numpy.squeeze()的用法

import numpy as np
x = np.array([[[0], [1], [2]]])
print('x.shape:',x.shape)  
x1 = np.squeeze(x)# 从数组的形状中删除单维条目，即把shape中为1的维度去掉
print('x1.shape',x1.shape)  
x2 = np.squeeze(x,2)# 从数组的形状中删除单维条目，三维删除
print('x2.shape',x2.shape)  
x3 = np.squeeze(x,0)# 从数组的形状中删除单维条目，一维删除
print('x3.shape',x3.shape)  

降维结果：

x.shape: (1, 3, 1)
x1.shape (3,)
x2.shape (1, 3)
x3.shape (3, 1)