【python】numpy.random.shuffle()

• 原地打乱，无返回值；
• 多维度的array，只对第一维进行打乱。（二维矩阵解释为对行向量shuffle）

打乱一个行向量组成的矩阵：

import numpy as np

matrix = np.expand_dims(np.arange(4), 0).repeat(2,axis=0).T
print("before:\n",matrix)
#before:
# [[0 0]
# [1 1]
# [2 2]
# [3 3]]
np.random.shuffle(matrix)
print("after:\n",matrix)
#after:
# [[3 3]
# [1 1]
# [2 2]
# [0 0]]

打乱一个列向量组成的矩阵：
方法一：稍麻烦

import numpy as np

matrix = np.expand_dims(np.arange(10), 0).repeat(2,axis=0)
print("before:\n",matrix)
# before:
# [[0 1 2 3 4 5 6 7 8 9]
# [0 1 2 3 4 5 6 7 8 9]]

index = np.arange(len(matrix[1]))
#index: [0 1 2 3 4 5 6 7 8 9]

np.random.shuffle(index)
#shuffled index: [9 8 4 2 0 1 7 6 3 5]

shuffled_matrix = matrix[:, index]
print("after:\n",shuffled_matrix)
#after:
# [[9 8 4 2 0 1 7 6 3 5]
# [9 8 4 2 0 1 7 6 3 5]]

方法二：简单些

import numpy as np

matrix = np.expand_dims(np.arange(10), 0).repeat(2,axis=0)
print("before:\n",matrix)
# before:
#  [[0 1 2 3 4 5 6 7 8 9]
#  [0 1 2 3 4 5 6 7 8 9]]

np.random.shuffle(matrix.T)
print("after:\n",matrix)
# after:
#  [[9 6 1 3 8 5 7 0 4 2]
#  [9 6 1 3 8 5 7 0 4 2]]