Why is bestscore not an unbiased final estimate after trying many hyperparameter configurations?
Classic MLLevel: AdvancedSearch-induced optimism
Why is best_score_ not an unbiased final estimate after trying many hyperparameter configurations?
Hints
- Selection consumes validation information.
- Evaluate the procedure, not only its winner.
Answer
This answer is part of mentoring
Advanced questions come with a full breakdown and a mock interview with a working quant researcher. The first call is free.
Source and licence: scikit-learn 1.7.2 — grid search · BSD-3-Clause · answer: Quantoros editorial