Amazon Personalize now provides offline model metrics for recommenders enabling you to evaluate the quality of recommendations. A recommender is a resource that provides recommendations optimized for specific use cases, such as “Frequently bought together” for Retail and “Top picks for you” for Media and Entertainment. Offline metrics are metrics that Amazon Personalize generates when you create a recommender. You can use offline metrics to analyze the performance of the recommender’s underlying model. Offline metrics allow you to compare the model with other models trained on the same data. The metrics provided include coverage, mean reciprocal rank, normalized discounted cumulative gain (NDCG) and precision.
Source:: Amazon AWS