Finally, compare multiple datasets against each other

We are able to compare multiple datasets against each other by loading the
labeled trips, clustering them and comparing the number of trips and how they
are clustered against each other.

The NREL dataset has the most compact clusters, the mini-pilot has the least,
and the staging dataset has the most variability that spans both highly compat
and very uncompact clusters.