tidymodels / tidymodels/embed

non-mean target encoding

Open
#245 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

feature target encoding
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

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. 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

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.