This software package is for solving the distance weighted discrimination problem of the following form:
where is the sample size, is the feature dimension, and is the penalty parameter. Here, the input is a matrix whose columns are the feature vectors, and is a vector containing binary classification label . The notation is defined to be . The exponent is typically set to or . The main algorithm for solving is a symmetric Gauss-Seidel based ADMM method. It will output the result for which is a hyperplane separating the two classes of data and is a slack variable to allow the possibility that the positive and negative data points may not be separated cleanly by the hyperplane.
For more details, please see the 'README' file in the package.
Lam, X.Y., Marron, J.S., Sun, D.F., and Toh, K.C. (2018), Fast algorithms for large scale generalized distance weighted discrimination, Journal of Computational and Graphical Statistics, forthcoming. https://arxiv.org/abs/1604.05473
Copyright: This version of DWDLarge is distributed under the GNU General Public License 2.0. For commercial applications that may be incompatible with this license, please contact the authors to discuss alternatives.
You can download the MATLAB package here: DWDLarge.zip
You can download the R package here: https://cran.r-project.org/web/packages/DWDLargeR/index.html
User's guide is included in the package. Below are some basic setup guide:
Unpack the software
The software requires a few mex-files that you may need to compile in MATLAB. To do so, run MATLAB in the directory "DWDLarge", then type:
After that, to see whether you have installed the software correctly, type the following in MATLAB:
By now, the DWDLarge package should be ready for use.