部署和提供模型Deploy and serve models

通过 MLflow 将模型部署到生产环境Deploy models to production with MLflow

有关通过 MLflow 部署模型的信息,请参阅记录、加载和部署 MLflow 模型For information about deploying models with MLflow, see Log, load, and deploy MLflow Models. 以下笔记本演示了如何使用 MLflow 模型注册表来生成、管理和部署模型。The following notebook illustrates how to use MLflow Model Registry to build, manage, and deploy a model.

MLflow 模型注册表示例MLflow Model Registry example

使用 MLflow 提供模型Serve models with MLflow

Azure Databricks 提供 MLflow 模型提供功能,使你能够将模型注册表中的机器学习模型作为 REST 终结点(它们基于模型版本的可用性和阶段自动更新)托管。Azure Databricks provides MLflow Model Serving, which allows you to host machine learning models from the Model Registry as REST endpoints that are updated automatically based on the availability of model versions and their stages. MLflow 模型提供功能可用于 Python MLflow 模型。MLflow Model Serving is available for Python MLflow models.

运行 Azure Databricks 作业Run a Azure Databricks job

你可以创建 Azure Databricks 作业,以立即或按计划运行笔记本或 JAR。You can create a Azure Databricks job to run a notebook or JAR either immediately or on a scheduled basis.