Suscripted Assignment in to Mxarray is not supported for code generation

Question

Joseph 2024-9-6

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2150709-suscripted-assignment-in-to-mxarray-is-not-supported-for-code-generation

回答： MULI 2024-9-10

Gamma=zeros(n_y*Np,n);
cont=1;
for k=1:n_y:n_y*Np
    row_i=k;
    row_f=k+n_y-1;
    
    Gamma(row_i:row_f,:)=C*A^(cont);
    cont=cont+1;
end

outputs.

Error:Subscripted assignment into an mxArray is not supported for code generation.

Function 'MATLAB Function' (#484.1885.1905), line 85, column 5:

"Gamma(row_i:row_f,:)"

Launch diagnostic report.

2 个评论
显示无隐藏无

Shivam Gothi 2024-9-6

Hello @Joseph,

Can you please share the complete code so that I can reproduce the error from my end and try to get a possible work around for it?

Walter Roberson 2024-9-6

I notice that each iteration of the loop, it sets n_y rows to a value, and the next iteration overwrites all but one of the rows. This feels like a waste of time.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

MULI 2024-9-10

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2150709-suscripted-assignment-in-to-mxarray-is-not-supported-for-code-generation#answer_1513589

Hi Joseph,

I understand that you are facing an error when trying to convert MATLAB code to C/C++ using MATLAB Coder.

This error is related to dynamic indexing that MATLAB Coder cannot handle due to its requirement for fixed-size arrays due to the following reasons:

Dynamic Indexing: Even though `Gamma` is preallocated with a fixed size, the assignment the line “Gamma(row_i:row_f, :) = C * A^(cont)” involves dynamic indexing, which MATLAB Coder cannot handle.

Matrix Operation Compatibility: The result of “C * A^con” must match the size of the submatrix “Gamma(row_i:row_f, :)”. If the dimensions do not match, it leads to an error.

To resolve this issue, you may follow these steps:

Make sure that the matrices `C` and `A` have fixed sizes. For instance, if `C` is `(n_y, m)` and `A` is `(m, m)`, the result `C * A^cont` should be `(n_y, m)`.
Ensure that `m` equals `n` so that the result of the multiplication fits into the submatrix of `Gamma`.