Fix sklearn 1.7+ compatibility: BaseEstimator type detection for ensemble #1512
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
sklearn 1.7+ changed estimator type detection from checking
_estimator_typeattribute to usingget_tags(). SinceBaseEstimatorinherits fromClassifierMixin, FLAML regression estimators were incorrectly tagged as classifiers, causingStackingRegressorto reject them during ensemble construction.Changes
__sklearn_tags__()inBaseEstimatorto dynamically return correct tags based on_estimator_typeinstance attribute set during initializationis_regressor()/is_classifier()checksExample
The fix is backward compatible with sklearn < 1.7 where the old
_estimator_typeattribute mechanism still works.Original prompt
💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.