Fast Null Linear Discriminant Analysis

This is the fast implementation of Null LDA method. [W,CPU_TIME]=FNLDA(Data,ClassLabel)

Null linear discriminant analysis (LDA) method is a popular dimensionality reduction method for solving small sample size problem. The implementation of nullLDA method is, however, computationally very expensive. This implementation is the computationally efficient way of computing null LDA transform of a data matrix of size N x Dim where Dim >> N. Computational complexity of FNLDA method is around Dim*N^2 flops.