Access variables with naming in workspace

I have a workspce full of data1, data2, data3.....,data100

They are all arrays with (1,n)

Instead of do it one by one via

```

matr(1,:)=data1

matr(2,:)=data2

etc

```

How can I use a loop to form this matrix

[Stephen23]

"I have a workspce full of data1, data2, data3.....,data100"

Having lots of numbered variables in the workspace is a sign that you are doing something wrong.

https://www.mathworks.com/matlabcentral/answers/304528-tutorial-why-variables-should-not-be-named-dynamically-eval

You did not tell us the most important information: how did you get all of these arrays into the workspace? The point where they are imported or created would be the correct place to fix your code, rather than what you are trying to do.

Whatever you do, do NOT use EVAL, ASSIGNIN, or follow any other subpar advice of that kind.

Answer 1

[Stephen23]

This appears to be a follow up from your question from seven hours ago:

https://www.mathworks.com/matlabcentral/answers/1804720-using-load-for-different-files-with-varying-names

The best solution by far is to not get into the situation where you have lots of numbered variables in the workspace:

https://www.mathworks.com/matlabcentral/answers/304528-tutorial-why-variables-should-not-be-named-dynamically-eval

For example, you can easily LOAD into an output variable, which makes your data very easy to access (unlike what you are doing now) It assumes exactly one variable per MAT file.

N = 25;
C = cell(1,N);
for k = 1:N
    F = sprintf('data%d.mat',k);
    C(k) = struct2cell(load(F));
end

All of your imported filedata wll be in the cell array C. Note that you can trivially loop over all of C, or access its contents individually. For example, the data for the second file:

C{2}

https://www.mathworks.com/help/matlab/matlab_prog/access-data-in-a-cell-array.html

Answer 2

[Chunru]

First of all, try to change the program that produces data1 to data100 by using multidimensional array or cell array if possible.

If that is not feasible for any reason, you can use eval:

matr = zeros(100, n)
for i=1:100
    matr(i,:) = eval("data"+i)
end

[Stephen23]

https://www.mathworks.com/matlabcentral/answers/304528-tutorial-why-variables-should-not-be-named-dynamically-eval

Reading this user's last question informs us that their bad data design is easy to avoid.

It would help both the OP and future readers of this forum, to avoid dynamically named variables.

[Rik]

I have not seen a situation 'in the wild' where eval could not be avoided. I don't see a reason why you would mention eval in this thread. You only risk teaching new users eval.

The only situation I have seen is when defining anonymous functions in an m file that Matlab 6.5 can also run.

[Chunru]

@Rik Imagine that you are using other people's program (which you may have p code only) or you really don't want to spend time to change that program (good or bad). You just want to quickly do your job.

[Chunru]

Every thing has a use under certain circumstances, "eval" and "assign" and "load" included, though you may want to avoid them in most times. This should be a good advice for life in general.

[Stephen23]

"...using other people's program (which you may have p code only) or you really don't want to spend time to change that program (good or bad)."

By reading the OPs previous questions we can see that (just like in 99.99% of situations involving attempts to access variable names) the OP simply painted themselves into a corner. As such, this OP's situation is easily avoided.

[Chunru]

In my post, I have advised "First of all, try to change the program that produces data1 to data100 by using multidimensional array or cell array if possible".

[Walter Roberson]

There is no point teaching someone to use eval until they have actually encountered a situation that could not avoid using eval.