二维数组索引访问，带步长值

Question

I have an array like:

a =np.array([[1,2,3,5,8,7,2,1,3,5,1,2,20,21],[3,8,3,9,8,7,10,1,3,5,1,2,20,21]])

requirement:

need to access index value with given step. For an example step value 3. Only index value not an elements value.

Required Output:

[[0,3,6,9,12],[0,3,6,9,12]]

Answer 1

Use a list comprehension to repeat the range for each row.

a =np.array([[1,2,3,5,8,7,2,1,3,5,1,2,20,21],[3,8,3,9,8,7,10,1,3,5,1,2,20,21]])
indexes = [list(range(0, a.shape[1], 3)) for _ in range(a.shape[0])]

Answer 2

You can do:

a = np.array([[1,2,3,5,8,7,2,1,3,5,1,2,20,21],[3,8,3,9,8,7,10,1,3,5,1,2,20,21]])
step = 3
indices = [np.arange(0, arr.size, step) for arr in a]

print(indices):

[array([ 0,  3,  6,  9, 12]), array([ 0,  3,  6,  9, 12])]

If you want lists instead of np arrays, just use .tolist() like:

[np.arange(0, arr.size, step).tolist() for arr in a]