How to form a matrix with the existing 7 matrices with following rules

Question

Tianshu Gao 2020-4-26

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/520783-how-to-form-a-matrix-with-the-existing-7-matrices-with-following-rules

评论： Tianshu Gao 2020-4-26

Now, I have 7 matrices (with specific values) in total, W(6x6), C1(3x3), C2(3x3), C3(3x3)，C4(3x3)，C5(3x3)，C6(3x3). I want to form a matrix A (18x18) with following rules:

A= [w(1,1)*C1 w(1,2)*C1 w(1,3)*C1 w(1,4)*C1 w(1,5)*C1 w(1,6)*C1

w(2,1)*C2 w(2,2)*C2 w(2,3)*C2 w(2,4)*C2 w(2,5)*C2 w(2,6)*C2

w(3,1)*C3 w(3,2)*C3 w(3,3)*C3 w(3,4)*C3 w(3,5)*C3 w(3,6)*C3

w(4,1)*C4 w(4,2)*C4 w(4,3)*C4 w(4,4)*C4 w(4,5)*C2 w(4,6)*C4

w(5,1)*C5 w(5,2)*C5 w(5,3)*C5 w(5,4)*C5 w(5,5)*C5 w(5,6)*C5

w(6,1)*C6 w(6,2)*C6 w(6,3)*C6 w(6,4)*C6 w(6,5)*C6 w(6,6)*C6]

Do I have a way to form matrix A without typing them manually? Because, in the future, I would probability solve 1000 C matrices, which is time consumming to type in manually. Thank you.

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Stephen23 2020-4-26

编辑：Stephen23 2020-4-26

"Do I have a way to form matrix A without typing them manually?"

Of course, it is easy once you avoid using numbered variables.

"Because, in the future, I would probability solve 1000 C matrices, which is time consumming to type in manually."

Yes, that would be very time consuming. That is why no experienced MATLAB user defines lots of numbered variable names: because it wastes their time (code writing time, debugging time, runtime). Indexing is so much simpler and more efficient.

https://www.mathworks.com/matlabcentral/answers/304528-tutorial-why-variables-should-not-be-named-dynamically-eval

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Stephen23 2020-4-26

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/520783-how-to-form-a-matrix-with-the-existing-7-matrices-with-following-rules#answer_428430

编辑：Stephen23 2020-4-26

在 MATLAB Online 中打开

This is easiest when you avoid anti-pattern numbered variables and use one array, e.g. a cell array:

W = rand(6,6);
C = {rand(3,3),rand(3,3),rand(3,3),rand(3,3),rand(3,3),rand(3,3)}; % one array
N = numel(C);
A = kron(W,ones(3,3)).*repmat(vertcat(C{:}),1,N); % or REPELEM instead of KRON

Checking the first output submatrix (the others you can check yourself):

>> size(A)
ans =
   18   18
>> A(1:3,1:3)
ans =
   0.181146   0.014799   0.190315
   0.100409   0.250922   0.047237
   0.046128   0.018101   0.180526
>> W(1,1)*C{1}
ans =
   0.181146   0.014799   0.190315
   0.100409   0.250922   0.047237
   0.046128   0.018101   0.180526