从更大的2D数组中提取每个2D正方形子数组的对角线

Question

This is my first programming class and I'm very excited to learn Python for Data Science. I cannot figure out how to write a loop that returns all diagonal numbers in the matrix. Below is the code, how close or far off am I? Thank you!

import numpy as np
cols = 0

matrixA = np.array([[2,0,0], [0,3,0], [0,0,4], [6,0,0], [0,7,0], [0,0,8]])

for rows in range(6):
    if rows == cols:
        print(matrixA[rows, cols])
    cols = cols + 1

Answer 1

Your current solution does not work because it does not take into account the fact that matrixA is not square. You will have to take care that your indices do not run out of bounds. Running it gives:

IndexError: index 3 is out of bounds for axis 1 with size 3

This is because the maximum value that cols is allowed to take here is 2 .

---

As an alternative, you could use np.diag :

print(x)
array([[2, 0, 0],
       [0, 3, 0],
       [0, 0, 4],
       [6, 0, 0],
       [0, 7, 0],
       [0, 0, 8]])

res = np.array([np.diag(x, -offset) for offset in range(0, *x.shape)])

print(res)
array([[2, 3, 4],
       [6, 7, 8]])

If you want a 1D result, call np.ravel :

print(res.ravel())
array([2, 3, 4, 6, 7, 8])

Answer 2

You don't need heavy library like numpy to achieve this simple task. In plain python you can do it via using zip and itertools.cycle(...) as:

>>> from itertools import cycle

>>> my_list = [[2,0,0], [0,3,0], [0,0,4], [6,0,0], [0,7,0], [0,0,8]]
>>> for i, j in zip(my_list, cycle(range(3))):
...     print(i[j])
...
2
3
4
6
7
8

Answer 3

Why have cols at all? It's always the same as rows, right?

for rows in range(6):
    print(matrixA[rows,rows])