拥有以下3D阵列(9,9,9):
>>> np.arange(729).reshape((9,9,9))
array([[[ 0 1 2 3 4 5 6 7 8]
[ 9 10 11 12 13 14 15 16 17]
[ 18 19 20 21 22 23 24 25 26]
[ 27 28 29 30 31 32 33 34 35]
[ 36 37 38 39 40 41 42 43 44]
[ 45 46 47 48 49 50 51 52 53]
[ 54 55 56 57 58 59 60 61 62]
[ 63 64 65 66 67 68 69 70 71]
[ 72 73 74 75 76 77 78 79 80]]
...
[[648 649 650 651 652 653 654 655 656]
[657 658 659 660 661 662 663 664 665]
[666 667 668 669 670 671 672 673 674]
[675 676 677 678 679 680 681 682 683]
[684 685 686 687 688 689 690 691 692]
[693 694 695 696 697 698 699 700 701]
[702 703 704 705 706 707 708 709 710]
[711 712 713 714 715 716 717 718 719]
[720 721 722 723 724 725 726 727 728]]])
如何重塑它看起来像这个2D数组(27,27):
答案 0 :(得分:4)
你需要进行6D
一次重塑,基本上将每个轴分成两个,然后转换为向后推偶数轴(第2,第4和第6)到结束和最终重塑回2D
-
a.reshape(-1,3,3,3,3,3).transpose(0,2,4,1,3,5).reshape(27,27)
示例运行 -
In [28]: a = np.arange(729).reshape((9,9,9))
In [29]: out = a.reshape(-1,3,3,3,3,3).transpose(0,2,4,1,3,5).reshape(27,27)
In [30]: out[0]
Out[30]:
array([ 0, 1, 2, 9, 10, 11, 18, 19, 20, 81, 82, 83, 90,
91, 92, 99, 100, 101, 162, 163, 164, 171, 172, 173, 180, 181, 182])
In [31]: out[1]
Out[31]:
array([ 3, 4, 5, 12, 13, 14, 21, 22, 23, 84, 85, 86, 93,
94, 95, 102, 103, 104, 165, 166, 167, 174, 175, 176, 183, 184, 185])
In [32]: out[2]
Out[32]:
array([ 6, 7, 8, 15, 16, 17, 24, 25, 26, 87, 88, 89, 96,
97, 98, 105, 106, 107, 168, 169, 170, 177, 178, 179, 186, 187, 188])
In [33]: out[3]
Out[33]:
array([ 27, 28, 29, 36, 37, 38, 45, 46, 47, 108, 109, 110, 117,
118, 119, 126, 127, 128, 189, 190, 191, 198, 199, 200, 207, 208, 209])
In [34]: out[-1]
Out[34]:
array([546, 547, 548, 555, 556, 557, 564, 565, 566, 627, 628, 629, 636,
637, 638, 645, 646, 647, 708, 709, 710, 717, 718, 719, 726, 727, 728])
通用案例解决方案
BSZ = [3,3] # Block size
p,q = BSZ
out = a.reshape(p,q,p,q,p,q).transpose(0,2,4,1,3,5).reshape(p**3,q**3)
答案 1 :(得分:1)
分而治之非常适用于此类问题。与numpy魔术相比,它更容易理解。这不仅可以解决27 * 27问题,还可以解决所有3个电源问题,如81 * 81的情况。首先显示代码。
import numpy as np
# np.arange(729).reshape((9,9,9)).flatten()
arr = np.arange(729)
result = np.empty((27,27))
def assign(width, start, end, anchor_x, anchor_y):
if width > 3:
sub_width = width / 3
for i in range(3):
for j in range(3):
assign(
sub_width,
start + (i * 3 + j) * sub_width ** 2,
start + (i * 3 + j) * sub_width ** 2 + sub_width ** 2,
anchor_x + i * sub_width,
anchor_y + j * sub_width)
else:
result[anchor_x:anchor_x+3, anchor_y:anchor_y+3] = arr[start:end].reshape(3,3)
assign(27, 0, 729, 0, 0)
print(result)
说明:
每次将一个大的n * n矩阵划分为9个较小的(n / 3)*(n / 3)矩阵,并逐个递归地求解它们。直到矩阵的宽度等于3,停止递归并将9个数字(arr[start:end]
)复制到result[anchor_x:anchor_x+3, anchor_y:anchor_y+3]