AI Platform Notebooks makes it easy to manage JupyterLab instances through a protected, publicly available notebook instance URL. A JupyterLab instance is a Deep Learning virtual machine instance with the latest machine learning and data science libraries pre-installed.
Setup
-
Before you begin
Select a Google Cloud Platform (GCP) project, set up billing, and enable the Compute Engine API.
-
Create a new notebook instance
Create a new notebook instance.
-
Install dependencies
Install software that your notebook is dependent upon.
-
Choosing a virtual machine (VM) image
Select the VM image that works best for your AI Platform Notebooks instance.
-
Registering a legacy notebook with Notebooks API
Register notebooks created before Notebooks API availability to take advantage of new features.
-
Create a new notebook instance using a custom container
Use your own container with AI Platform Notebooks.
-
Use a notebook instance within a service perimeter
Use VPC Service Controls to set up a service perimeter for your AI Platform Notebooks instance.
-
Use a shielded VM with AI Platform Notebooks
Help protect your AI Platform Notebooks instance with Shielded VM.
Management
-
Save a notebook to GitHub
Save a notebook to GitHub.
-
Shut down a notebook instance
Shut down a notebook instance.
-
Manage hardware accelerators for a notebook instance
Manage hardware accelerators for a notebook instance.
-
Using customer-managed encryption keys (CMEK)
Configure a new AI Platform Notebooks instance to use CMEK.
-
Use SSH to access JupyterLab
Use SSH to access your JupyterLab instance.
-
AI Platform Notebooks IAM roles
Learn about the AI Platform Notebooks IAM roles.
-
Manage access using IAM roles
Grant and revoke access to AI Platform Notebooks resources.
-
AI Platform Notebooks audit logging
Manage AI Platform Notebooks audit logs.