MaartenGr / MaartenGr/BERTopic
Prob returned by fit_transform
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 7.8k
- Forks
- 920
- Avg merge
- 22h 24m
- Merged PRs (30d)
- 5
Description
I ran the fit_tranform() but the returned array prob did not add up to 1. For a data point assigned to a non -1 cluster, I assume the prob should add up to 1. I checked the source code and it seems the probability calculation is from hdbscan package. If the prob represents the chance of a data belonging to a cluster, the total probability should be 1. If not, can you explain what those values represent. In my case, a big portion of prob sum to a value < 0.5.
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
Start at BERTopic's fit_transform() path and inspect where the HDBSCAN probability values are obtained. Determine what the returned prob values represent and whether their behavior is intentional; document the explanation and any relevant usage guidance once confirmed.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100