Under certain `min_dist`/ `spread` params UMAP fails to separate data
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 8.3k
- Forks
- 871
- Avg merge
- 1d 13h
- Merged PRs (30d)
- 5
Description
Thanks for UMAP. I was playing with the parameters and noticed an odd pattern that under certain parameters the UMAP 'collapses' and all the points become arranged on a single line, e.g.
Regular seperation, min_dist=1.1 and spread=2.1:
Collapsing behaviour min_dist=2.1 and spread=0.1 (also with min_dist=1.1):
This isn't really a practical problem for me since I can just select different parameters, but it seems odd, and I'm not sure if it a known curiosity or not.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No source files or tests are identified in the report. Start by reproducing UMAP with the stated min_dist and spread values, then trace the embedding behavior to determine whether the line collapse is expected or a bug; done means explaining or correcting that behavior with a regression test.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- data-visualization, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100