How do I provision a SageMaker AI Project?

Resolution

Note: If you receive errors when you run AWS Command Line Interface (AWS CLI) commands, then see Troubleshooting errors for the AWS CLI. Also, make sure that you're using the most recent AWS CLI version.

Before you provision a SageMaker AI Project, make sure to turn on project templates for your SageMaker AI Studio domain. When you create a domain, you activate the templates import after you take one of the following actions:

• You turn on SageMaker AI Project templates and Amazon SageMaker AI JumpStart for the account.
• You call the EnableSagemakerServicecatalogPortfolio API for an existing SageMaker AI Studio domain.

Grant your SageMaker AI Studio users the required permissions to use Projects. Confirm that you turned on the following options under the Domain Settings tab:

• Enable Amazon SageMaker AI project templates and Amazon SageMaker JumpStart for this account
• Enable Amazon SageMaker AI project templates and Amazon SageMaker JumpStart for Studio users

You can also use the GetSagemakerServicecatalogPortfolioStatus API, or run the get-sagemaker-servicecatalog-portfolio-status AWS CLI command:

aws sagemaker get-sagemaker-servicecatalog-portfolio-status

After you turn on the projects templates, you can provision a SageMaker AI Project directly from a SageMaker AI Studio domain. Or, you can create a SageMaker AI Project through an API or the AWS CLI.

Note: If you use an API or the AWS CLI, then you must provide the ProductId and ProvisioningArtifactId.

To get the the product and provisioning artifact IDs, complete the following steps:

1. Open the Service Catalog console.
2. In the navigation pane, under Administration, choose Portfolios.
3. Choose the Imported tab.
4. In the search bar, enter Amazon SageMaker Solutions and MLOps products.
5. Choose Amazon SageMaker Solutions and MLOps products.
6. Review the Products page to get the Product ID.
7. Select the product or project template that you want to use, for example MLOps template for model building and training.
8. Review the Product list page to get the Provisioning Artifact ID.

Use the project and provisioning artifact IDs in your CreateProject API or run the create-project AWS CLI command:

aws sagemaker create-project --project-name your-project--service-catalog-provisioning-details ProductId="product-id-example",ProvisioningArtifactId="pa-id-example"

Note: Replace your-project with your project name, product-id-example with your product ID, and pa-id-example with your provisioning artifact ID.