BigQuery ML 是一项 Google Cloud服务,可让您使用标准 SQL 查询在 BigQuery ML 中创建和执行机器学习模型。借助 Vertex AI,您可以在统一平台中使用预训练和自定义工具。在 Model Registry 中注册 BigQuery ML 模型后,您便可以将它们与其他机器学习模型一起管理,以进行版本控制、评估和部署。
借助此集成,您可以选择要注册到 Model Registry 的 BigQuery ML 模型。注册 BigQuery ML 模型后,您可以将 BigQuery ML 模型部署到端点以进行在线推理。
[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-09-04。"],[],[],null,["# BigQuery ML and Model Registry\n\n[BigQuery ML](/bigquery/docs/bqml-introduction) is a Google Cloud\nservice that lets you create and execute machine learning models in\nBigQuery ML by using standard SQL queries. With\nVertex AI, you can use pre-trained and custom tooling within a\nunified platform. After you register your\nBigQuery ML models in the Model Registry,\nyou can manage them with your other ML models for versioning,\nevaluation, and deployment.\n\nWith this integration, you can choose which BigQuery ML models to\nregister to the Model Registry. After you've registered\nyour BigQuery ML models, you can deploy your\nBigQuery ML model to an endpoint for online inference.\n\nTo learn how to integrate your BigQuery ML models with\nVertex AI Model Registry, see\n[Manage BigQuery ML models in Vertex AI](/bigquery/docs/managing-models-vertex).\n\nLimitations\n-----------\n\n- You can't register\n [remote models](/bigquery/docs/bqml-introduction#remote_models).\n\n- The following models can be registered in\n Model Registry, but they can't be deployed in\n Vertex AI:\n\n - [Imported XGBoost models](/bigquery/docs/reference/standard-sql/bigqueryml-syntax-create-xgboost)\n - [`ARIMA_PLUS` models](/bigquery/docs/reference/standard-sql/bigqueryml-syntax-create-time-series)\n - [`ARIMA_PLUS_XREG` models](/bigquery/docs/reference/standard-sql/bigqueryml-syntax-create-multivariate-time-series)"]]