How to run the same script but for different datafiles consecutively

Question

sittmo 2013-8-29

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/85929-how-to-run-the-same-script-but-for-different-datafiles-consecutively

So I have the following script:

clc;
clear;
LM = xlsread('LagrangeMultiplier.xlsx','LM');
CM = xlsread('LagrangeMultiplier.xlsx','CM');
DT = xlsread('LagrangeMultiplier.xlsx','DT');
for i = 1:size(LM,2)
    JPD(i,:) = exp(-1-sum(LM(:,i)))*mvncdf(DT, Inf*ones(size(DT,1),1),0,CM);
end

Basically I have datafiles named:

LagrangeMultiplier1.xlsx
LagrangeMultiplier2.xlsx
LagrangeMultiplier3.xlsx

and so on until LagrangeMultiplier100.xlsx

Currently I'm running this script manually for each file, ie, I have LagrangeMultiplier1.xlsx as the name in the script, then I copy the values from the JPD workspace into excel, then put LagrangeMultiplier2.xlsx as the datafile name and so on. How can I automate this so that MATLAB automatically runs the script for all 100 datafiles and stores all 100 arrays for JPD into the workspace?

Thanks

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

Geert 2013-8-30

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/85929-how-to-run-the-same-script-but-for-different-datafiles-consecutively#answer_95515

编辑：Geert 2013-8-30

在 MATLAB Online 中打开

@ Azzi: It is indeed not recommended to create severable variables by the

eval([ 'JPD',num2str(ii),' = JPD;'])

command. I thought Simon wanted to do it anyway, so hence I delivered ;)

@ Simon: From your screenshot it looks like your data is one-dimensional. You can hence change the code as follows to obtain what you want:

nb_files = 100;
% preallocate space for JPD
LM = xlsread('LagrangeMultiplier1.xlsx','LM');
JPD = zeros(length(LM),nb_files);
% read in all the files and store them as columns in JPD variable
for kk=1:nb_files
  file=sprintf('LagrangeMultiplier%d.xlsx',kk);
  LM = xlsread(file,'LM');
  CM = xlsread(file,'CM');
  DT = xlsread(file,'DT');
  for ii = 1:length(LM)
      JPD(ii,kk) = exp(-1-sum(LM(:,ii)))*mvncdf(DT, Inf*ones(size(DT,1),1),0,CM);
  end
end

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

sittmo 2013-8-30

Perfect, just what I was looking for, thanks Geert!

请先登录，再进行评论。

Answer 2

Azzi Abdelmalek 2013-8-29

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/85929-how-to-run-the-same-script-but-for-different-datafiles-consecutively#answer_95450

在 MATLAB Online 中打开

for k=1:100
  file=sprintf('LagrangeMultiplier%d.xlsx',k);
  LM = xlsread(file,'LM');
  CM = xlsread(file,'CM');
  DT = xlsread(file,'DT');
  for i = 1:size(LM,2)
      JPD(i,:) = exp(-1-sum(LM(:,i)))*mvncdf(DT, Inf*ones(size(DT,1),1),0,CM);
  end
  out{k}=JPD
end

3 个评论
显示 1更早的评论隐藏 1更早的评论

Ilham Hardy 2013-8-30

在 MATLAB Online 中打开

change

out{k}=JPD

into

JPDall{k}=JPD

to accomodates your need..

sittmo 2013-8-30

编辑：sittmo 2013-8-30

Thanks, however it still doesn't do what I intended, the result I get is shown in this screenshot: http://imageshack.us/photo/my-images/832/umzs.jpg/

However as you can see, I need to click on JPDall{1,1} to see the 1348x1 column of values. Since I want to copy each cell of values into excel, I would need to open each of the 9 JPDall cells individually and copy each into excel, that would take quite a long time when I increase it to 100 cells. Is there a way to output all JPDall cells into one variable, so that I have 1348x9 matrix of values so I can copy it all at once into excel.

请先登录，再进行评论。

Answer 3

Geert 2013-8-29

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/85929-how-to-run-the-same-script-but-for-different-datafiles-consecutively#answer_95453

编辑：Geert 2013-8-29

在 MATLAB Online 中打开

The following code will save your JPD variable for LagrangeMultiplier1.xlsx as JPD1 in the workspace, the JPD variable for LagrangeMultiplier2.xlsx as JPD2, and so on...

clc;
clear;
nb_files = 100;
for ii = 1:nb_files
    filename = ['LagrangeMultiplier',num2str(ii),'.xlsx'];
    LM = xlsread(filename,'LM');
    CM = xlsread(filename,'CM');
    DT = xlsread(filename,'DT');
      JPD = [];
      for jj = 1:size(LM,2)
          JPD(jj,:) = exp(-1-sum(LM(:,jj)))*mvncdf(DT, Inf*ones(size(DT,1),1),0,CM);
      end
      eval([ 'JPD',num2str(ii),' = JPD;'])
  end

4 个评论
显示 2更早的评论隐藏 2更早的评论

Ilham Hardy 2013-8-30

I think what he meant is just like Azzi's answer.. by putting it in one cell.

sittmo 2013-8-30

What I mean is, have JPD1, JPD2... all within one variable called JPDall or something, because currently I have to open JPD1 to see the values in JPD1, then open JPD2 to see the values in JPD2... what I was hoping is to have all those values under one variable JPDall so that when I open JPDall I see the values for all JPD1, JPD2 ...

请先登录，再进行评论。