Linear classifier built on the KL expansion of the common covariance matrix
W = KLLDC(A,N)
Finds the linear discriminant function W for the dataset A. This is done by computing the LDC on the data projected on the first eigenvectors of the averaged covariance matrix of the classes. Either first N eigenvectors are used or the number of eigenvectors is determined such that ALF, the percentage of the total variance is explained. (Karhunen Loeve expansion)
If N (ALF) is NaN it is optimised by REGOPTC.