The PREDICT keyword for Apache Spark pools on Azure Synapse simplifies the process of enriching your data with predictive models for batch scoring at scale. PREDICT makes it possible to save time and reduce risk by allowing you to select MLFLow packaged models from AzureML model registry for scoring. It also drastically simplifies the handoff between an ML model creator and model user. This means you can easily score a large variety of ML models on Spark and streamline the batch scoring process.
Learn more.