尝试使用np.random.choice选择索引，然后将这些索引处的值设置为0：

np.random.seed(10)
idx = np.random.choice(np.arange(5*10), size=5, replace=False)

rng.ravel()[idx] = 0

字符串
输出量：

array([[15, 19,  6,  4, 16, 16, 10,  3, 16, 10],
       [ 3,  3,  8, 14,  8, 16,  1,  9, 10, 19],
       [ 5, 16,  2,  0, 15, 11, 18, 15, 18, 16],
       [ 3, 18, 17,  3, 19, 15,  6,  0,  8, 18],
       [15,  5,  0, 17,  0,  6,  3,  0,  5, 10]])

型

当然

idx = np.random.choice(rng.ravel(), 10, replace= False)
print(idx)
rng.ravel()[idx] = 0
rng

字符串
输出

[10 17  3  6 15 15 15 16 15 15]

array([[15, 19,  6,  0, 16, 16,  0,  3, 16, 10],
       [ 0,  3,  8, 14,  8,  0,  0,  0, 10, 19],
       [ 5, 16,  2,  7, 15, 11, 18, 15, 18, 16],
       [ 3, 18, 17,  3, 19, 15,  6,  3,  8, 18],
       [15,  5, 10, 17, 13,  6,  3, 19,  5, 10]], dtype=int64)

型
所以我只取6，而不是取10个零，因为15在我的idx中出现了5次。

也许你可以试试这个：

import numpy as np
from numpy.random import default_rng

rng = default_rng(seed=10)
a=rng.integers(1,20,(5,10)) #creates the initial matrix
print("This is the initial matrix")
print(a)

rep= default_rng(seed=5)
row=rep.integers(0,5,(10,)) 
print("This is the random row index")
print(row)

column=rep.integers(0,10,(10,))
print("This is the random column index")
print(column)
a[row,column]=0 #set the random seats to 0

字符串
你在矩阵中随机得到5个位置，替换为值零（0）