Access matrix outside of for loop

Question

Alexandra Locherer 2022-9-1

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1793075-access-matrix-outside-of-for-loop

评论： Sushranth Prakash Hebbar 2022-9-5

Hello everybody,

I count signals and co-localizations of these signals in big images. I do this with nested loops (I now it's not the optimal way how to do it) which make my code very slow. I want to use parfor to accelerate the run time.

I made a Matrix outside my main for loop in which I save the numbers. After the main for loop I take this Matrix and save it in an ecxel table. By using parfor I'm no longer abel to define my matrix outside of the loop. This gives me the problem that I can't acces my data in the end.

Is there a way to acces data generated in an for loop outside of it?

Thank you for any tips

imagefilesEpo = dir('*Alexa Fluor 488.tif');
imagefilesDCN = dir('*Alexa Fluor 647.tif');
nfiles=length(imagefilesEpo);
Matrix_with_Number_of_colocalization=zeros([nfiles 4]);
parpool(3)
parfor jj=1:nfiles
    
     
%    Epo = imagefilesEpo(jj);
    currentfilenameEpo = imagefilesEpo (jj).name;
       
    currentimageEpo = imread(currentfilenameEpo);
    Epo = currentimageEpo;
%     DCN = imagefilesDCN(jj);
    currentfilenameDCN = imagefilesDCN (jj).name;
      
    currentimageDCN = imread(currentfilenameDCN);
    DCN = currentimageDCN;
    %% Epo und DCN zeigen
%     figure('Name','Overlapping Epo - DCN, both unmodified')
%     imshowpair(Epo, DCN)
    
    %% blur
    
    %Epo =imgaussfilt(Epo,5);
    
    %% Threshold Epo with 0.35
    
    Epo035 = imbinarize(Epo, 0.35);
    
    %% max size Epo
    
    Epo035max1900 = bwareafilt (Epo035, [1,1900]);
    
    
    %% Threshold DCN with 0.25
    
    DCN025 = imbinarize(DCN, 0.25); 
    
    %% Boundarys
    % remove the % below only if you want to see the boarder on the picture
    
    [Epo_R,Epo_C]=bwboundaries(Epo035max1900,'noholes');
        % figure;
        % imshow(Epo035max1900)
        % hold on
        % for k = 1:length(Epo_R)
        %    boundary = Epo_R{k};
        %    plot(boundary(:,2), boundary(:,1), 'r', 'LineWidth', 0.002)
        % end
    
    [Marker_R,Marker_C]=bwboundaries(DCN025,'noholes');
        % figure;
        % imshow(DCN025)
        % hold on
        % for k = 1:length(Marker_R)
        %    boundary = Marker_R{k};
        %    plot(boundary(:,2), boundary(:,1), 'r', 'LineWidth', 0.002)
        % end
    
    %% anzahl column in Matrix M = anzahl zusammenhängende elemente
    
    rEpo=size(Epo_R,1);
    rDCN=size(Marker_R,1);
    %% anzahl pixel cxcl14015
    pixelDCN025 = sum(DCN025(:) == 1);
    
    
    %% Koordinaten der zusammenhängende Regionen von Epo als vektoren speichern
    
    Koordinaten_Epo=zeros();
    a=1;
    b=2;
    
    for i=1:rEpo
        [E_row,E_col]=find(Epo_C==i);
        for j=1:size(E_row)
        Koordinaten_Epo(j,a)=E_row(j);
        end
        for k=1:size(E_col)
        Koordinaten_Epo(k,b)=E_col(k);
        end
        a=a+2;
        b=b+2;
    
    end
    
    
    %% Koordinaten von Marker in array speichern
    
    Koordinaten_Marker=zeros();
    
    [M_row,M_col]=find(Marker_C>=1);
    
    for l=1:pixelDCN025
        Koordinaten_Marker(l,1)=M_row(l);
        Koordinaten_Marker(l,2)=M_col(l);
    end
    
    
    %% colocalization zählen
    
    Num_colocalization=0;
    o=1;
    p=2;
    
   
    while p <=size(Koordinaten_Epo,2)
        for m=1:size(Koordinaten_Marker,1)
            if p <=size(Koordinaten_Epo,2)
                for q=1:size(Koordinaten_Epo,1)
                    if (Koordinaten_Marker(m,1) == Koordinaten_Epo(q,o)) && (Koordinaten_Marker(m,2) == Koordinaten_Epo(q,p))
                        Num_colocalization=Num_colocalization+1;
                        p=p+2;
                        o=o+2;
                        break
                    end
                end
            end
        end
            if p <=size(Koordinaten_Epo,2)
                o=o+2;
                p=p+2;
            end
    end
    
     % saving the datas in a table 
    Matrix_with_Number_of_colocalization(jj,1)=Num_colocalization;
    Matrix_with_Number_of_colocalization(jj,2)=rEpo;
    Matrix_with_Number_of_colocalization(jj,3)=rDCN;
    percentage=100/(rEpo/Num_colocalization);
    Matrix_with_Number_of_colocalization(jj,4)=percentage;
    row_name(jj)={strcat(extractBetween(currentfilenameEpo,1,29))};
    
end
% save the data
row_name=row_name';
RowName=cell2table(row_name);
Table_with_Number_of_colocalization=array2table(Matrix_with_Number_of_colocalization);
Colocalisation_EPO_DCN=[RowName,Table_with_Number_of_colocalization];
Colocalisation_EPO_DCN.Properties.VariableNames = ["Experiment","Colocalization","Epo","DCN","Colocalization in %"];
% save the final table as excel
filename = 'Colocalisation_EPO_DCN.xlsx';
writetable(Colocalisation_EPO_DCN,filename,'Sheet',1,'Range','A1');

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

Sushranth Prakash Hebbar 2022-9-5

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1793075-access-matrix-outside-of-for-loop#answer_1042030

I understand that you want to access the 'Matrix_with_Number_of_colocalization' variable outside the 'parfor' loop.

When a 'parfor' loop is used, the variables are classified into different categories.

For more information on this, you can refer to the following documentation link:

https://www.mathworks.com/help/releases/R2021b/coder/ug/classification-of-variables-in-parfor-loops.html

2 个评论
显示无隐藏无

Alexandra Locherer 2022-9-5

Hey thank you for the answer. I just wanted to close the question. I save my data with a function in to different files. These I can access from outside the parfor loop.

Sushranth Prakash Hebbar 2022-9-5

Yes. Once, the data has been saved in different files, it should be accessible even outside the 'parfor' loop.

请先登录，再进行评论。

Access matrix outside of for loop

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

回答（1 个）

2 个评论
显示无隐藏无

另请参阅

类别

标签

产品

版本

Community Treasure Hunt

Access matrix outside of for loop

0 个评论 显示 -2更早的评论隐藏 -2更早的评论

回答（1 个）

2 个评论 显示 无隐藏 无

另请参阅

类别

标签

产品

版本

Community Treasure Hunt

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

2 个评论
显示无隐藏无