MATLAB is running out of memory but it should not be

by

For a data matrix of size n-by-p, PRINCOMP will return a coefficient matrix of size p-by-p where each column is a principal component expressed using the original dimensions, so in your case you will create an output matrix of size:

1036800*1036800*8 bytes ~ 7.8 TB

Consider using PRINCOMP(X,'econ') to return only the PCs with significant variance

Alternatively, consider performing PCA by SVD: in your case n<<p, and the covariance matrix is impossible to compute. Therefore, instead of decomposing the p-by-p matrix XX', it is sufficient to only decompose the smaller n-by-n matrix X'X. Refer to this paper for reference.

EDIT:

Here’s my implementation, the outputs of this function match those of PRINCOMP (the first three anyway):

function [PC,Y,varPC] = pca_by_svd(X)
    % PCA_BY_SVD
    %   X      data matrix of size n-by-p where n<<p
    %   PC     columns are first n principal components
    %   Y      data projected on those PCs
    %   varPC  variance along the PCs
    %

    X0 = bsxfun(@minus, X, mean(X,1));     % shift data to zero-mean
    [U,S,PC] = svd(X0,'econ');             % SVD decomposition
    Y = X0*PC;                             % project X on PC
    varPC = diag(S'*S)' / (size(X,1)-1);   % variance explained
end

I just tried it on my 4GB machine, and it ran just fine:

» x = rand(16,1036800);
» [PC, Y, varPC] = pca_by_svd(x);
» whos
  Name             Size                     Bytes  Class     Attributes

  PC         1036800x16                 132710400  double              
  Y               16x16                      2048  double              
  varPC            1x16                       128  double              
  x               16x1036800            132710400  double

Update:

The princomp function became deprecated in favor of pca introduced in R2012b, which includes many more options.

More Related Contents:

  1. Why is Matlab’s inv slow and inaccurate?
  2. Matlab – PCA analysis and reconstruction of multi dimensional data
  3. PCA in matlab selecting top n components
  4. Fast Algorithms for Finding Pairwise Euclidean Distance (Distance Matrix)
  5. Does Matlab eig always returns sorted values?
  6. BSXFUN on memory efficiency with relational operations
  7. Is MATLAB OOP slow or am I doing something wrong?
  8. How to elegantly ignore some return values of a MATLAB function
  9. How do you concatenate the rows of a matrix into a vector?
  10. How to execute multiple statements in a MATLAB anonymous function?
  11. How to load a large xlsx file with Apache POI?
  12. How to make the blackboard text appear clearer using MATLAB?
  13. Sort a matrix with another matrix
  14. Fortran minimization of a function with additional arguments
  15. Multiply a 3D matrix with a 2D matrix
  16. resizing 3D matrix (image) in MATLAB
  17. How to plot inequalities
  18. Call a function by an external application without opening a new instance of Matlab
  19. Is this a Matlab bug? Do you have the same issue? [duplicate]
  20. How to draw an arrow in Matlab?
  21. Matlab: implementing what CTRL+C does, but in the code
  22. How can I divide each row of a matrix by a fixed row?
  23. How can I load 100 files with similar names and/or string in just one step in MATLAB?
  24. Octave/Matlab: Adding new elements to a vector
  25. using precomputed kernels with libsvm
  26. Remove noise from wav file, MATLAB
  27. Convert output from symbolic math (sym) to float
  28. How to draw a line on an image in matlab?
  29. How to perform interpolation on a 2D array in MATLAB
  30. Automatically maximize a figure

Leave a Comment