Repeating each element of a numpy array 5 times

2019-01-11 13:35发布

问题:

import numpy as np

data = np.arange(-50,50,10)
print data

[-50 -40 -30 -20 -10   0  10  20  30  40]

I want to repeat each element of data 5 times and make new array as follows:

ans = [-50 -50 -50 -50 -50 -40 -40 ... 40]

How can I do it?

What about repeating the whole array 5 times?

ans =  [-50 -40 -30 -20 -10   0  10  20  30  40 -50 -40 -30 -20 -10   0  10  20  30  40 -50 -40 -30 -20 -10   0  10  20  30  40 -50 -40 -30 -20 -10   0  10  20  30  40 -50 -40 -30 -20 -10   0  10  20  30  40 .......]

回答1:

In [1]: data = np.arange(-50,50,10)

To repeat each element 5 times use np.repeat:

In [3]: np.repeat(data, 5)
Out[3]: 
array([-50, -50, -50, -50, -50, -40, -40, -40, -40, -40, -30, -30, -30,
       -30, -30, -20, -20, -20, -20, -20, -10, -10, -10, -10, -10,   0,
         0,   0,   0,   0,  10,  10,  10,  10,  10,  20,  20,  20,  20,
        20,  30,  30,  30,  30,  30,  40,  40,  40,  40,  40])

To repeat the array 5 times use np.tile:

In [2]: np.tile(data, 5)
Out[2]: 
array([-50, -40, -30, -20, -10,   0,  10,  20,  30,  40, -50, -40, -30,
       -20, -10,   0,  10,  20,  30,  40, -50, -40, -30, -20, -10,   0,
        10,  20,  30,  40, -50, -40, -30, -20, -10,   0,  10,  20,  30,
        40, -50, -40, -30, -20, -10,   0,  10,  20,  30,  40])

Note, however, that sometimes you can take advantage of NumPy broadcasting instead of creating a larger array with repeated elements.

For example, if

z = np.array([1, 2])
v = np.array([[3], [4], [5]])

then to add these arrays to produce

 [[4 5]
  [5 6]
  [6 7]]

you do not need to use tile:

In [12]: np.tile(z, (3,1))
Out[12]: 
array([[1, 2],
       [1, 2],
       [1, 2]])

In [13]: np.tile(v, (1,2))
Out[13]: 
array([[3, 3],
       [4, 4],
       [5, 5]])

In [14]: np.tile(z, (3,1)) + np.tile(v, (1,2))
Out[14]: 
array([[4, 5],
       [5, 6],
       [6, 7]])

Instead, NumPy will broadcast the arrays for you:

In [15]: z + v
Out[15]: 
array([[4, 5],
       [5, 6],
       [6, 7]])


回答2:

Simply use np.repeat:

In [5]: data.repeat(5)

Out[5]: 
array([-50, -50, -50, -50, -50, -40, -40, -40, -40, -40, -30, -30, -30,
       -30, -30, -20, -20, -20, -20, -20, -10, -10, -10, -10, -10,   0,
         0,   0,   0,   0,  10,  10,  10,  10,  10,  20,  20,  20,  20,
        20,  30,  30,  30,  30,  30,  40,  40,  40,  40,  40])