(aside image)

DCA: discriminative component analysis

The DCA software package implements the discriminative component analysis method described in the paper Discriminative Components of Data (with the difference that in this package, conjugate gradient is used for optimization instead of stochastic gradient). The algorithm has been developed by the Statistical Machine Learning and Bioinformatics group at the Department of Information and Computer Science, Aalto University. When you use the code, please cite the following paper:

Jaakko Peltonen and Samuel Kaski. Discriminative Components of Data. IEEE Transactions on Neural Networks, 16:68-83, 2005.

The software package is written in Python and C. The software package is licensed under the GNU Lesser General Public License version 2.1; see the file LICENSE included in the package for the full license terms. The package has been developed under GNU/Linux; contact Jaakko Peltonen if you have any trouble installing it on another platform.

The software is still in beta: everything should work, but there may still be some rough edges, particularly in the documentation. This is experimental software provided as is; we welcome any comments and corrections but cannot give any guarantees about the code.

Documentation

See the readme.txt file included in the package.

Getting and installing the software

Download the ZIP archive [dca.zip], and extract it into an empty directory of your choosing. You then need to compile a library needed by the package: in a UNIX-like environment this can be done with the provided Makefile. You may need to edit the Makefile to make it correspond to your local Python installation. After you have edited the Makefile, typing 'make' at the command line in the directory where you extracted the files should compile everything. See the included readme.txt file for instructions on using the software.

Support

If you have any questions, suggestions, or bug reports, please direct them to Jaakko Peltonen.