You are discussing how you evaluate machine learning models after training and before release. The interviewer wants to understand how you think about metric selection, tradeoffs, and what different evaluation results imply in practice.
What is your experience with machine learning model evaluation techniques?