[Numpy-discussion] dimension aligment

Charles R Harris charlesr.harris@gmail....
Tue May 20 12:56:23 CDT 2008


On Tue, May 20, 2008 at 11:47 AM, Thomas Hrabe <thrabe@burnham.org> wrote:

>
> Hi all,
>
> just a simple question regarding the alignment of dimensions:
>
>
> given a 3d array
> a =
> numpy.array([[[1,2,3],[4,5,6]],[[7,8,9],[10,11,12]],[[13,14,15],[16,17,18]],[[19,20,21],[22,23,24]]])
> a.shape
> returns (4,2,3)
>
> so I assume the first digit is the 3rd dimension, second is 2nd dim and
> third is the first.
>

Only if you count from the right. I would call the first digit the first
dimension.

>
> how is the data aligned in memory now?
> according to the strides it should be
> 1,2,3,4,5,6,7,8,9,10,...
> right?
>
Like a C array, contiguous and the rightmost dimension varies fastest.

>
> if I had an array of more dimensions, the first digit returned by shape
> should always be the highest dim.
>
Yes. Athough first is less ambiguous than highest

Chuck
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://projects.scipy.org/pipermail/numpy-discussion/attachments/20080520/502046d4/attachment.html 


More information about the Numpy-discussion mailing list