This directory contains algorithm for preprocessing:
* standardize: normalization of the columns of the data matrix (feature)  by subtracting the mean and dividing by the standard deviation.
* normalize: normalization of the lines of the data matrix by dividing by the Euclidean norm.
* shif_n_scale: global data matrix normalization by subtracting shift and dividing by scale.
* pc_extract: computation of new features based on principal component analysis.
* subsample: selection of a subset of the training patterns.