Generally Available: Model Serving on Azure Databricks
Data opublikowania: 07 marca, 2023
Model Serving on Azure Databricks is now generally available
Model Serving on Azure Databricks is now generally. Azure Databricks Model Serving deploys machine learning models as a REST API, allowing you to build real-time ML applications like personalized recommendations, customer service chatbots, fraud detection, and more - all without the hassle of managing serving infrastructure. The first feature to launch under this model is Serverless Real-Time Inference.
This feature uses Serverless compute, which means that the endpoints and associated compute resources are managed and run in Azure Databricks. Read more