How to use np.transpose()

Question

I am trying to understand how to use np.transpose() properly.

I have a np array of shape (4,28,28,8,8). So this is 4 images of shape (224,224) that I viewed as shown by the previous shape.

I would like to revert back to (4,224,224). I feel the best way to go about is to use np.transpose() and reshape() functions. But I am hitting a roadblock as to how to revert back correctly.

Help. Please and thank you.

EDIT: (4,224,224) is 4 (this variable is subject to change as the number of images I load can change it can be 4, it can be 1000) images of shape (224,224). I used listdir() to load images. While loading I resized to the current shape of (224,224). I am going to perform operations on the shape (4,28,28,8,8) which is technically, 4 images, of shape (224,224) broken into (28,28) blocks, each containing (8,8) blocks. This shape I got by using view_as_blocks provided by scikit-image. Once I perform the operations which require that shape, I must revert back to (4,224,224). Where I am stuck.

What is the meaning of each dimension in `(4,28,28,8,8)`? This affects the answer. — Jeremy McGibbon, Oct 03 '17 at 16:56
Do you want to reshape from (4,28,8,28,8)? Transpose parameter `(0,1,3,2,4)` should work. — hpaulj, Oct 03 '17 at 17:02
@JeremyMcGibbon I have edited the question with an explanation of how I got the shape and why I need to revert back. — TheTank, Oct 03 '17 at 17:07
How's this different from your Aug 22 question? It was marked a duplicate, with a transpose answer, https://stackoverflow.com/q/41510395/901925 — hpaulj, Oct 04 '17 at 03:31

score 1 · Accepted Answer · answered Oct 03 '17 at 17:03

1

What you need is probably :

b=np.swapaxes(a,3,2).reshape(4,28*8,28*8)

a beeing your array.

answered Oct 03 '17 at 17:03

B. M.

18,243
2
35
54

So there is no use of using transpose at all? – TheTank Oct 03 '17 at 17:08
@JeremyMcGibbon Can you please suggest a way I can convert (4, 28, 28, 8, 8) to (4, 224, 224)? Or how to access the inner blocks? – TheTank Oct 03 '17 at 17:14
This solution looks like it should work. `swapaxes` does the job of `transpose` here. – Jeremy McGibbon Oct 03 '17 at 17:15
`transpose`is good for 2 dimensions; for more dimensions, it reverses all the dimensions. `swapaxes` is more clear in your case, but you can also use `transpose(a,(0,1,3,2,4))`. – B. M. Oct 04 '17 at 13:01
@B.M. I used transpose in the end but this worked too. Thank you. – TheTank Oct 06 '17 at 16:19

How to use np.transpose()

1 Answers1