Sparse Coding Toolbox——Open-Source！

SPArse Modeling Software

About

What is SPAMS?

SPAMS (SPArse Modeling Software) is an optimization toolbox for solving various sparse estimation problems.

• Dictionary learning and matrix factorization (NMF, sparse PCA, ...)
• Solving sparse decomposition problems with LARS, coordinate descent, OMP, SOMP, proximal methods
• Solving structured sparse decomposition problems (l1/l2, l1/linf, sparse group lasso, tree-structured regularization, structured sparsity with overlapping groups,...).

See the documentation for all the features. 
It is developped by Julien Mairal (INRIA), with the collaboration of Francis Bach (INRIA), Jean Ponce (Ecole Normale Supérieure), Guillermo Sapiro (University of Minnesota), Rodolphe Jenatton (INRIA) and Guillaume Obozinski (INRIA). It is coded in C++ with a Matlab interface. Recently, interfaces for R and Python have been developed by Jean-Paul Chieze(INRIA).

License

Version 2.1 and later are open-source under licence GPLv3. 
It is therefore possible to redistribute the sources with any other software, as long as it under GPL licence. 
For other usage, please contact the authors.

Related publications

You can find here some publications at the origin of this software. 
The "matrix factorization" and "sparse decomposition" modules were developed for the following papers:

J. Mairal, F. Bach, J. Ponce and G. Sapiro. Online Learning for Matrix Factorization and Sparse Coding. Journal of Machine Learning Research, volume 11, pages 19-60. 2010.

J. Mairal, F. Bach, J. Ponce and G. Sapiro. Online Dictionary Learning for Sparse Coding. International Conference on Machine Learning, Montreal, Canada, 2009

The "proximal" module was developed for the following papers:

J. Mairal, R. Jenatton, G. Obozinski and F. Bach. Network Flow Algorithms for Structured Sparsity. Adv. Neural Information Processing Systems (NIPS). 2010.

R. Jenatton, J. Mairal, G. Obozinski and F. Bach. Proximal Methods for Sparse Hierarchical Dictionary Learning. International Conference on Machine Learning. 2010.

The feature selection tools for graphs were developed for"

J. Mairal and B. Yu. Supervised Feature Selection in Graphs with Path Coding Penalties and Network Flows. arXiv:1204.4539v1. 2012.

News

05/23/2012: SPAMS v2.3 is released. 
03/24/2012: SPAMS v2.2 is released with a Python and R interface, and new compilation scripts for a better Windows/Mac OS compatibility. 
06/30/2011: SPAMS v2.1 goes open-source! 
11/04/2010: SPAMS v2.0 is out for Linux and Mac OS! 
02/23/2010: Windows 32 bits version available! Elastic-Net is implemented. 
10/26/2009: Mac OS, 64 bits version available!