The function d.spls.split divides the data X into a calibration and a validation set using
the Kennard and Stone strategy for each group at a time and according to the number of calibration desired
from each group.
d.spls.split(X, Xtype, Listecal = NULL)a numeric vector giving the row indices of the input data selected for calibration
a numeric matrix.
a vector of index specifying to which group belongs each observation.
a vector specifying how many observations from each group should be selected as calibration.
Louna Alsouki François Wahl
d.spls.type,d.spls.calval