How to save the data when we are using dir('*.hdf')

I have opened all files using MOD= dir('*.hdf'); but cannot able to use this to open these files or to extract some values form these files. Here is my script

MOD= dir('*.hdf');
for i=1:365
  Name=MOD(i).name 
  if i ==1
    AE(i)=hdfread(Name,'/mod08/Data Fields/Angstrom_Exponent_Land_Mean', 'Index', {[1  1],[1  1],[180  360]});
    AOD(i)=hdfread(Name, '/mod08/Data Fields/Optical_Depth_Land_And_Ocean_Mean', 'Index', {[1  1],[1  1],[180  360]});
  else
    AE(:,:,i)=hdfread(Name, '/mod08/Data Fields/Angstrom_Exponent_Land_Mean', 'Index', {[1  1],[1  1],[180  360]});
    AOD(:,:,i)=hdfread(Name, '/mod08/Data Fields/Optical_Depth_Land_And_Ocean_Mean', 'Index', {[1  1],[1  1],[180  360]});
  end
end

When I run this program, there is an error showing, Subscripted assignment dimension mismatch.

Error in test2 (line )

      AE(i)=hdfread(MOD(i).name
      ,'/mod08/Data
      Fields/Angstrom_Exponent_Land_Mean',
      'Index', {[1  1],[1  1],[180  360]});

Answer 1

[dpb]

While not your specific problem (at least yet), using the magic number of 365 for the loop isn't good practice; use

for i=1:length(MOD)  % iterate over returned file directory structure

Also, there's no need to create a second variable; just use hdfread(MOD(i).name, ... as you did in the code that generated the error instead of that as you posted; it has nothing to do with the error as you discovered.

AE(i)=hdfread(MOD(i).name, ...

tries to assign the entire output of the dataset read into a single element of a double array. That, needless to say, won't fit if there's more than a single value in the dataset. Use

AE=hdfread(MOD(i).name, ...

to return the dataset to the variable AE but then process that set of data before moving on to the next. If you must have more than a single set in memory at the same time, you'll have to do something to prevent overwriting subsequent values into the same array -- one way to do this would be to use a cell array as

AE{i}=hdfread(MOD(i).name, ...

where the "curlies" on the subscript create a cell array. You then would have to dereference those cell array references to get access to the data content itself.

Knowing nothing of the dataset nor what is intended to be done, we can't help in constructing an efficient way to process the data but the above will fix the immediate problem of reading the file(s).

[ANKUR KUMAR]

Thank you, Sir. My problem is resolved. But could you please tell me, why this command doesn't work.

% code
AE(i)=hdfread(MOD(i).name, ...

[dpb]

I did above -- the () parens are indexing into an array and refer to the one element of that array which is a single value. The RH side of the expression is all of the data from the HDF dataset which contains multiple values of whatever size that dataset happens to be. It's an attempt of a "many to one" assignment that just can't be--just as you can't fit two eggs into a single compartment in an egg carton, you can't store multiple numbers into a single memory location.

OTOH, a cell array is a series of egg cartons of flexible size -- each carton can be of whatever size is needed for the number of eggs the hens laid that day. See

doc cell

for basic syntax and the section on cell arrays in the Data Types section for background.