Improve performance of code to solve equation

Question

Sitherion 2016-12-8

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/315959-improve-performance-of-code-to-solve-equation

评论： Walter Roberson 2016-12-8

在 MATLAB Online 中打开

I am quite novice to MATLAB and I am struggling to find a solution to this equation.

Where the matrices' dimensions are {λ} = N×K, {Y} = N×D, {π} = 1×K, and {μ} = D×K.

What I have created looks more like a monstrosity than an efficient MATLAB code, and that is due to a serious lack of MATLAB skills.

Actually, I decided to solve this step by step, in order to get myself familiarised with how MATLAB works. I ended up with an ultra inefficient code that I cannot even combine it. Even the final result is wrong, as I want to create a NxK matrix. Any guidance would be much appreciated.

%Dimensions:
nn = 10;
dd = 7;
kk = 5;
%Initial variables:
lambda0 = rand(nn,kk);
sigma = rand(1);
Y = rand(nn,dd);
mu = rand(dd,kk);
%Calculate pies:
First_part = log(pie(:)./(1.-pie(:))); % Kx1 vector
%Calculate Second part:
for n = 1:nn
for d = 1:dd
lambda_mu(d,:) = lambda0(n,:).*mu(d,:); %lambdamu is a DxK matrix
end
lambda_mu2(n,:) = sum(lambda_mu,2); %This is a NXD matrix
end
%Y-lambdamu2:
for n = 1:nn
YY(n,:) = Y(n,:)-lambda_mu2(n,:); % This is a NxD vector
end
%YY*mu:
Second = (YY*mu)./sigma^2; % NxK
%Third:
Third = (mu'*mu)./2*sigma^2; % KxK
%Final:
for n=1:nn
Final_part = transpose(First_part(:))+Second(n,:)-Third;
end

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

Walter Roberson 2016-12-8

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/315959-improve-performance-of-code-to-solve-equation#answer_246437

Your notation is not consistent with the possibility of y, mu, and lambda all being 2D matrices with subscript indicating columns and superscript (n) indicating row. Your sigma would have to be summing columns in that case, but the sum would have to be subtracted from the row y superscript (n).

The intended meaning of your superscripts and subscripts in the equation are not clear.