tslearn-team / tslearn-team/tslearn
cluster number reduced over iteration
Open
Nobody has claimed this yet.
KShape
- Dominant language
- Python
- Stars
- 3.2k
- Forks
- 384
- Avg merge
- 3d 12h
- Merged PRs (30d)
- 11
Description
- As shown in the screenshot below, it reports like this when I turn on verbose. Instead of keep decreasing as in the example, the inertia increased. Does this mean the algorithm didn’t converge?
- Sometimes the final cluster number is less than the number I assigned (as in the screenshot below), which usually happens when I assign K as larger than 5 (I guess some similar clusters merged). Does this theoretically make sense? I feel this is caused by the initialization, is there any suggestion on defining the random state (I also found the result varying even if I fix the seeds)?
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
The issue provides no source file or test to start from. Reproduce the verbose clustering cases described in the screenshots with K above 5 and fixed seeds; done means determining whether the increasing inertia and reduced final cluster count are expected or represent a defect.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100