什麼是 Azure Machine Learning?What is Azure Machine Learning?

在本文中,您會了解 Azure Machine Learning,這是一種雲端式環境,可用來定型、部署、自動化、管理及追蹤 ML 模型。In this article, you learn about Azure Machine Learning, a cloud-based environment you can use to train, deploy, automate, manage, and track ML models.

Azure Machine Learning 可用於任何一種機器學習,從傳統 ML 到深度學習、受監督和不受監督的學習。Azure Machine Learning can be used for any kind of machine learning, from classical ml to deep learning, supervised, and unsupervised learning. 無論您偏好使用 SDK 撰寫 Python 或 R 程式碼或在 Studio 中使用零程式碼/低程式碼選項,都可以在 Azure Machine Learning 工作區中建立、定型及追蹤機器學習和深度學習模型。Whether you prefer to write Python or R code with the SDK or work with no-code/low-code options in the studio, you can build, train, and track machine learning and deep-learning models in an Azure Machine Learning Workspace.

開始訓練您的本機電腦,然後向外擴增到雲端。Start training on your local machine and then scale out to the cloud.

此服務也會與熱門的深度學習和增強式開放原始碼工具 (例如 PyTorch、TensorFlow、scikit-learn 和 Ray RLlib) 交互操作。The service also interoperates with popular deep learning and reinforcement open-source tools such as PyTorch, TensorFlow, scikit-learn, and Ray RLlib.

提示

免費試用!Free trial! 如果您沒有 Azure 訂用帳戶,請在開始前先建立一個免費帳戶。If you don’t have an Azure subscription, create a free account before you begin. 立即試用免費或付費版本的 Azure Machine LearningTry the free or paid version of Azure Machine Learning today. 即可取得用於 Azure 服務的點數。You get credits to spend on Azure services. 信用額度用完之後,您可以保留帳戶並使用免費的 Azure 服務After they're used up, you can keep the account and use free Azure services. 除非您明確變更您的設定且同意付費,否則我們絕對不會從您的信用卡收取任何費用。Your credit card is never charged unless you explicitly change your settings and ask to be charged.

什麼是機器學習?What is machine learning?

機器學習是一項資料科學技術,可讓電腦使用現有資料來預測未來的行為、結果和趨勢。Machine learning is a data science technique that allows computers to use existing data to forecast future behaviors, outcomes, and trends. 使用機器學習,電腦不需要明確進行程式設計就能學習。By using machine learning, computers learn without being explicitly programmed.

機器學習的預測可讓應用程式和裝置更聰明。Forecasts or predictions from machine learning can make apps and devices smarter. 例如,當您線上購物時,機器學習服務可根據您已經購買的產品,協助推薦其他您可能會想要的產品。For example, when you shop online, machine learning helps recommend other products you might want based on what you've bought. 或是當您的信用卡被刷過時,機器學習服務可將該筆交易與交易資料庫進行比對,協助偵測詐騙。Or when your credit card is swiped, machine learning compares the transaction to a database of transactions and helps detect fraud. 而且,當您的吸塵器機器人清潔房間時,機器學習服務可協助它判斷作業是否已完成。And when your robot vacuum cleaner vacuums a room, machine learning helps it decide whether the job is done.

適用於每個工作的機器學習工具Machine learning tools to fit each task

Azure Machine Learning 為開發人員和資料科學家提供其機器學習工作流程需要的所有工具,包括:Azure Machine Learning provides all the tools developers and data scientists need for their machine learning workflows, including:

您甚至可使用 MLflow 來追蹤計量及部署模型 或 Kubeflow,以建置端對端工作流程管線You can even use MLflow to track metrics and deploy models or Kubeflow to build end-to-end workflow pipelines.

以 Python 或 R 建置 ML 模型Build ML models in Python or R

使用 Azure Machine Learning Python SDKR SDK,開始訓練您的本機電腦。Start training on your local machine using the Azure Machine Learning Python SDK or R SDK. 然後可以擴增至雲端。Then, you can scale out to the cloud.

透過許多可用的計算目標 (例如 Azure Machine Learning Compute 和 Azure Databricks) 及進階的超參數微調服務,您可以使用雲端功能更快地建置更好的模型。With many available compute targets, like Azure Machine Learning Compute and Azure Databricks, and with advanced hyperparameter tuning services, you can build better models faster by using the power of the cloud.

您也可以使用 SDK,自動進行模型定型和微調You can also automate model training and tuning using the SDK.

在 Studio 中建置 ML 模型Build ML models in the studio

Azure Machine Learning Studio 是 Azure Machine Learning 中的 Web 入口網站,適用於可供模型定型、部署和資產管理的低程式碼和無程式碼選項。Azure Machine Learning studio is a web portal in Azure Machine Learning for low-code and no-code options for model training, deployment, and asset management. Studio 會與 Azure Machine Learning SDK 整合,以提供順暢的體驗。The studio integrates with the Azure Machine Learning SDK for a seamless experience. 如需詳細資訊,請參閱什麼是 Azure Machine Learning StudioFor more information, see What is Azure Machine Learning studio.

MLOps:部署和生命週期管理MLOps: Deploy & lifecycle management

當您有正確的模型時,您可以在 Web 服務中、在 IoT 裝置上或從 Power BI 輕鬆使用它。When you have the right model, you can easily use it in a web service, on an IoT device, or from Power BI. 如需詳細資訊,請參閱有關如何部署和部署位置的文章。For more information, see the article on how to deploy and where.

接著,您可以使用適用於 Python 的 Azure Machine Learning SDKAzure Machine Learning Studio機器學習 CLI 來管理所部署的模型。Then you can manage your deployed models by using the Azure Machine Learning SDK for Python, Azure Machine Learning studio, or the machine learning CLI.

這些模型可被取用並即時非同步地傳回大量資料的相關預測。These models can be consumed and return predictions in real time or asynchronously on large quantities of data.

另外,透過進階機器學習管線,您可以在資料準備、模型定型與評估及部署的每個步驟上共同作業。And with advanced machine learning pipelines, you can collaborate on each step from data preparation, model training and evaluation, through deployment. 管線可讓您執行下列作業:Pipelines allow you to:

  • 自動化雲端中的端對端機器學習程序Automate the end-to-end machine learning process in the cloud
  • 重複使用元件,並且只在需要時重新執行步驟Reuse components and only rerun steps when needed
  • 在每個步驟中使用不同的計算資源Use different compute resources in each step
  • 執行批次評分工作Run batch scoring tasks

如果您想要使用指令碼來自動化機器學習工作流程,機器學習 CLI 會提供命令列工具來執行一般工作,例如提交定型回合或部署模型。If you want to use scripts to automate your machine learning workflow, the machine learning CLI provides command-line tools that perform common tasks, such as submitting a training run or deploying a model.

若要開始使用 Azure Machine Learning,請參閱以下的後續步驟To get started using Azure Machine Learning, see Next steps.

與其他服務整合Integration with other services

Azure Machine Learning 可與 Azure 平台上的其他服務搭配運作,也可以與 Git 和 MLFlow 等開放原始碼工具整合。Azure Machine Learning works with other services on the Azure platform, and also integrates with open source tools such as Git and MLFlow.

安全通訊Secure communications

您的 Azure 儲存體帳戶、計算目標和其他資源可以在虛擬網路內安全地使用,以定型模型及執行推斷。Your Azure Storage account, compute targets, and other resources can be used securely inside a virtual network to train models and perform inference. 如需詳細資訊,請參閱虛擬網路隔離和隱私權概觀For more information, see Virtual network isolation and privacy overview.

後續步驟Next steps