I have an array like this:
a = np.array([[1,2,3,4,5],
[6,0,8,9,10],
[11,12,13,14,15],
[16,0,0,19,20]])
And I would like to remove columns and rows where there is a 0 value, so the new a should be like:
array([[1,4,5],
[11,14,15]])
How to work this out using indexing?
>>> a[(a != 0).all(axis=1)][:,(a != 0).all(axis=0)]
array([[ 1, 4, 5],
[11, 14, 15]])
Finding the elements of a that are non-zero is really easy:
>>> (a != 0)
array([[ True, True, True, True, True],
[ True, False, True, True, True],
[ True, True, True, True, True],
[ True, False, False, True, True]], dtype=bool)
And then you can just use all, specifying the axis, to find the rows you want to keep:
>>> (a != 0).all(axis=1)
array([ True, False, True, False], dtype=bool)
and the same thing for the columns:
>>> (a != 0).all(axis=0)
array([ True, False, False, True, True], dtype=bool)
