本頁面說明如何在Google Cloud 主控台中透過 Cloud Marketplace 建立深度學習 VM 映像檔執行個體,而不必使用指令列。
事前準備
Sign in to your Google Cloud account. If you're new to
Google Cloud,
create an account to evaluate how our products perform in
real-world scenarios. New customers also get $300 in free credits to
run, test, and deploy workloads.
In the Google Cloud console, on the project selector page,
select or create a Google Cloud project.
您可以選擇「Enable access to JupyterLab via URL instead of SSH (Beta)」。啟用這項 Beta 版功能後,您就能使用網址存取 JupyterLab 執行個體。 Google Cloud 專案中具有編輯者或擁有者角色的使用者都可存取這個網址。這項功能目前僅適用於美國、歐盟和亞洲。
選取開機磁碟類型和開機磁碟大小。
選取所需的網路設定。
按一下 [Deploy] (部署)。
如果您選擇安裝 NVIDIA 驅動程式,安裝作業需要 3 到 5 分鐘才能完成。
VM 部署完成後,這個頁面會更新,並提供存取該執行個體的操作說明。
後續步驟
如要瞭解如何透過 Google Cloud 主控台或指令列連線至新的深度學習 VM 執行個體,請參閱「連線至執行個體」一文。執行個體名稱是您指定的「部署作業名稱」,後面加上 -vm。
[[["容易理解","easyToUnderstand","thumb-up"],["確實解決了我的問題","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["難以理解","hardToUnderstand","thumb-down"],["資訊或程式碼範例有誤","incorrectInformationOrSampleCode","thumb-down"],["缺少我需要的資訊/範例","missingTheInformationSamplesINeed","thumb-down"],["翻譯問題","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["上次更新時間:2025-06-16 (世界標準時間)。"],[[["This guide outlines how to create a Deep Learning VM instance directly from the Google Cloud Marketplace within the console, eliminating the need for command-line operations."],["Before creating the instance, you must select a specific Deep Learning VM image based on your preferred framework and processor type, and check that enough GPU quota is available if you are planning to use GPUs."],["The instance creation process involves selecting a deployment name, zone, machine type, optional GPU settings, and machine learning framework, then it includes selecting a boot disk and networking settings before deployment."],["If you are planning to use GPUs, you will need to install the NVIDIA drivers, which can be done automatically on the first startup, and you also have the choice of enabling JupyterLab access via URL."],["After deployment, you will be provided instructions to access the instance, and the instance name is created by appending `-vm` to the deployment name that was chosen during setup."]]],[]]