A churn model uses age, monthly fee, and usage count.
Handle missing values and apply the same transformations during training and inference.
Choose two appropriate practices.
Assess missing and anomalous values and use a reproducible preprocessing pipeline consistently during training and inference.
Detailed explanation
This creates training-serving skew.
Inconsistent preprocessing changes the meaning of model inputs.
Choose imputation, exclusion, or another policy based on the data meaning and impact.
Zero may not represent a missing value and can introduce bias.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01ガイドDomain 1.3のデータ前処理・特徴量を確認する。
Official references
What to confirm
欠損処理と訓練・推論の処理一致の重要性を説明できる。
Key points
- Missing values
- Reproducibility
- Training-inference consistency
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
Training-serving skew
Different preprocessing or units between training and production can reduce model quality.
Reproducibility
A recorded and repeatable pipeline makes data treatment observable and testable.