Python numpy.squeeze()用法及代碼示例

當我們要從數組形狀中刪除一維條目時，將使用numpy.squeeze()函數。

用法： numpy.squeeze(arr, axis=None )

代碼1：

# Python program explaining 
# numpy.squeeze function 
  
import numpy as geek 
  
in_arr = geek.array([[[2, 2, 2], [2, 2, 2]]]) 
   
print ("Input array:", in_arr)  
print("Shape of input array:", in_arr.shape)   
  
out_arr = geek.squeeze(in_arr)  
  
print ("output squeezed array:", out_arr) 
print("Shape of output array:", out_arr.shape) 

輸出：

Input array: [[[2 2 2]
  [2 2 2]]]
Shape of input array: (1, 2, 3)
output squeezed array: [[2 2 2]
 [2 2 2]]
Shape of output array: (2, 3)

代碼2：

# Python program explaining 
# numpy.squeeze function 
import numpy as geek 
in_arr = geek.arange(9).reshape(1, 3, 3)  
  
print ("Input array:", in_arr)   
out_arr = geek.squeeze(in_arr, axis = 0)  
  
print ("output array:", out_arr)   
print("The shapes of Input and Output array:")  
  
print(in_arr.shape, out_arr.shape)

輸出：

Input array: [[[0 1 2]
  [3 4 5]
  [6 7 8]]]
output array: [[0 1 2]
 [3 4 5]
 [6 7 8]]
The shapes of Input and Output array:
(1, 3, 3) (3, 3)

注意：

ValueError:
If axis is not None, and an axis being squeezed is not of length 1.

代碼3：

# Python program explaining 
# numpy.squeeze function 
# when value error occurs 
import numpy as geek 
  
in_arr = geek.arange(9).reshape(1, 3, 3)  
  
print ("Input array:", in_arr)   
out_arr = geek.squeeze(in_arr, axis = 1)  
  
print ("output array:", out_arr)   
print("The shapes of Input and Output array:") 
   
print(in_arr.shape, out_arr.shape)

輸出：

ValueError                                Traceback (most recent call last)
 in ()
      5 
      6 print ("Input array:", in_arr)
----> 7 out_arr = geek.squeeze(in_arr, axis=1)
      8 print ("output array:", out_arr)
      9 print("The shapes of Input and Output array:")

~\Anaconda3\lib\site-packages\numpy\core\fromnumeric.py in squeeze(a, axis)
   1196     try:
   1197         # First try to use the new axis= parameter
-> 1198         return squeeze(axis=axis)
   1199     except TypeError:
   1200         # For backwards compatibility

ValueError:cannot select an axis to squeeze out which has size not equal to one