Large performance impact with Map during Clustering

[Thad Guidry]

There's also a fairly sizeable performance impact with Map going on throughout Clustering operations during SimilarityDistance functions.

I did some preliminary research and it looks like the way we do or utilize Map in certain areas of the Clustering code parts could probably be enhanced or given a deeper look.  I noticed also Stefano's old notes in commits about the performance not being ideal yet, but never circled back over after this:

2010-03-09

getting closer to the desired functionality... still way too slow though

 

Thad

https://www.linkedin.com/in/thadguidry/