1.3Standard
Compare data processing, models, and hyperparameters.
Explain the selected candidate later.
Choose two practices.
× Incorrect
Record data, code, settings, seeds, environment, metrics, and time, and compare quality, safety, cost, latency, bias, and adoption rationale.
Detailed explanation
IncorrectChange evaluation data and settings for every comparison
Results are not comparable.
CorrectCompare quality, safety, cost, latency, and bias and record the adoption rationale
A single metric does not decide the model.
CorrectRecord data, code, settings, random seed, environment, metrics, and time
Experiments can be reproduced and compared.
IncorrectKeep only the highest score and delete other experiments
Failure analysis and comparison are lost.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 1.3の実験管理、MLOps、モデル選択を確認する。
Official references
What to confirm
モデル候補の比較条件と採用理由を再現可能な形で管理できる。
Key points
- Experiment tracking
- Reproducibility
- Rationale
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
Experiment tracking
Recording model, data, settings, environment, and results for each trial.
Hyperparameter
A setting chosen before or around training that controls how learning proceeds.
Question IDAWS-AIF-371
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