machine-learning video for a retail company uses a recommendation model deployed on an Amazon SageMaker endpoint to provide product suggestions to customers in
A retail company uses a recommendation model deployed on an Amazon SageMaker endpoint to provide product suggestions to customers in real time. The company has developed a new version of the recommendation model and wants to evaluate its performance using live customer data without affecting the current production model. The company needs to determine whether the new model provides better recommendations before replacing the production model. What do you recommend?