random.shuffer与random.permutation 区别

np.random.shuffle(arr) #在原数组上改变顺序,无返回值 同random.shuffle()
arr = np.random.permutation([1, 4, 9, 12, 15]) #不在原数组上进行，返回新的数组，不改变自身数组

import numpy as np

arr = np.arange(10) 

print(arr) #[0 1 2 3 4 5 6 7 8 9]

np.random.shuffle(arr) #在原数组上改变顺序,无返回值

print(arr) #[9 5 8 2 6 1 4 0 7 3]

arr = np.arange(12).reshape(3,4)

print(arr)
# [[ 0  1  2  3]
#  [ 4  5  6  7]
#  [ 8  9 10 11]]

np.random.shuffle(arr)  #多维时,默认是列维度

print(arr)
# [[ 0  1  2  3]
#  [ 8  9 10 11]
#  [ 4  5  6  7]]

arr = np.random.permutation(10) #可直接生成一个随机排列的数组

print(arr) #[3 5 9 8 1 6 0 4 7 2]

arr = np.random.permutation([1, 4, 9, 12, 15]) #不在原数组上进行，返回新的数组，不改变自身数组

print(arr) #[ 1  4 12  9 15]

#应用 当训练数据与label分别保存在不同变量中时,可以用来同步乱序再训练

X = np.array(["A","B","C","D","E","F"])

Y = np.array(["a","b","c","d","e","f"])

ind = range(len(X))

new_ind = np.random.permutation(ind)

new_x = X[new_ind]

new_y = Y[new_ind]

print(new_x) #['D' 'A' 'B' 'F' 'C' 'E']

print(new_y) #['d' 'a' 'b' 'f' 'c' 'e']

arr = np.arange(9).reshape((3, 3))

print(arr)
# [[0 1 2]
#  [3 4 5]
#  [6 7 8]]

arr2 = np.random.permutation(arr)  #同样默认列维度

print(arr)
# [[0 1 2]
#  [3 4 5]
#  [6 7 8]]

print(arr2)
# [[0 1 2]
#  [6 7 8]
#  [3 4 5]]

转载于:https://my.oschina.net/u/3726752/blog/3089264