What's an efficient way to pick a specific slice in a high dimensional array

Question

tiwwexx 2021-8-10

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/895867-what-s-an-efficient-way-to-pick-a-specific-slice-in-a-high-dimensional-array

评论： tiwwexx 2021-8-10

Hello all, I have an array that can be anywhere from 9 to 16 dimensional. I want to concatinate one of the dimensions into a previous dimension. For example say I have an array where size(array)=[100 100 100 5 5 5 2]. I can "get rid of" the 7th dimension of size 2 by using cat(2, array(:,:,:,:,:,:,1),array(:,:,:,:,:,:,2)). However, Using all of those colins as place holders is very cumbersome and the size of my array can vary but I always want to concatinate the 7th dimension into the second dimension of the array. Does anyone know a slick way to do this?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Dave B 2021-8-10

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/895867-what-s-an-efficient-way-to-pick-a-specific-slice-in-a-high-dimensional-array#answer_764352

在 MATLAB Online 中打开

That's a scary looking array!

Not sure how slick it is, but if you're looking to turn those 6 colons into a number 6, you can leverage subsref. It makes more code but maybe easier to parameterize it if that's your goal?

a=rand(8, 7, 6, 5, 4, 3, 2); % using a smaller array than yours...
subsref(a,substruct('()', [repelem({':'},6) 1]));

(Just a proof of concept that it works):

b=cat(2, a(:,:,:,:,:,:,1),a(:,:,:,:,:,:,2));
c=cat(2, ...
    subsref(a,substruct('()',[repelem({':'},6) 1])), ...
    subsref(a,substruct('()',[repelem({':'},6) 2])));
isequal(b,c)
ans = logical
   1

Something to consider: is there a form of reshape that solves this problem for you?

d = cat(3,repmat(1,4),repmat(2,4),repmat(3,4))
d = 
d(:,:,1) =

     1     1     1     1
     1     1     1     1
     1     1     1     1
     1     1     1     1


d(:,:,2) =

     2     2     2     2
     2     2     2     2
     2     2     2     2
     2     2     2     2


d(:,:,3) =

     3     3     3     3
     3     3     3     3
     3     3     3     3
     3     3     3     3
dreshape=reshape(d,4,12)
dreshape = 4×12
     1     1     1     1     2     2     2     2     3     3     3     3
     1     1     1     1     2     2     2     2     3     3     3     3
     1     1     1     1     2     2     2     2     3     3     3     3
     1     1     1     1     2     2     2     2     3     3     3     3