And And - 3 months ago 12
Python Question

Converting 2D Numpy array to a list of 1D columns

What would be the best way of converting a 2D numpy array into a list of 1D columns?

For instance, for an array:

array([[ 0, 5, 10],
[ 1, 6, 11],
[ 2, 7, 12],
[ 3, 8, 13],
[ 4, 9, 14]])


I would like to get:

[array([0, 1, 2, 3, 4]), array([5, 6, 7, 8, 9]), array([10, 11, 12, 13, 14])]


This works:

[a[:, i] for i in range(a.shape[1])]


but I was wondering if there is a better solution using pure Numpy functions?

Answer

I can't think of any reason you would need

[array([0, 1, 2, 3, 4]), array([5, 6, 7, 8, 9]), array([10, 11, 12, 13, 14])]

Instead of

array([[0, 1, 2, 3, 4], [5, 6, 7, 8, 9], [10, 11, 12, 13, 14]])

Which you can get simply with a.T

If you really need a list, then you can use list(a.T)

Comments