On Wed, May 22, 2013 at 11:15 AM, eat wrote:
> FWIW, apparently bug related to dtype of np.eye(.)
sort of -- the issue shows up when assigning a float64 array (default
for eye()) to a rank-0 array with a custom dtype that has a single
object filed that is an arraynumpy should do the right th
Thanks, I filed a new issue on the bug tracker.
Nicolas
On May 22, 2013, at 8:15 PM, eat wrote:
> Hi,
>
> FWIW, apparently bug related to dtype of np.eye(.)
>
>
> On Wed, May 22, 2013 at 8:07 PM, Nicolas Rougier
> wrote:
>
>
> Hi all,
>
> I got a weird output from the following scr
Hi,
FWIW, apparently bug related to dtype of np.eye(.)
On Wed, May 22, 2013 at 8:07 PM, Nicolas Rougier
wrote:
>
>
> Hi all,
>
> I got a weird output from the following script:
>
> import numpy as np
>
> U = np.zeros(1, dtype=[('x', np.float32, (4,4))])
>
> U[0] = np.eye(4)
> print U[0]
> # out
On Wed, May 22, 2013 at 10:07 AM, Nicolas Rougier
> U = np.zeros(1, dtype=[('x', np.float32, (4,4))])
>
> U[0] = np.eye(4)
> print U[0]
> # output: ([[0.0, 1.875, 0.0, 0.0], [0.0, 0.0, 0.0, 0.0], [0.0, 0.0, 0.0,
> 1.875], [0.0, 0.0, 0.0, 0.0]],)
I get the same thing. Note:
In [86]: U[0].shape
O
Hi all,
I got a weird output from the following script:
import numpy as np
U = np.zeros(1, dtype=[('x', np.float32, (4,4))])
U[0] = np.eye(4)
print U[0]
# output: ([[0.0, 1.875, 0.0, 0.0], [0.0, 0.0, 0.0, 0.0], [0.0, 0.0, 0.0,
1.875], [0.0, 0.0, 0.0, 0.0]],)
U[0] = np.eye(4, dtype=np.float
