An ML engineer needs to predict apartment prices for a particular location. Which evaluation metric is appropriate for measuring the model’s predictive performance?

Mean absolute error (MAE) is the appropriate metric because predicting apartment prices is a regression task, where the goal is to predict a continuous numerical value. MAE measures the average magnitude of the errors in a set of predictions, without considering their direction. It is robust to outliers and directly interpretable as the average absolute difference between predicted and actual values. Accuracy, AUC, and F1 score are classification metrics used for problems where the model predicts discrete categories or classes (e.g., whether an email is spam or not). They are not suitable for evaluating the performance of a regression model.