MATLAB Answers

0

How to convert arrays out of cells

Oliver Makan さんによって質問されました 2019 年 11 月 14 日
最新アクティビティ Adam Danz
さんによって 編集されました 2019 年 11 月 15 日 13:31
Hi Guys,
I wanted to convert the fields of the structure into variables to correlate them with each other.
Manually I just click on them an transfer them into workspace as a cell array. Like you see at the bottom I get a code like this with incremental values.
Is there a possibilitiy to do this with an for loop or something like this that it generates my cells automatically out of the structure.
Thank you for your help.
000003.jpg

  0 件のコメント

サインイン to comment.

1 件の回答

Adam Danz
回答者: Adam Danz
2019 年 11 月 14 日
編集済み: Adam Danz
2019 年 11 月 15 日 13:31

Generating variables within a loop involves dynamic variable naming which should be avoided at all cost. Instead, since your arrays appear to be all the same size, it would be much better to pack them into a matrix or a cell array. A matrix would make it very easy to compute correlations between variables. For example, corrcoef(A) computes correlation coefficients for each column of A. Here's a demo using a structure that is similar to yours.
% Create structure
SAM.Inn(1).Ri = rand(1,100);
SAM.Inn(1).Diff = rand(1,100);
SAM.Inn(2).Ri = rand(1,100);
SAM.Inn(2).Diff = rand(1,100);
SAM.Inn(3).Ri = rand(1,100);
SAM.Inn(3).Diff = rand(1,100);
SAM.Inn(4).Ri = rand(1,100);
SAM.Inn(4).Diff = rand(1,100);
% Put all "Ri" into a matrix where each column is a field from SAM.Inn
Ri = cell2mat({SAM.Inn.Ri}.'); % Or, see Stephen's comment below.
% Compute correlation coefficients
cc = corrcoef(Ri);
If you'd rather work with cell arrays or if your data do not have the same number of observations,
Ri = {SAM.Inn.Ri};

  1 件のコメント

Stephen Cobeldick
2019 年 11 月 15 日
Without the intermediate cell array and slow cell2mat:
Ri = vertcat(SAM.Inn.Ri).';

サインイン to comment.



Translated by