rasbt/mlxtend

Add option for termination criterion with tolerance

Open

#1,079 建立於 2024年1月18日

在 GitHub 查看
 (1 留言) (2 反應) (0 負責人)Python (4,676 star) (839 fork)batch import
New Featureeasyhelp wanted

描述

Describe the workflow you want to enable

Add an option for a termination criterion similar to "tol" implemented in sklearn.feature_selection.SequentialFeatureSelector: https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.SequentialFeatureSelector.html

Describe alternatives you've considered, if relevant

While the sklearn SequentialFeatureSelector can do this, it misses the history callback you get in this package, which i think is very nice and helpful.

Additional context

The version of mlxtend i am using is 0.23.1

貢獻者指南

技術棧
pythonnumpypandas
領域
machine learningdata
議題類型
feature
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
3
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
1-2 days
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
fresh
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
mostly clear
前置要求
knowledge of mlxtend's SequentialFeatureSelectorfamiliarity with sklearn's tol parameterbasic Python
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
60
研究方向
Examine mlxtend's sequential feature selector.py file, specifically the SequentialFeatureSelector class. Implement a tolerance parameter that stops feature selection when the performance improvement is below the threshold, similar to sklearn's implementation (https://scikit learn.org/stable/modules/generated/sklearn.feature selection.SequentialFeatureSelector.html). Ensure the history callback continues to work with the new criterion. No existing PRs or comments provide additional guidance.