Feb. 26, 2023
7:15 a.m.
On Sat, Feb 25, 2023 at 8:17 PM Louis Petingi <Louis.Petingi@csi.cuny.edu> wrote:
Hi Robert
Just a follow up. I was able (my student) to get the 1 vector from the 0 eigenvector. Even though the normalized or this set of eigenvectors will work we could try the two sets. Not sure if multiplying the unit vectors by a scalar is the correct approach.
Yeah, that essentially amounts to multiplying all of the eigenvectors by `np.sqrt(n)`. I don't imagine it will do much to change the results of K-means or other clustering algorithm unless if it has absolute distance thresholds embedded in it. But it's absolutely a free choice for you to precondition things for your clustering algorithm. -- Robert Kern