Re: Reading variable from workspace

"Elnaz " <ebsadeghian@gmail.com> wrote in message
news:ltqe1r$pub$1@newscl01ah.mathworks.com...
> This is exactly what I'm doing:
>
> % this loads 10 huge matrices, all of type double, same number of rows but
> different number of columns
> for i = 1:10
> load(['filename',num2str(i),'.mat']);
> end

LOAD into a structure or into an array of structures.

for i = 1:10
    D(i) = load(['filename',num2str(i),'.mat']);
end

> Then the computations:
>
> % generate 7 random streams as messages
> msg1 = (-1).^(rand(7,10000)>0.5);
>
> % multiply each stream by a corresponding huge matrix 'x', I know that
> this is error but I provide it just to show what I'm trying to achieve.
> for i = 1:n
> msg2(i,:) = msg(i,:)*['x',num2str(i)];

name = ['x',num2str(i)];
msg(i, :)*D(i).(name)

Or if this is actually the computation you want to perform:

msg2 = cell(1, numberOfFilesToProcess);
for whichone = 1:numberOfFilesToProcess
    D = load(filename{whichone});
    variablesLoadedIntoD = fieldnames(D);
    if isscalar(variablesLoadedIntoD)
        x = D(i).(variablesLoadedIntoD{1});
    else
        error('Each MAT-file should contain only one variable');
    end
    n = size(x, 1);

    % This avoids multiplying by a long vector.
    msg1 = randi([0 1], 1, n);
    msg2{whichone} = sum(x(msg1 == 0, :), 1) - sum(x(msg1 == 1, :), 1);
end

> end
>
> As you see I have the same problem with my msg2 matrix since the rows will
> be of different lengths. I can preallocate it with the maximum length but
> then inside my code I need to extract the nonzero portion each time I'm
> using msg2 (more hassle).

So don't. Process each file separately. In this case, the FOR loop above
would be in your outer function and the inner body of that loop would be
turned into a function that accepts a filename and returns its processed
vector that the outer function can store in the cell array.

If you do this, because each step is independent of the others, this could
be a candidate for parallelization using PARFOR.

http://www.mathworks.com/help/distcomp/parfor.html

--
Steve Lord
slord@mathworks.com
To contact Technical Support use the Contact Us link on
http://www.mathworks.com