How to combine 2 array by turns from
arr_1 = np.full((5,), 0)
arr_2 = np.full((5,), 1)
to
[0,1,0,1,0,1,0,1,0,1]
? thanks
CodePudding user response：
Given two arrays you can stack and then transpose them. Then just flatten it out:
import numpy as np
arr_1 = np.full((5,), 0)
arr_2 = np.full((5,), 1)
np.stack([arr_1, arr_2]).T.ravel()
#array([0, 1, 0, 1, 0, 1, 0, 1, 0, 1])
This works by first stacking:
np.stack([arr_1, arr_2])
# array([[0, 0, 0, 0, 0],
# [1, 1, 1, 1, 1]])
Then transposing:
np.stack([arr_1, arr_2]).T
# array([[0, 1],
# [0, 1],
# [0, 1],
# [0, 1],
# [0, 1]])
Then ravel() reads them off the contiguous flattened array.
Edit
As Nin17 points out in the comments stack() takes an axis argument, which avoids the transpose:
np.stack([arr_1, arr_2], 1).ravel()
# array([0, 1, 0, 1, 0, 1, 0, 1, 0, 1])
This is a nice performance improvement.
CodePudding user response：
You can do that like so:
import numpy as np
a = np.array([0, 1])
repeated = np.tile(a, 5)
Result:
array([0, 1, 0, 1, 0, 1, 0, 1, 0, 1])