scikit-learn / scikit-learn/scikit-learn
Stratification in Nystroem kernel approximation
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 67.3k
- Forks
- 27.4k
- Avg merge
- 1d 15h
- Merged PRs (30d)
- 58
Description
Would it be possible to add stratification in Nystroem kernel approximation? The fit method already takes y as an argument but doesn't do anything with it. Perhaps instead it could be used so that if y is supplied, the sampling in X is guided such that the data points reflect an even number of samples per class.
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 the Nystroem estimator's fit method and trace how samples are selected from X when y is supplied. Define the expected behavior for balanced per-class sampling while preserving the existing behavior without y, then verify the result with the relevant Nystroem tests or by adding coverage for both cases.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100