rasbt / rasbt/mlxtend

Refactoring bootstrap_point632_score into iterator and scorer components

Open
#455 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

New Feature
Dominant language
Python
Stars
5.2k
Forks
916
Avg merge
17h 31m
Merged PRs (30d)
4

Description

unfortunately, the bootstrap_point632_score function can currently not be used in the ExhaustiveFeatureSelector etc.

The bootstrap_point632_score function is more like scikit-learn's cross_val_score which basically bundles a scorer with a cv method into one function, but here we need both separately.

So, in order to make this work with the ExhaustiveFeatureSelector interface, one would have to factorize the bootstrap_point632_score into an BootstrapIterator + a 0.632 scoring function (the 0.632 method scores and weights both the training and the out of bag samples), like cross_val_score can be refactored into a KFold/StratifiedKFold iterator + scoring function (which only scores the test folds).

Contributor guide

No contributing guide indexed for this repository

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 locating the implementation of bootstrap_point632_score and the ExhaustiveFeatureSelector interface. Compare the requested separation with scikit-learn's cross_val_score and its iterator/scorer pattern. Done means the bootstrap procedure is exposed as a BootstrapIterator, the 0.632 weighting is a separate scorer, and both work through the selector interface.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, scikit-learn
Domain
machine-learning
Issue type
Refactor
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.