Python numpy.diag_indices()

Python numpy.diag_indices()函数返回下标，以访问最小维度为2的数组主对角线上的元素。以元组的形式返回索引。
访问数组的主对角线。

语法:

numpy.diag_indices(n, n_dim = 2)

参数 :

n :数组的大小，每个维度都需要有二个元素的索引。
n_dim : [int, optional]维数的数目。

返回 :

Indices(as tuples) to access diagonal elements.

代码 1 :

# Python Program illustrating
# working of diag_indices()
 
import numpy as geek
 
# Creates a 5 X 5 array and returns indices of
# main diagonal elements
d = geek.diag_indices(5)
print("Indices of diagonal elements as tuple : ")
print(d, "\n")
 
array = geek.arange(16).reshape(4,4)
print("Initial array : \n", array)
 
# Here we can manipulate diagonal elements
# by accessing the diagonal elements
d = geek.diag_indices(4)
array[d] = 25
print("\n New array : \n", array)

输出 :

Indices of diagonal elements as tuple : 
(array([0, 1, 2, 3, 4]), array([0, 1, 2, 3, 4])) 

Initial array : 
 [[ 0  1  2  3]
 [ 4  5  6  7]
 [ 8  9 10 11]
 [12 13 14 15]]

 New array : 
 [[25  1  2  3]
 [ 4 25  6  7]
 [ 8  9 25 11]
 [12 13 14 25]]

代码2:操作2D数组

# Python Program illustrating
# working of diag_indices()
 
import numpy as geek
 
# Manipulating a 2D array
d = geek.diag_indices(3, 2)
 
array = geek.arange(12).reshape(4, 3)
 
array[d] = 111
print("Manipulated array : \n", array)

输出 :

Manipulated array : 
 [[111   1   2]
 [  3 111   5]
 [  6   7 111]
 [  9  10  11]]

代码3:操作3D数组

# Python Program illustrating
# working of diag_indices()
 
import numpy as geek
 
# Setting diagonal indices
d = geek.diag_indices(1, 2)
print("Diag indices : \n", d)
 
# Creating a 3D array with all ones
array = geek.ones((2, 2, 2), dtype=geek.int)
print("Initial array : \n", array)
 
# Manipulating a 3D array
array[d] = 0
print("New array : \n", array)

输出 :

Diag indices : 
 (array([0]), array([0]))
Initial array : 
 [[[1 1]
  [1 1]]

 [[1 1]
  [1 1]]]
New array : 
 [[[0 0]
  [1 1]]

 [[1 1]
  [1 1]]]