Currently, we only check for this when the matrix is sparse - in the future, would be better to have this check regardless of how the eigendecomposition has been computed. Options
- use
networkx
- always comptue the top two eigenvalues -> problem with backwards iteration, need to modify/get rid
Currently, we only check for this when the matrix is sparse - in the future, would be better to have this check regardless of how the eigendecomposition has been computed. Options
networkx