scikit-learn-contrib/category_encoders

Feature request: min_group_size for other category encoders, e.g. one hot

Open

#279 opened on Oct 30, 2020

View on GitHub
 (11 comments) (2 reactions) (1 assignee)Python (2,322 stars) (397 forks)batch import
enhancementgood first issue

Description

CountEncoder() has a min_group_size parameter that sets a minimum number of obs in a group required in order not for the group to be lumped together with other small groups. It would be nice to have this for the other encoder classes.

Contributor guide

Tech stack
python
Domain
datamachine learning
Issue type
feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
blocked
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
familiarity with category encodersunderstanding of CountEncoder.min group sizeknowledge of sklearn transformer APIPython
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
45
Research direction
The issue requests adding a min group size parameter to other encoders like OneHotEncoder. The implementer should study the CountEncoder's code in category encoders/count.py to understand the current implementation of min group size. Then, adapt the logic for each target encoder (e.g., OneHotEncoder in one hot.py) by adding a similar parameter and grouping logic. Existing tests in tests/ should be extended to cover the new parameter for each encoder. The discussion in the issue comments may provide additional context on desired behavior.