使用 Azure Machine Learning 的 MLOpsMLOps with Azure Machine Learning

MLOps (機器學習作業) 以 DevOps 準則和做法為基礎,可提高工作流程的效率,例如持續整合、傳遞和部署。MLOps (machine learning operations) is based on DevOps principles and practices that increase workflow efficiencies like continuous integration, delivery, and deployment. MLOps 會將這些原則套用至機器學習程式,以便:MLOps applies these principles to the machine learning process in order to:

  • 更快速地實驗和開發模型。Experiment and develop models more quickly.
  • 更快速地將模型部署到生產環境。Deploy models to production more quickly.
  • 實務和精簡品質保證。Practice and refine quality assurance.

Azure Machine Learning 提供下列 MLOps 功能:Azure Machine Learning provides the following MLOps capabilities:

  • 建立可重現的管線。Create reproducible pipelines. 機器學習管線可讓您針對資料準備、定型和評分程式,定義可重複且可重複使用的步驟。Machine Learning pipelines enable you to define repeatable and reusable steps for your data preparation, training, and scoring processes.
  • 建立可重複使用的軟體環境 ,以定型和部署模型。Create reusable software environments for training and deploying models.
  • 從任何地方註冊、封裝及部署模型。Register, package, and deploy models from anywhere. 您可以追蹤使用模型所需的相關中繼資料。You can track the associated metadata required to use the model.
  • 捕捉端對端生命週期的治理資料。Capture the governance data for the end-to-end lifecycle. 記錄的資訊可能包括正在發佈模型的人員、進行變更的原因,以及在生產環境中部署或使用模型的時間。The logged information can include who is publishing models, why changes were made, and when models were deployed or used in production.
  • 通知並警示生命週期中的事件。Notify and alert on events in the lifecycle. 例如,您可以取得實驗完成、模型註冊、模型部署和資料漂移偵測的警示。For example, you can get alerts for experiment completion, model registration, model deployment, and data drift detection.
  • 監視應用程式,以瞭解操作與機器學習相關的問題。Monitor applications for operational and machine learning-related issues. 比較定型和推斷之間的模型輸入、探索模型特定的計量,並在您的機器學習基礎結構上提供監視和警示。Compare model inputs between training and inference, explore model-specific metrics, and provide monitoring and alerts on your machine learning infrastructure.
  • 使用 Azure Machine Learning 和 Azure 管線將端對端機器學習生命週期自動化。Automate the end-to-end machine learning lifecycle with Azure Machine Learning and Azure Pipelines. 透過管線,您可以經常更新模型、測試新的模型,並與其他應用程式和服務持續推出新的機器學習模型。With pipelines, you can frequently update models, test new models, and continuously roll out new machine learning models alongside your other applications and services.

使用 Azure Machine Learning 進行 MLOps 的最佳做法Best practices for MLOps with Azure Machine Learning

模型與程式碼不同,因為它們有有機的生命週期,除非維持,否則會下降。Models differ from code because they have an organic shelf life and will deteriorate unless maintained. 部署完成後,他們可以加入真正的商業價值,而且當資料科學家提供採用標準工程實務的工具時,這會變得更容易。After they're deployed, they can add real business value, and this gets easier when data scientists are given the tools to adopt standard engineering practices.

使用 Azure MLOps 可協助您:MLOps with Azure helps you:

  • 建立可重現的模型和可重複使用的訓練管線。Create reproducible models and reusable training pipelines.
  • 簡化品質控制和 a/B 測試的模型封裝、驗證和部署。Simplify model packaging, validation, and deployment for quality control and a/B testing.
  • 說明並觀察模型行為,並將重新定型程式自動化。Explain and observe model behavior, and automate the retraining process.

MLOps 改進了機器學習解決方案的品質與一致性。MLOps improves the quality and consistency of your machine learning solutions. 若要深入瞭解如何使用 Azure Machine Learning 來管理模型的生命週期,請參閱 MLOps:使用 Azure Machine learning 進行模型管理、部署和監視To learn more about how to use Azure Machine Learning to manage the lifecycle of your models, see MLOps: model management, deployment, and monitoring with Azure Machine Learning.

下一步Next steps

若要深入瞭解,請閱讀及探索下列資源:Learn more by reading and exploring the following resources: