Error when I concatenated matrices

Question

Viridiana Torres 2016-4-27

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/281253-error-when-i-concatenated-matrices

评论： Viridiana Torres 2016-4-27

Hello! does anyone know why when I try to execute this command I got the next error

Error using horzcat
Dimensions of matrices being concatenated are not consistent
Error in @(x)[x,nan(1,max(m)-numel(x))]

where

m=cellfun(@numel,R);
B = cellfun(@(x)[x,nan(1,max(m)-numel(x))],R,'un',0);

where m is 30*1 double and R is 30*1 cell composed by numbers

Thanks a lot!

3 个评论
显示 1更早的评论隐藏 1更早的评论

Muhammad Usman Saleem 2016-4-27

your matrix are not favorable for multiplication. Read some basics about multiplication of A and B matrix.

Rows of matrix A must be same as columns of matrix B.

Viridiana Torres 2016-4-27

Thanks muhammad my mistake is not about how to multiplicate A and B, that is something that I know. My mistake is related with how to interpret the codes in Matlab.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Stephen23 2016-4-27

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/281253-error-when-i-concatenated-matrices#answer_219620

编辑：Stephen23 2016-4-27

在 MATLAB Online 中打开

This error can occur whenever any of the elements of R are column vectors or matrices:

>> R = {[1;2;3;4],[5;6]};
>> m = cellfun(@numel,R)
m =
     4     2
>> B = cellfun(@(x)[x,nan(1,max(m)-numel(x))],R,'un',0)
Error using horzcat
Dimensions of matrices being concatenated are not consistent.
Error in @(x)[x,nan(1,max(m)-numel(x))]

Think about it: you use numel to get the total number of elements in the array and use this to pad the array using a row vector so NaN's, but you always pad using horizontal concatenation [ , ]. But note that numel does not say anything about the orientation of the vectors, or how many rows/columns/pages/... an array has. So you could end up trying to concatenate a column vector and a row vector. Or an array and a row vector.

Basically you are doing this:

>> [[5;6],nan(1,2)]
Error using horzcat
Dimensions of matrices being concatenated are not consistent.

But you need to think about the shape of the two arrays that you are trying to concatenate:

>> [5;6]
ans =
     5
     6
>> nan(1,2)
ans =
   NaN   NaN

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Viridiana Torres 2016-4-27

Dear Stephen thanks a lot for your answer and the way you explain it, I really appreciate when someone tries to explain properly its answer. Thanks a lot!

请先登录，再进行评论。