Measure a customer-support summarizer.
Find quality differences by language, region, product, and input length.
Choose two appropriate practices.
Measure important slices and inspect sample size, uncertainty, and label quality so average performance does not hide harm.
Detailed explanation
Averages can hide subgroup failures.
Averages can hide subgroup failures.
Small or noisy slices need careful interpretation.
Small or noisy slices need careful interpretation.
Some users can still experience serious problems.
Some users can still experience serious problems.
The issue would be hidden.
The issue would be hidden.
Try it yourself
An example you can run in a temporary verification environment.
AWS公式AIF-C01 Domain 3.3・4.1のグループ別評価とバイアスを確認する。Expected result
全体平均に隠れる条件別の品質差を調査できる。Key points
- Slices
- Uncertainty
- Average limits
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
- Command output formatting can vary slightly by distribution or tool version.
- Run the example in a temporary directory or process when possible.
Foundation review
Read the scope first
Check whether the command acts on the current shell, a new process, an existing process, or a file.
Verify the observable result
Use the supplied command and compare the output with the expected result.