-
Notifications
You must be signed in to change notification settings - Fork 875
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.
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
[WIP] Rewrite stacking estimators with class hierarchy #445
base: master
Are you sure you want to change the base?
Conversation
Hello @kota7! Thanks for submitting the PR.
|
Thanks for the PR! Just noticed a lot of failing unit tests. I think this may be because sklearn 0.20 has recently become available via miniconda, and mlxtend (also) tests the lastest sklearn version. There were some bugfixes to some estimators plus a slight change in the iris dataset (two data points changed) which explained some of the different results. Adjusted the unit tests yesterday and will rerun the CI tests here, then we can see if there are some issues remaining. |
Description
This is work-in-progress
Refactors existing stacking estimators with class inheritance structure. More specifically, classes are related as below. The real code is a bit more complex with multiple inheritance.
In addition, following arguments are renamed:
refit
-->use_clones
inStackingRegressor
andStackingCVRegressor
Following option is added
verbose
toStackingRegressor
Although there are still some tests fail and bug fix is going, I would like to have your thoughts as of now.
Fixes #444
Related issues or pull requests
Pull Request Checklist
./docs/sources/CHANGELOG.md
file (if applicable)./mlxtend/*/tests
directories (if applicable)mlxtend/docs/sources/
(if applicable)nosetests ./mlxtend -sv
and make sure that all unit tests pass (for small modifications, it might be sufficient to only run the specific test file, e.g.,nosetests ./mlxtend/classifier/tests/test_stacking_cv_classifier.py -sv
)flake8 ./mlxtend