Numpy reorganize array

Question

Let's say I have got the following numpy array

A = np.array([[0,1,2,3,4],[5,6,7,8,9],[10,11,12,13,14],[15,16,17,18,19],[20,21,22,23,24],[25,26,27,28,29]])

out[]: array([[ 0,  1,  2,  3,  4],
       [ 5,  6,  7,  8,  9],
       [10, 11, 12, 13, 14],
       [15, 16, 17, 18, 19],
       [20, 21, 22, 23, 24],
       [25, 26, 27, 28, 29]])

I would like to reorganize it in such a way that the output is

out[] : array([[0,5],
         [1,6],
         [2,7],
         [3,8],
         [4,9],
         [10,15],
         [11,16],
         [12,17],
         [13,18],
         [14,19],
         [20,25],
         [21,26],
         ....,
         [24,29]])

I have been trying different combinations of np.reshape, tranpose, flatten, np.swapaxes, but with no success.

The real array has tens or sometimes hundreds of rows.

Originally, the data is given as DataFrame, but I realized that converting to numpy array could be a better alternative... Can it be done directly using pandas?

So the first, third, etc. row should be the first column and the second, fourth, etc. row should be the second column? — bayes2021
– bayes2021, Commented Jan 11, 2023 at 6:45
Yes, in general for a, let's say 100x5 dimensional array, I would like to group every 10 rows and put them as columns (this subgroup would be 5x10), then the 10 subgroups are later appended, such that I end up with a 50x10 dimensional array — ktitimbo
– ktitimbo, Commented Jan 11, 2023 at 6:58
@KelvinTitimbo You can refer this answer for general logic to solve these kind of problems stackoverflow.com/a/47978032/5462372 — MSS
– MSS, Commented Jan 11, 2023 at 9:43

MSS · Accepted Answer · 2023-01-12 06:42:52Z

1

Yes this can be done in this way using as_strided.

from numpy.lib.stride_tricks import as_strided

A = np.array([[0,1,2,3,4],[5,6,7,8,9],[10,11,12,13,14],[15,16,17,18,19],[20,21,22,23,24],[25,26,27,28,29]])
A = A.reshape(3, 2, 5)
#print(A)
S = A.itemsize
out = as_strided(A, shape=(3,5,2), strides=(2*A.shape[1]*S ,S, A.shape[1]*S)).copy()
out = out.reshape(15,2)
print(out)
>> [[ 0  5]
 [ 1  6]
 [ 2  7]
 [ 3  8]
 [ 4  9]
 [10 15]
 [11 16]
 [12 17]
 [13 18]
 [14 19]
 [20 25]
 [21 26]
 [22 27]
 [23 28]
 [24 29]]

edited Jan 12, 2023 at 6:42

answered Jan 11, 2023 at 9:33

MSS

3,6932 gold badges28 silver badges57 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Numpy reorganize array

1 Answer 1

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related