在python中，numpy.array（matrix [U * V * W]）中的操作与numpy.array（matrix [U * V]）中的操作有何不同？

Question

我的代码适用于2d和3d类型列表数组，但不适用于numpy.ndarray类型的3d数组：

def rotate(lst):
    si = 0
    ei = len(lst)-1
    sj = 0
    ej = len(lst[0])-1
    ix = 0
    jx = 1
    i = 0
    j = 0

    pre = lst[0][0]
    for x in range(1, (len(lst[0])) * (len(lst))):

        if i == ei and ix == 1:  # V
            ix = 0
            jx = -1
            ej -= 1
        elif j == ej and jx == 1:  # >
            ix = 1
            jx = 0
            si += 1
        elif i == si and ix == -1:  # ^
            ix = 0
            jx = 1
            sj += 1
        elif j == sj and jx == -1:  # <
            ix = -1
            jx = 0
            ei -= 1

        print(pre,end=',')
        t = lst[i + ix][j + jx]
        lst[i + ix][j + jx] = pre
        pre = t
        i += ix
        j += jx

    lst[0][0] = pre
    return lst

print(rotate(msg))

提供2D输入时

> msg = numpy.array([
>                     [1,2,3],
>                     [8,9,4],
>                     [7,6,5],
>                   ])

op：

> 1,2,3,4,5,6,7,8,
> 
> [[9 1 2]  [7 8 3]  [6 5 4]]

在输入3d时

> msg = numpy.array([
>                     [[1, 1], [2, 2], [3, 3]],
>                     [[8, 8], [9, 9], [4, 4]],
>                     [[7, 7], [6, 6], [5, 5]],
>                   ])

op：

> [2 2],[2 2],[2 2],[2 2],[2 2],[2 2],[2 2],[2 2], 
>
> [[[2 2]   [2 2]   [2 2]]
>  [[2 2]   [2 2]   [2 2]]
>  [[2 2]   [2 2]   [2 2]]]

我想要以下操作：

> [[[9 9]   [1 1]   [2 2]]
>  [[7 7]   [8 8]   [3 3]]
>  [[6 6]   [5 5]   [4 4]]]

在检查我发现执行lst[i + ix][j + jx] = pre后发现t的值被更改了。