x = fn_round(x[,y[,'floor|ceil'])
eliminate floating point error rounds to the nearest multiple of y if a second argument is given
Thomas Deneux
Copyright 2009-2017
Published with MATLAB® R2015a