ML model deployment
ML model deployment refers to putting a model into production. Once an ML model has been deployed into production, it can be used to predict new data. We recommend using the Vertex AI console or API to deploy a trained ML model. With Vertex AI, we can serve the model in production with batch prediction; we recommend specifying the appropriate hardware for your model and determining how to pass inputs to the model. With Vertex AI, we can also serve the model with online endpoint prediction; we recommend using Vertex AI Feature Store’s online serving API and turning on automatic scaling with a minimum of two nodes.
