On Fri, Nov 13, 2015 at 12:50 PM, Nathaniel Smith wrote:
> On Nov 13, 2015 10:06 AM, "Charles R Harris"
> wrote:
> >
> > Hi All,
> >
> > I think 1.10.0 and 1.10.1 are sufficiently buggy that they should be
> removed from circulation as soon as 1.10.2 comes out. The inner product bug
> for non co
On Nov 13, 2015 10:06 AM, "Charles R Harris"
wrote:
>
> Hi All,
>
> I think 1.10.0 and 1.10.1 are sufficiently buggy that they should be
removed from circulation as soon as 1.10.2 comes out. The inner product bug
for non contiguous arrays is particularly egregious. It is not customary to
remove ou
On 11/13/15 at 12:16pm, Charles R Harris wrote:
> On Fri, Nov 13, 2015 at 11:22 AM, Manolo Martínez
> wrote:
>
> > On 11/13/15 at 11:06am, Charles R Harris wrote:
> > > The inner product bug
> > > for non contiguous arrays is particularly egregious.
> >
> > Could you please post a link to the rel
On Fri, Nov 13, 2015 at 11:22 AM, Manolo Martínez
wrote:
> On 11/13/15 at 11:06am, Charles R Harris wrote:
> > The inner product bug
> > for non contiguous arrays is particularly egregious.
>
> Could you please post a link to the related issue? I have been seeing
> very strange things with scipy.
On 11/13/15 at 11:06am, Charles R Harris wrote:
> The inner product bug
> for non contiguous arrays is particularly egregious.
Could you please post a link to the related issue? I have been seeing
very strange things with scipy.integrate.odeint, and I wonder if they
are related.
Thanks,
Manolo
_
Hi,
On Fri, Nov 13, 2015 at 10:06 AM, Charles R Harris
wrote:
> Hi All,
>
> I think 1.10.0 and 1.10.1 are sufficiently buggy that they should be removed
> from circulation as soon as 1.10.2 comes out. The inner product bug for non
> contiguous arrays is particularly egregious. It is not customary
Hi All,
I think 1.10.0 and 1.10.1 are sufficiently buggy that they should be
removed from circulation as soon as 1.10.2 comes out. The inner product bug
for non contiguous arrays is particularly egregious. It is not customary to
remove outdated Numpy releases from sourceforge and pypi, but I'd lik
On 11/11/2015 18:38, Sebastian Berg wrote:
Sounds fine to me, and considering the squeeze logic (which I think is
unfortunate, but it is not something you can easily change), I would be
for simply adding logic to accept a single integral argument and
otherwise not change anything.
[...]
As said
