SPArse Modeling Software
now open-source!

Downloads

The sources of SPAMS v2.1 are now released under GPLv3 licence. We still provide binaries for Linux 64 bits, with linking to the Intel MKL library. For other plateforms, MAC OS, Windows, you need to compile it yourself. Some compilation scripts can be found in the contribution section.
The binary package was compiled using the Intel compiler version 11. On recent computers, a better performance can be obtained by compiling yourself the library with the latest Intel compiler.

Current version: 2.1

Tar-gzipped, sources,
Tar-gzipped, Linux 64bits, Intel Compiler, MKL,

+ Bug corrected in mexLasso, when reaching the end of the regularization path. Numerical Stability has been improved.

The version using Intel MKL are recommended for a better performance. There is no need of installing the MKL library, everything needed is included in the package. Mac and Windows versions have not been tested intensively. Please feel free to report any problem you encounter. If you want to report speed results obtained with this toolbox, please use the Linux 64bits version.

Current version: 2.0

+ Proximal Toolbox

Older version: 1.02

Tar-gzipped, Linux 64bits, Intel Compiler, MKL,
Tar-gzipped, Linux 64bits, GNU Compiler, ATLAS,
Tar-gzipped, Linux 32bits, Intel Compiler, MKL,
Tar-gzipped, Linux 32bits, GNU Compiler, ATLAS,
Tar-gzipped, Mac OS 32bits, Intel Compiler, MKL,
Tar-gzipped, Mac OS 64bits, Intel Compiler, MKL,
Tar-gzipped, Windows 32bits, Intel Compiler, MKL,

+ Elastic-Net, a few heuristics are added.

Warning: the file test_TrainDL in the win32 version was incorrect. This has been corrected.

Warning: It seems that Matlab >= R2009b is required for the win32 version.

Older version: 1.01

Tar-gzipped, Linux 64bits, Intel Compiler, MKL,
Tar-gzipped, Linux 64bits, GNU Compiler, ATLAS,
Tar-gzipped, Linux 32bits, Intel Compiler, MKL,
Tar-gzipped, Linux 32bits, GNU Compiler, ATLAS,
Tar-gzipped, Mac OS 32bits, Intel Compiler, MKL,
Tar-gzipped, Mac OS 64bits, Intel Compiler, MKL,

+ functions are robust to bad input parameters.

Older version: 1.00

Last modification: 2011-07-03 19:23:37.000000000 +0200