

Cheers,
Nico




On Thu, Oct 14, 2010 at 11:00 AM, Pauli Virtanen <pav at iki.fi> wrote:
> Thu, 14 Oct 2010 09:40:55 +0200, Nico Schlömer wrote:
> [clip]
>> vdot() returns a *scalar* here, and I've got no idea how it is computed.
>
> It's
>
>        >>> print np.dot( X.ravel().conjugate(), Y.ravel() )
>        (16.8356363983+3.7240082208j)
>
> And the documentation is wrong, although the examples do show that it
> flattens the inputs.
>
>> Why isn't  vdot( x, y ) == dot( x.conjugate(), y )?
>
> No idea, but it's probably always been like this.
>
>        Pauli
>
> _______________________________________________
> SciPy-User mailing list
> SciPy-User at scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-user
>