d=dir(fullpath(dirname,'*.mat'));
for i=1:length(d)
load(d(i).name % will leave whatever variable is in mat-file in memory
[p,n]=fileparts(d(i).name); % get path, name w/o extension
csvwrite([fullfile(p,n) '.csv'],X) % write to name w/ .csv extension
end
The above use X as a placeholder for the variable containing the data LOADed; use whatever is that name. Of course, also assumes all the files are symmetric with the same name; if not use the functional form of load and the appropriate structure name retrieved.
And, of course, this begs the question of why one would do this... .mat files are much more accurate in keeping full precision whereas .csv files are a limited number of decimal places; are much faster to read/write and take up far less disk space besides (albeit that's pretty much a "don't care" item any more).
