3d pictures seem to cluster into 2 groups

[micah541]

Just an observation, really. I tried a number of examples. Perhaps the process should be somewhat iterative? How to check if we're finished?

[micah541]

I added a scalar curvature function. I'm hoping this gives us some useful information. My hope is that when a metric space is already well clustered (say sampling from a single normal distribution) then the scalar curvature should be roughly a positive constant. If the metric space still needs clustering, then one would expect to find a few points with negative scalar curvature.

The test does seem to give some useful information, in some cases.