1.3Standard
Search over tree depth and regularization.
Do not overfit the final test set.
Choose two appropriate practices.
× Incorrect
Use training and validation data for candidate selection, hold out the test set, and record the search space, seed, metric, and stopping rule.
Detailed explanation
IncorrectTune until the test score is best
The test set becomes part of development.
CorrectRecord search space, random seed, metric, and stopping condition
The search can be reproduced and controlled.
CorrectCompare candidates on training and validation data and reserve the test set
Model selection remains separate from final evaluation.
IncorrectTreat the search score as a production guarantee
Deployment conditions still need validation.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 1.3のモデル選択、検証、再現性を確認する。
Official references
What to confirm
探索と最終テストの役割を分け、再現可能な条件を記録できる。
Key points
- Validation
- Test
- Reproducibility
Notes
- Environment: AWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
- Compare the definition with the AIF-C01 objective and the service documentation before applying it.
- Keep the concept separate from a particular answer choice or implementation step.
Foundation review
Hyperparameter
A training or model-structure setting chosen outside the learned parameters.
Search space
The defined ranges and choices explored during model selection.
Question IDAWS-AIF-174
EnvironmentAWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
Last technical review2026-08-20
Report an error or rights issue