numpy array question

[Peter Otten]

jagmit sandhu wrote:

> python newbie. I can't understand the following about numpy arrays:
> 
> x = np.array([[0, 1],[2,3],[4,5],[6,7]])
> x
> array([[0, 1],
>        [2, 3],
>        [4, 5],
>        [6, 7]])
> x.shape
> (4, 2)
> y = x[:,0]
> y
> array([0, 2, 4, 6])
> y.shape
> (4,)
> 
> Why is the shape for y reported as (4,) ? I expected it to be a (4,1)
> array. thanks in advance

Why do you expect every array to be a 2D matrix? Why not 3D or 4D? In the 
shape of y the 4 could be followed by an infinite amount of ones

(4, 1, 1, 1, 1, 1,...)

Instead numpy uses as many dimensions as you provide:

>>> import numpy as np
>>> np.array(42).shape
()
>>> np.array([42]).shape
(1,)
>>> np.array([[42]]).shape
(1, 1)
>>> np.array([[[42]]]).shape
(1, 1, 1)

I think this is a reasonable approach. Dont you?