AUTHORS: A. d’Aspremont, L. El Ghaoui, M. I. Jordan, G. R. G. Lanckriet.
CONTENT: This package contains the numerical code used in the paper. The M-files, source code and MEX binaries for Linux, Mac OS X and Windows are available below.
Version 0.6: (December 2008) Freshly updated MATLAB/MEX binaries for Win32 and MATLAB R2008a, MATLAB has once again changed its BLAS/LAPACK distribution... PathSPCA, another much simpler code for computing a full path of approximate solutions based on the results in the JMLR paper "Optimal Solutions for Sparse Principal Component Analysis." by A. d’Aspremont, F. Bach and L. El Ghaoui is available here. This code is available in pure MATLAB and Python, produces slighty less accurate solutions but scales much better. By default, try this one first.
Version 0.5: (May 2008) Updated MATLAB/MEX binaries for intel Macs, Win32, and LINUX. Directly calls ARPACK, much faster but harder to compile on Win32. Salman Siddiqui has compiled a version for 64-bit machines which can be downloaded here.
Version 0.4: (July 2007) Updated MATLAB/MEX binaries for intel Macs.
Version 0.3: (July 2006) Uses partial eigenvalue
decomposition (with ARPACK) and Padé approximations to
compute the matrix exponential. Partial eigenvalue
decomposition is significantly faster (2x to 10x) on most
problems. Works with MATLAB 7.3. The code is compatible with
previous versions of MATLAB but might need to be recompiled
because of changes in the way MATLAB handles MEX files.
Version 0.2: (Nov. 2005) Works with Sedumi 1.1 and
MATLAB 7.1, unified sources for win32 and Mac.