tidymodels / tidymodels/embed

User defined cuts for tree-based methods

Open
#202 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

documentation feature
Dominant language
R
Stars
146
Forks
23
PR merge metrics
No merged PRs in 30d

Description

Feature

It isn't currently clear how step_collapse_cart and other methods for numeric predictors determine how many groups to select. It could be useful to add a parameter cut_depth that would allow users to specify how far down the decision tree to make cuts to collapse/group values. Alternatively, a groups parameter that allows users to specify the number of groups they want to retain, which would return <= groups based on cutting the tree could also be a useful feature for users.

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

Start by reading step_collapse_cart and the other numeric-predictor methods to understand how tree cuts currently determine groups. Clarify whether the API should expose cut_depth, groups, or both, then define and test the expected behavior for retaining at most the requested number of groups.

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
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.