A classifier scores 99% on training data but 65% on unseen test data.
Investigate the difference before deployment.
Which description is most appropriate?
The gap is a sign of possible overfitting: the model fits training examples but generalizes poorly.
Detailed explanation
High training performance with much lower test performance indicates poor generalization.
High training performance with much lower test performance indicates poor generalization.
The gap alone does not prove whether leakage exists.
The gap alone does not prove whether leakage exists.
This destroys the independent evaluation.
This destroys the independent evaluation.
The cause and generalization must be investigated first.
The cause and generalization must be investigated first.
Try it yourself
An example you can run in a temporary verification environment.
AWS公式AIF-C01ガイドDomain 1.3の過学習・汎化・評価データを確認する。Expected result
訓練とテスト性能の差を過学習の兆候として説明できる。Key points
- Overfitting
- Generalization
- Test data
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.