Finding first zero element ignoring leading zeros in numpy array

Question

I'm trying to figure out the fastest way to find the first zero element along the second axis of a 2d NumPy array ignoring the leading block of zeros.

For example, I have following array:

a = np.array([[ 0. ,  0. ,  0. ],
       [ 0. ,  3. ,  0. ],
       [ 2. ,  0. ,  0. ],
       [ 0. ,  1.7,  1.5],
       [ 1. ,  1.5,  1.6]])

and what I would like to get is:

array([3, 2, 4])

In fact, ignoring all the leading zeros. This could be achieved in a for loop counting the number of leading zeros and subsequently slicing each column, it however looks pretty ugly and not so NumPy oriented.

because I'm ignoring the first 3 zeros and the array does not contain any other 0 element — David
– David, Commented Jul 27, 2017 at 16:10
Am I missing something? It does contain additional 0 elements. I still don't see how you can arrive at 4 for the last element. Can you look over your post and make sure your code is correct? — A Magoon
– A Magoon, Commented Jul 27, 2017 at 16:17
@AMagoon I am guessing its like an invalid specifier, i.e. if no second island of 0s found, output the last index. David can you confirm/comment on this? — Divakar
– Divakar, Commented Jul 27, 2017 at 16:20
@A Magoon, the code is correct. Sorry, perhaps I had to explain better this. The explanation is in @Divakar's comment — David
– David, Commented Jul 27, 2017 at 16:23

Divakar · Accepted Answer · 2017-07-27 16:19:39Z

With a play of masks -

def first_zero_index_along_cols(a):
    m1 = a==0
    m2 = (~m1).cumsum(0)>0
    mask = m1 & m2
    return np.where(mask.any(0), mask.argmax(0), a.shape[0]-1)

Sample runs -

Case #1 :

In [503]: a
Out[503]: 
array([[ 0. ,  0. ,  0. ],
       [ 0. ,  3. ,  0. ],
       [ 2. ,  0. ,  0. ],
       [ 0. ,  1.7,  1.5],
       [ 1. ,  1.5,  1.6]])

In [504]: first_zero_index_along_cols(a)
Out[504]: array([3, 2, 4])

Case #2 :

In [505]: a[2:,1] = 0

In [506]: a
Out[506]: 
array([[ 0. ,  0. ,  0. ],
       [ 0. ,  3. ,  0. ],
       [ 2. ,  0. ,  0. ],
       [ 0. ,  0. ,  1.5],
       [ 1. ,  0. ,  1.6]])

In [507]: first_zero_index_along_cols(a)
Out[507]: array([3, 2, 4])

Case #3 :

In [508]: a[:,1] = 0

In [509]: a
Out[509]: 
array([[ 0. ,  0. ,  0. ],
       [ 0. ,  0. ,  0. ],
       [ 2. ,  0. ,  0. ],
       [ 0. ,  0. ,  1.5],
       [ 1. ,  0. ,  1.6]])

In [510]: first_zero_index_along_cols(a)
Out[510]: array([3, 4, 4])

Collectives™ on Stack Overflow

Finding first zero element ignoring leading zeros in numpy array

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related