SDNMEAN Nearest mean classifier P=SDNMEAN(DATA) INPUT DATA labeled data set OUTPUT P Nearest mean pipeline (output is a similarity) OPTIONS 'priors' class priors (default: use priors from the training set) 'scaled' use scaled diagonal covariance (classifier is not sensitive to scaling) DESCRIPTION SDNMEAN is a nearest mean classifier implemented as a normal model with identity covariance matrix. SDNMEAN is therefore sensitive to scaling. Alternatively, with 'scaled' option, the scaled diagonal covariance is used making SDNMEAN scaling-insensitive. READ MORE http://perclass.com/doc/guide/classifiers.html#sdnmean