Horizontally concatinating the fields of a structure with different dimensions.

Question

ARN 2019-1-14

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/439663-horizontally-concatinating-the-fields-of-a-structure-with-different-dimensions

评论： Stephen23 2019-1-14

采纳的回答： Stephen23

i have a [1*1] structure containing 30 fields,

variable00,,...... variable29 with size [8193*2, 8192*2....8189*2] (random size every time)

I want to concatinate all the data in one field 'allvariable' but with the dimension missmatch i am unable to do that.

Also is there a way to concatinate only the first rows?

One random data picture is attached.

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

Stephen23 2019-1-14

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/439663-horizontally-concatinating-the-fields-of-a-structure-with-different-dimensions#answer_356299

编辑：Stephen23 2019-1-14

The different number of rows does not prevent them from being concatenated vertically:

>> S.f1 = [0,1;2,3;4,5];
>> S.f2 = [6,7;8,9];
>> C = struct2cell(S);
>> vertcat(C{:})
ans =
   0   1
   2   3
   4   5
   6   7
   8   9

If you want to concatenate horizontally either use indexing:

>> D = cellfun(@(m)m(1:2,:),C,'uni',0);
>> horzcat(D{:})
ans =
   0   1   6   7
   2   3   8   9

or download catpad. from FEX:

>> catpad(2,C{:})
ans =
     0     1     6     7
     2     3     8     9
     4     5   NaN   NaN

2 个评论
显示无隐藏无

ARN 2019-1-14

Thanks! desired answer is what i get from catpad but with zeros instead of NaN in the answer which i got from following.

g(isnan(g))=0;

Isn't there any simple way to do that? Lets say i am not allowed to use catpad.

Stephen23 2019-1-14

"Isn't there any simple way to do that? Lets say i am not allowed to use catpad."

The simple solution is to use catpad (it already exists, it works, it only requires one/two lines for you to get what you need). If you really do not want to use catpad then you could certainly use a few loops (or a few cellfun calls) to get the size of each matrix, then pad them as required, and then afterwards concatenate together. You would have to decide if that is "simple" or not.

请先登录，再进行评论。