Thanks to visit codestin.com
Credit goes to github.com

Skip to content

set force_all_finite=False for SelectFromModel.transform #11026

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

Conversation

alexpantyukhin
Copy link

@alexpantyukhin alexpantyukhin commented Apr 24, 2018

Fixes #10985

Copy link
Member

@jnothman jnothman left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

please add a test.

Also, modify the PR description text to say something like Fixes #xxxx. This way GitHub recognises the relationship between the PR and the issue it fixes.

@alexpantyukhin
Copy link
Author

@jnothman Thank you! I will do.

@jnothman jnothman changed the title set force_all_finite=True for SelectFromModel.transform set force_all_finite=False for SelectFromModel.transform Apr 25, 2018
@alexpantyukhin alexpantyukhin force-pushed the force_all_finite_false_for_SelectFromModel branch 6 times, most recently from ce85344 to cde9b83 Compare April 25, 2018 20:27
@alexpantyukhin alexpantyukhin force-pushed the force_all_finite_false_for_SelectFromModel branch from cde9b83 to 2642357 Compare April 25, 2018 20:48
Copy link
Member

@jnothman jnothman left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think we should allow NaN in fitting too. Make est = make_pipeline(Imputer(), ...Classifier()) in the test

@alexpantyukhin
Copy link
Author

@jnothman Do you think it should be in the same test?

@jnothman
Copy link
Member

Yes

@alexpantyukhin
Copy link
Author

I found that Imputer is deprecated and there is no Classifier. Also there are many inheritance from the BaseEstimator. Sorry, I'm a little bewildered what estimators should be listed in the make_pipline(...)

@jnothman
Copy link
Member

jnothman commented Apr 26, 2018 via email

@alexpantyukhin alexpantyukhin force-pushed the force_all_finite_false_for_SelectFromModel branch 3 times, most recently from da02c6b to d61c94e Compare April 27, 2018 20:50
@alexpantyukhin
Copy link
Author

alexpantyukhin commented Apr 27, 2018

@alexpantyukhin alexpantyukhin force-pushed the force_all_finite_false_for_SelectFromModel branch from d61c94e to e209dfa Compare April 28, 2018 06:32
@alexpantyukhin alexpantyukhin force-pushed the force_all_finite_false_for_SelectFromModel branch from e209dfa to f628d5e Compare April 28, 2018 10:01
@amueller
Copy link
Member

yes, that's why you'd pipeline it with SimpleImputer. But I'm not sure that'll actually work, because there is no coef_ or feature_importances_ on the pipeline. Which raises the question on whether there should be ....

@jnothman
Copy link
Member

Tests are failing @alexpantyukhin. Let us know if you need help working out why.

Base automatically changed from master to main January 22, 2021 10:50
@thomasjpfan
Copy link
Member

Thanks for working on this issue. Closing since original issue was resolved by #11635

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Is there any reason for SelectFromModel.transform to use force_all_finite=True in check_array?
4 participants