fab-cli-tutorial-data-science

fab-cli-tutorial-data-science

Scripts to deploy the data science tutorial into your Fabric workspace.

This includes:

• 1 workspace
• 1 lakehouse, non-schema
• 4 notebooks
• 1 ML experiment (auto-created by notebooks)
• 3 ML models (auto-created by notebooks)
• 1 semantic model
• 1 Power BI report

Instructions

Make sure you have the Fabric CLI installed:

$ pip install ms-fabric-cli

Deploy from local

1. Clone the repository and jump to demo folder
2. Deploy demo

   $ ./fab-cli-tutorial-data-science/scripts/setup.sh --capacity-name <capacity_name> --postfix 87

Deploy using GitHub actions

1. Fork the repository
2. Create three secrets in your repository: FAB_TENANT_ID, FAB_CLIENT_ID, and FAB_CLIENT_SECRET
3. Go to the Actions tab and set up parameters
4. Run the deploy-fab-cli-tutorial-data-science workflow

Notes

See notes below for additional info:

• Use --capacity-name <capacity_name> --postfix 87 --spn_auth_enabled true --upn_objectid <user-objectid> to deploy using SPN auth from local.
• upn_objectid is used to grant user permission during service principal deployment.