8000 Some estimator checks on meta-estimators are not invalid · Issue #20217 · scikit-learn/scikit-learn · GitHub
[go: up one dir, main page]

Skip to content
8000

Some estimator checks on meta-estimators are not invalid #20217

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

Open
thomasjpfan opened this issue Jun 7, 2021 · 1 comment
Open

Some estimator checks on meta-estimators are not invalid #20217

thomasjpfan opened this issue Jun 7, 2021 · 1 comment
Labels
module:test-suite everything related to our tests

Comments

@thomasjpfan
Copy link
Member
thomasjpfan commented Jun 7, 2021

Now that we are trying to move meta-estimators to delegate validation to inner estimators, I think that some estimator checks are invalid for meta-estimators. For example, check_complex_data on a meta-estimator that delegates validation, actually checks that the inner estimator is validating. Should we skip the input validation checks for meta-estimators?

CC @ogrisel

XREF #20205

@thomasjpfan thomasjpfan added the module:test-suite everything related to our tests label Jun 7, 2021
@glemaitre
Copy link
Member

Is not it tricky since that some meta-estimator still validate data: for instance if the meta-estimator needs to index X or y, it might require to do so).

It might be better to have a tag for this validate_input? As well, we saw an inconsistency in the usage of MetaEstimatorMixin. It would be great to use it everywhere or just remove it and use the tag instead.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
4375
Labels
module:test-suite everything related to our tests
Projects
None yet
Development

No branches or pull requests

2 participants
0