Linear Discriminant Analysis for the Small Sample Size Problem as described in
[CC] = nan_train_lda_sparse(X,G,par,tol)
(sparse) training data Xmat
group coding Xmat of the training data
(sparse) test data Xmat
group coding Xmat of the test data
if par = 0 then classification exploits sparsity too
tolerance to distinguish zero eigenvalues
Wrong classification rate (in %)
LDA transformation vectors
Algorithm 1 of J. Duintjer Tebbens, P. Schlesinger: 'Improving Implementation of Linear Discriminant Analysis for the High Dimension/Small Sample Size Problem', Computational Statistics and Data Analysis, vol. 52, no. 1, pp. 423-437, 2007.
J. Duintjer Tebbens, P. Schlesinger: 'Improving
Implementation of Linear Discriminant Analysis for the High Dimension/Small Sample Size
Problem', Computational Statistics and Data Analysis, vol. 52, no. 1,
pp. 423-437, 2007.