How to find the csv that lesser length of rows then 800 and append the same file chunk to fill the missing values.

2 次查看（过去 30 天）

shery khan 2022-7-1

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1751735-how-to-find-the-csv-that-lesser-length-of-rows-then-800-and-append-the-same-file-chunk-to-fill-the-m

编辑： shery khan 2022-7-2

Question No 1 is how can i find those files that have less number of rows or rows are not equal to 800? Question No 2 is after finding those file I want to fill those rows in chunks like if file contain 632 Rows. 32 rows are discarded and append 200 consecutive chunk (this chunk is copied from same file like 1 t0 200 ) at the end of file which make it 800. meaning we are replacing last 32 rows with first 200 rows of this file to complete 800 length of file

Before this i have completed these missing rows with mean code is attached

home_dir = 'D:\DS\Activity_Recognition';
 
sensors = dir(home_dir);
for sens =3:length(sensors)
    classes = dir(fullfile(home_dir, sensors(sens).name));
    for cls=3:length(classes)
        data = dir(fullfile(home_dir, sensors(sens).name, classes(cls).name));
        toWrite = cell(zeros(1,1));
        for dat = 3:length(data)
            
            tmp = load(fullfile(home_dir, sensors(sens).name, classes(cls).name, data(dat).name));
            
            tmp = tmp(:,2:end);
% % %      
    n_rows_target = 800; % extend to 800 rows, with mean of each column
    n_rows = size(tmp,1);
    if n_rows < n_rows_target
    
    mean_tmp = mean(tmp,1);
  tmp(n_rows+1:n_rows_target,:) = repmat(mean_tmp,n_rows_target-n_rows,1);
  
    end
%for 5 Sec
            for ch=1:4
             
               
              xTmp = tmp(((ch-1)*200)+1:ch*200, 1);
                yTmp = tmp(((ch-1)*200)+1:ch*200, 1);
               zTmp = tmp(((ch-1)*200)+1:ch*200, 1);
                xFeat = computeStatFeats(xTmp, length(xTmp)); 
                yFeat = computeStatFeats(yTmp, length(yTmp));
                zFeat = computeStatFeats(zTmp, length(zTmp));
                toWrite{dat-2, 1} = [xFeat yFeat zFeat];
                
            end
            fprintf('Computing %s Features: %d\n', classes(cls).name, length(data)-dat-2);
          
        end
        toWrite = cell2mat(toWrite);
        writematrix(toWrite, strcat(sensors(sens).name, '_', strrep(classes(cls).name, ' ', '_'), '.csv'));
    end
end