2

Having the following 3D array (9,9,9):

np.arange(729).reshape((9,9,9))

[[[  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]]]

How do I reshape it to look like this 2D array (27,27):

[[0 1 2 3 4 5 6 7 8 81 82 83 84 85 86 87 88 89 162 163 164 165 166 167 168 169 170]     
 [9 10 11 12 13 14 15 16 17 90 91 92 93 94 95 96 97 98 171 172 173 174 175 176 177 178 179]
 ...
 [558 559 560 561 562 563 564 565 566 639 640 641 642 643 644 645 646 647 720 721 722 723 724 725 726 727 728]
Improve this question
1
  • From the example you give, it looks like you're doing more than just reshaping -- it also looks like you're transforming the array as well. For example, the first row skips from 8 to 81. What are the transformation rules you are trying to implement? Commented Mar 6, 2017 at 19:10

3 Answers 3

Reset to default
2

You can firstly reshape the array to a 4d array, swap the second and third axises and then reshape it to 27 X 27:

a.reshape(3,3,9,9).transpose((0,2,1,3)).reshape(27,27)

#array([[  0,   1,   2,   3,   4,   5,   6,   7,   8,  81,  82,  83,  84,
#         85,  86,  87,  88,  89, 162, 163, 164, 165, 166, 167, 168, 169,
#        170],
#       [  9,  10,  11,  12,  13,  14,  15,  16,  17,  90,  91,  92,  93,
#         94,  95,  96,  97,  98, 171, 172, 173, 174, 175, 176, 177, 178,
#        179],
#  ... 
#       [558, 559, 560, 561, 562, 563, 564, 565, 566, 639, 640, 641, 642,
#        643, 644, 645, 646, 647, 720, 721, 722, 723, 724, 725, 726, 727,
#        728]])
Improve this answer
Sign up to request clarification or add additional context in comments.

3 Comments

Your solution works perfectly, thank you, but I made a mistake on my question and edited it.
Can you ask a new question instead of editing the current one, Which makes the answer invalid? I can help you with if you give me the link to it.
sorry about that and thank you stackoverflow.com/questions/42652082/…
1 
import numpy as np

x = np.arange(729).reshape(9, 9, 9)

y = x.transpose(1, 0, 2).reshape(27, 27)
y[y[:,2].argsort()]

Explanations

I used numpy.transpose to permute the stride and shape information for each axis.

>>> x.strides
(324L, 36L, 4L)
>>> x.transpose(1, 0, 2).strides
(36L, 324L, 4L)

more info in this answer.

Then I used numpy.reshape the 3D (9L, 9L, 9L) in 2D as expected

>>> x.reshape(27, 27)
(27L, 27L)

Of course, the combination of functions (like transpose and reshape) is very common in numpy. It allows you do to this matrix transformation in a one-liner:

x.transpose(1, 0, 2).reshape(27, 27)

EDIT

As @PaulPanzer point it out, the array was unsorted.

To sort array column by column, one can use:

y[y[:,2].argsort()]

But maybe it isn't the easiest answer anymore.

Improve this answer

5 Comments

Nice. Very clean answer.
@James clean it may well be, but also wrong. Look at the second row of the result.
@PaulPanzer damn you right, I'm wrong ^^' didnt notice
Happens to the best of us ;-)
@PaulPanzer correct it! however, I'm not sure this is the best answer anymore... Thx for pointing this out though
0

If you are ok with moving data in a list you can use:

np.hstack([x for x in np.arange(729).reshape((9,9,9))])
Improve this answer

1 Comment

This also gives a wrong result, never mind the shape being wrong, element #27 in the flattened output is not as specified. Actually, it is wrong in the same way as the other answer.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.