Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Refactor stacking estimators with class inheritance #444

Open
kota7 opened this issue Sep 24, 2018 · 1 comment · May be fixed by #445
Open

Refactor stacking estimators with class inheritance #444

kota7 opened this issue Sep 24, 2018 · 1 comment · May be fixed by #445
Labels
Enhancement

Comments

@kota7
Copy link
Contributor

kota7 commented Sep 24, 2018 •
edited

Currently we have four stacking-based estimators: StackingRegressor and StackingClassifier, and CV variant of each. They are independently coded, while core steps are common.

This makes enhancement work time-consuming, since contributors often need to make similar update on four source code. I would suggest refactoring the four estimators with appropriate class inheritance structure. This will make issues like #439 and #255 much simpler to implement.
@rasbt
Copy link
Owner

rasbt commented Sep 24, 2018

Totally agree. They were implemented one at a time, over time, and the API changed quite a bit based on what's useful and what not. Now after 1-2 years that it's kind of stabilized and they remain pretty similar "core"-wise, it'd be good to consolidate them via a parent Stacking class

@kota7 kota7 linked a pull request Sep 25, 2018 that will close this issue
Open
5 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
Enhancement
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

[WIP] Rewrite stacking estimators with class hierarchy kota7/mlxtend
2 participants
@rasbt @kota7