Why can the original PCA object remain unfitted when a cached pipeline containing it has been fitted?
Classic MLLevel: IntermediateFitted clone inspection
Why can the original PCA object remain unfitted when a cached pipeline containing it has been fitted?
Hints
- Object identity can change.
- The fitted pipeline owns the fitted instance.
Answer
Caching clones intermediate transformers before fitting. Inspect the fitted instance through pipe.named_steps rather than the original Python object.
Source and licence: scikit-learn 1.7.2 — compose · BSD-3-Clause · answer: Quantoros editorial