Graphical Lasso implementation in FORTRAN
Found an implementation: https://github.com/JClavel/glassoFast
Was wondering whether it is worth implementing it into pygeodyn, instead of using scikit-learn implementation. Would it leads to fastest calculations?