8000 Add MissingIndicator convenience into SimpleImputer · Issue #12294 · scikit-learn/scikit-learn · GitHub
[go: up one dir, main page]
Skip to content

Add MissingIndicator convenience into SimpleImputer #12294

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
Closed
amueller opened this issue Oct 4, 2018 · 9 comments
Closed

Add MissingIndicator convenience into SimpleImputer #12294

amueller opened this issue Oct 4, 2018 · 9 comments
Labels
help wanted

Comments

@amueller
Copy link
Member
amueller commented Oct 4, 2018

Following up on #8075 I think we should have a convenience option to add MissingIndicator to SimpleImputer without having to build a FeatureUnion or ColumnTransformer.
@jnothman
Copy link
Member
jnothman commented Oct 5, 2018 •
edited
Loading

Such as an append_indicators parameter?

@amueller
Copy link
Member Author
amueller commented Oct 5, 2018

@jnothman yes, that's what I was thinking. Do you agree?

@amueller amueller mentioned this issue Oct 5, 2018
Closed
@jnothman
Copy link
Member
jnothman commented Oct 6, 2018 via email

@amueller amueller mentioned this issue Oct 19, 2018
Closed
@thomasjpfan
Copy link
Member

To have full control over the missing indicators, features and error_on_new would need to be added to SimpleImputer. Would this create too much bloat to the SimpleImputer API?

@jnothman
Copy link
Member
jnothman commented Nov 5, 2018

To have full control over the missing indicators, features and error_on_new would need to be added to SimpleImputer. Would this create too much bloat to the SimpleImputer API?

We could consider allowing append_indicators=[feature1, feature2] but I think that's excessive for an initial implementation.

@adanhawth
Copy link
Contributor

Do you mind if I take this one?

@jeremiedbb
Copy link
Member

There's already #12583

@adanhawth
Copy link
Contributor

@amueller @jnothman: as @jeremiedbb pointed out, would this be a strict duplicate of #11886 and to be closed upon successful merge of #12583 or is there some other feature that this issue should speak to? Perhaps those enhancements mentioned by @thomasjpfan?

@jnothman
Copy link
Member

It is a duplicate.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
help wanted
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
5 participants
@jnothman @amueller @thomasjpfan @jeremiedbb @adanhawth
0