Introduce a customer-churn prediction model into production.
Organize data preparation, training, evaluation, deployment, and operations.
Which description of a typical machine learning lifecycle is appropriate?
The lifecycle covers defining the goal, preparing data, training, evaluating, deploying, and monitoring iteratively.
Detailed explanation
This causes leakage and invalidates the evaluation.
Goals and metrics should be defined before development.
Model development is iterative and includes evaluation and operational monitoring.
Data and business conditions can change after deployment.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01ガイドのDomain 1.3で、AI/ML開発ライフサイクルの各段階を確認する。
Official references
What to confirm
開発前の目的定義から本番後の監視までを一連のライフサイクルとして説明できる。
Key points
- Data preparation
- Separating evaluation data
- Production monitoring
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
Independent evidence
Evaluation data must remain separate from training and tuning so the result is credible.
Post-deployment monitoring
Production inputs and eventual outcomes reveal changes that a pre-release test cannot show.