Interleave numpy arrays

Question

I'm trying to interleave arrays as below.

import numpy as np

x = np.array([1,2,3,4,5])
y = np.array([4,6,2,6,9],[5,9,8,7,4],[3,2,5,4,9])

Desired result:

[[1,2,3,4,5],[4,6,2,6,9],[1,2,3,4,5],[5,9,8,7,4],[1,2,3,4,5],[3,2,5,4,9]]

Is there an elegant way to do this?

This is my the way I wrote, but I was looking to improve this line. data=np.array([x,y[0],x,y[1],x,y[2]]) Any other way to write this?

x=np.array([1,2,3,4,5])     
y=np.array([[4,6,2,6,9],[5,9,8,7,4],[3,2,5,4,9]]) 

data=np.array([x,y[0],x,y[1],x,y[2]])
print(data)

In what way are you threading lists? What do you mean with *threading* here? Please provide clear sample input and sample output and explain what the relation is between the input and output. — Willem Van Onsem, Jan 28 '18 at 15:25
I am a newbie still learning, but I do have a data set where I have multiple y-data set (like y above) with shared single x-axis. To plot all together, I'd like to create [x,y1,x,y2,x,y3....] in an elegant way (shorter code?) I'm not sure I do explain well for your response. I really appreciate your help. — mocs, Jan 28 '18 at 15:28

Tai · Accepted Answer · 2018-01-28T16:23:42.787

4

You can try to use np.insert

import numpy as np

x = np.array([1,2,3,4,5])
y = np.array([[4,6,2,6,9],[5,9,8,7,4],[3,2,5,4,9]])
np.insert(y, obj=(0, 1, 2), values=x, axis=0)

array([[1, 2, 3, 4, 5],
       [4, 6, 2, 6, 9],
       [1, 2, 3, 4, 5],
       [5, 9, 8, 7, 4],
       [1, 2, 3, 4, 5],
       [3, 2, 5, 4, 9]])

(0, 1, 2) refers to the indexes in y that you would like to insert into before insertion.

EDIT : One can use obj=range(y.shape[0]) for arbitrary length of y. Thanks for Chiel's suggestion.

Please see the tutorial for more information.

edited Jan 28 '18 at 16:23

answered Jan 28 '18 at 16:10

Tai

7,124
3
25
44

1

You could generalize it even more by replacing the `obj=(0,1,2)` by `obj=range(y.shape[0])`, in case `y` has an arbitrary number of rows. For my taste, you could also remove the argument names for `obj` and `values`. – Chiel Jan 28 '18 at 16:21
@Chiel added. Thank you for the suggestion. – Tai Jan 28 '18 at 16:24

Patrick Artner · Answer 2 · 2018-01-28T16:07:43.740

Adapted from the answer https://stackoverflow.com/a/5347492/7505395 to Interweaving two numpy arrays:

import numpy as np
x=np.array([1,2,3,4,5])     
y=np.array([[4,6,2,6,9],[5,9,8,7,4],[3,2,5,4,9]]) # fixed missing []

x_enh = np.array([x]*len(y)) # blow up x to y size 

c = np.empty((y.size * 2,), dtype=y.dtype).reshape(y.size,5) # create correctly sized empty
c[0::2] = x_enh   # interleave using 2 steps starting on 0
c[1::2] = y       # interleave using 2 steps starting on 1

print(c)

Output:

[[1 2 3 4 5]
 [4 6 2 6 9]
 [1 2 3 4 5]
 [5 9 8 7 4]
 [1 2 3 4 5]
 [3 2 5 4 9]]

Interleave numpy arrays

2 Answers2

Linked