Skip to content
Quantoros

A search evaluates both recall and precision. How is the final configuration selected for refitting?

Classic MLLevel: IntermediateMulti-metric model selection

A search evaluates both recall and precision. How is the final configuration selected for refitting?

Hints
  • Measurement and selection are separate.
  • The final ranking needs an explicit policy.

Answer

Specify a refit metric name or a supported callable selection rule; alternatively disable refitting. Reporting several metrics does not by itself define how to trade them off.

Source and licence: scikit-learn 1.7.2 — grid search · BSD-3-Clause · answer: Quantoros editorial

Related questions