Allows to build Multi-Data-Driven Sparse Partial Least Squares models. Multi-blocks with
high-dimensional settings are particularly sensible to this. It comes with visualization
functions and uses 'Rcpp' functions for fast computations and 'doParallel' to parallelize cross-validation.
This is based on H Lorenzo, J Saracco, R Thiebaut (2019)
A sparse PLS formulation for mono and multi-block data sets with missing samples