non-mean target encoding
Nobody has claimed this yet.
- Dominant language
- R
- Stars
- 146
- Forks
- 23
- PR merge metrics
- No merged PRs in 30d
Description
This would naturally cover quantile encoding. Maybe be included in the upcoming analytical version.
I have seen people use a lot of interesting things like min, max, nunique. https://www.kaggle.com/competitions/playground-series-s4e12/discussion/554328
another example: https://www.kaggle.com/code/cdeotte/first-place-single-model-lb-38-81#Feature-Engineer-Columns
this would be easy to do without shrinkage, but I haven't spend enough time thinking about how it would work in that scenario
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 names no files, tests, or entry points. Locate the existing encoding implementation, then clarify the analytical-version scope, supported aggregations such as min, max, and nunique, and how shrinkage should work; done should include agreed behavior covered by tests.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- r
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100