Hacker News new | ask | show | jobs
by pmaze 642 days ago
I did, there was a first round of UMAP to 50 dimensions. Running HDBSCAN on the full embeddings gave bad results, lots of singleton clusters.
1 comments
maCDzP 634 days ago
Interesting, I got the opposite result. The full embeddings gave two or three clusters. How did you work with the hyper parameters of HDBSCAN?
link