Vertex AI Model Registry 是一个中央代码库,您可以在其中管理机器学习模型的生命周期。借助 Model Registry,您可以大致了解模型,以便更好地组织、跟踪和训练新版本。如果您想要部署模型版本,可以直接从存储库将其分配给端点,也可以使用别名将模型部署到端点。
Vertex AI Model Registry 支持自定义模型和所有 AutoML 数据类型 - 表格、图片和视频。Model Registry 还可支持 BigQuery ML 模型。如果您在 BigQuery ML 中训练了模型,则可以向 Model Registry 注册这些模型,而无需从 BigQuery ML 导出模型或将其导入 Model Registry。
在模型版本详情页面中,您可以进行评估、部署到端点、设置批量推理以及查看特定模型的详细信息。Vertex AI Model Registry 提供了一个简单易用的界面,方便您管理效果最佳的模型并将其部署到生产环境。
常规工作流
有许多有效的工作流用于 Model Registry。如需开始使用,建议您遵循以下准则,以了解您可以在 Model Registry 中执行的操作以及在模型训练过程中的各个阶段可以执行的操作。
将模型导入 Model Registry。
创建新模型,为模型版本分配默认别名,准备好投入到生产环境。
添加其他别名或标签,以帮助您管理和整理模型和模型版本。
将模型部署到端点以进行在线推理。
运行批量推理并启动模型评估流水线。
在模型详情页面查看模型详细信息并查看性能指标。
如需详细了解如何将 BigQuery ML 模型与 Vertex AI 集成,请参阅 BigQuery ML 文档。
使用 Dataplex Universal Catalog 搜索和发现模型
Dataplex Universal Catalog 是一个用于存储、管理和访问元数据的平台。借助 Dataplex Universal Catalog,您可以跨项目和区域搜索 Vertex AI 模型。
[[["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):2025-09-02。"],[],[],null,["# Introduction to Vertex AI Model Registry\n\n| To see an example of getting started with Vertex AI Model Registry,\n| run the \"Get started with Vertex AI Model Registry\" notebook in one of the following\n| environments:\n|\n| [Open in Colab](https://colab.research.google.com/github/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_registry/get_started_with_model_registry.ipynb)\n|\n|\n| \\|\n|\n| [Open in Colab Enterprise](https://console.cloud.google.com/vertex-ai/colab/import/https%3A%2F%2Fraw.githubusercontent.com%2FGoogleCloudPlatform%2Fvertex-ai-samples%2Fmain%2Fnotebooks%2Fofficial%2Fmodel_registry%2Fget_started_with_model_registry.ipynb)\n|\n|\n| \\|\n|\n| [Open\n| in Vertex AI Workbench](https://console.cloud.google.com/vertex-ai/workbench/deploy-notebook?download_url=https%3A%2F%2Fraw.githubusercontent.com%2FGoogleCloudPlatform%2Fvertex-ai-samples%2Fmain%2Fnotebooks%2Fofficial%2Fmodel_registry%2Fget_started_with_model_registry.ipynb)\n|\n|\n| \\|\n|\n| [View on GitHub](https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/model_registry/get_started_with_model_registry.ipynb)\n\nThe Vertex AI Model Registry is a central repository where you can manage\nthe lifecycle of your ML models. From the Model Registry,\nyou have an overview of your models so you can better organize, track,\nand train new versions. When you have a model version you would like to deploy,\nyou can assign it to an endpoint directly from the registry,\nor using aliases, deploy models to an endpoint.\n\nThe Vertex AI Model Registry supports custom models and all\nAutoML data types - tabular, image, and video. The\nModel Registry\ncan also support BigQuery ML models. If you have models trained in\nBigQuery ML, you can register them with the\nModel Registry without needing to export them from\nBigQuery ML or import them into the Model Registry.\n\nFrom the model version details page you can evaluate, deploy to an endpoint,\nset up batch inference, and view specific model details. The Vertex AI Model Registry\nprovides a straightforward and streamlined interface to manage and deploy your\nbest models to production.\n\nCommon workflow\n---------------\n\nThere are many valid workflows for working in the Model Registry.\nTo get started, you might want to follow these guidelines to understand what you can\ndo in the Model Registry and at what stage in your model-training journey.\n\n- Import models to the Model Registry.\n- Create new models, assign a model version the default alias, ready for production.\n- Add other aliases, or labels to help you manage and organize your models and model versions.\n- Deploy your models to an endpoint for online inference.\n- Run batch inference, and start your model evaluation pipeline.\n- View your model details and view performance metrics from the model details page.\n\nTo learn more about how to integrate your BigQuery ML models with\nVertex AI, see the\n[BigQuery ML documentation.](/bigquery-ml/docs/managing-models-vertex)\n\nSearch and discover models using Dataplex Universal Catalog\n-----------------------------------------------------------\n\nDataplex Universal Catalog is a platform for storing, managing, and accessing your\nmetadata. Dataplex Universal Catalog provides a way to search\nfor your Vertex AI models across projects and regions.\n\nFor more information, see [About data catalog management in\nDataplex Universal Catalog](/dataplex/docs/catalog-overview).\n\nWhat's next\n-----------\n\nTo get started using Vertex AI Model Registry, see:\n\n- [Import models to Vertex AI](/vertex-ai/docs/model-registry/import-model)\n- [Model versioning with Model Registry](/vertex-ai/docs/model-registry/versioning)\n- [How to use model version aliases](/vertex-ai/docs/model-registry/model-alias)\n- [BigQuery ML and Model Registry](/vertex-ai/docs/model-registry/model-registry-bqml)\n- [Copy a model in Vertex AI Model Registry](/vertex-ai/docs/model-registry/copy-model)"]]