[Numpy-discussion] Can not update a submatrix

Nadav Horesh nadavh@visionsense....
Thu Jan 31 05:21:31 CST 2008


Thanks for your explanation. It explains also the following:

>>> R = N.arange(9).reshape(3,3)
>>> ax = [0,2]
>>> U = R[ax,:]
>>> U
array([[0, 1, 2],
       [6, 7, 8]])
>>> U[:] = 100
>>> U
array([[100, 100, 100],
       [100, 100, 100]])
>>> R    # No change since U is a copy
array([[0, 1, 2],
       [3, 4, 5],
       [6, 7, 8]])
>>> R[ax,:] = 100  # Works through __setitem__
>>> R
array([[100, 100, 100],
       [  3,   4,   5],
       [100, 100, 100]])

   Nadav


On Wed, 2008-01-30 at 13:35 -0700, Timothy Hochberg wrote:

> 
> 
> 
> On Jan 30, 2008 12:43 PM, Anne Archibald <peridot.faceted@gmail.com>
> wrote:
> 
>         On 30/01/2008, Francesc Altet <faltet@carabos.com> wrote:
>         > A Wednesday 30 January 2008, Nadav Horesh escrigué:
>         > > In the following piece of code:
>         > > >>> import numpy as N
>         > > >>> R = N.arange(9).reshape(3,3)
>         > > >>> ax = [1,2]
>         > > >>> R
>         > >
>         > > array([[0, 1, 2],
>         > >        [3, 4, 5],
>         > >        [6, 7, 8]])
>         > >
>         > > >>> R[ax,:][:,ax] = 100
>         > > >>> R
>         > >
>         > > array([[0, 1, 2],
>         > >        [3, 4, 5],
>         > >        [6, 7, 8]])
>         > >
>         > > Why R is not updated?
>         >
>         > Because R[ax] is not a view of R, but another copy of the
>         original
>         > object (fancy indexing does return references to different
>         objects).
>         > In order to get views, you must specify only a slice of the
>         original
>         > array.  For example:
>         
>         
>         
>         This is not exactly correct.
>         
> 
> [...a fine explanation by Anne...]
> 
> Another way to look at this is note that:
> 
> 
> R[ax,:][:,ax] = 100
> 
> 
> 
> is translated to:
> 
> 
> 
> R.__getitem__((ax, :)).__setitem__((:,ax), 100)
> 
> 
> 
> '__setitem__' is capable of setting values using fancy indexing, but
> as has been noted, '__getitem__' makes a copy of R when using fancy
> indexing, and therefore the original does not get modified.
> 
> [Technically, ':' gets translated to slice(None,None,None), but I
> didn't want to write that out].
> 
> -- 
> .  __
> .   |-\
> .
> .  tim.hochberg@ieee.org 
> 
> _______________________________________________
> Numpy-discussion mailing list
> Numpy-discussion@scipy.org
> http://projects.scipy.org/mailman/listinfo/numpy-discussion
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://projects.scipy.org/pipermail/numpy-discussion/attachments/20080131/102e9f08/attachment.html 


More information about the Numpy-discussion mailing list