function weightsLDA = getLDAweights(matrixData,clabels) % Feature extraction based on LDA % % Inputs: % % matrixData - d*t*n tensor containing n matrices of size d*t, where n is the number of samples. % clabels - n-dimensional vector of binary coded category labels % % Ouput: % % weightsLDA - d*t matrix of t-dimensional directions estimated with LDA % % % Jukka-Pekka Kauppi % jukka-pekka.kauppi@helsinki.fi % University of Helsinki, Department of Computer Science % 23.8.2012 siz = size(matrixData); d = siz(1); t = siz(2); n = siz(3); weightsLDA = zeros(d,t); for m = 1:d X = squeeze(matrixData(m,:,:))'; % n*t matrix weightsLDA(m,:) = computeLDA(X,clabels); end