machine_learning
Complete ML workflow template
How to use it
machine_learning is exposed by the Fabric Analytics MCP MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the machine_learning tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.
FULL FABRIC ANALYTICS MCP LISTINGOther tools in Fabric Analytics MCP (39)
Batch job configuration
Batch job ID
Microsoft Fabric bearer token
Basic notebook with minimal setup
SQL or Spark code to execute
Optional token for pagination
Use your own notebook definition
Custom notebook definition (required if template is 'custom')
Structured data exploration template
Name of the dataset to query
Optional description
Display name for the new item
Optional environment ID to attach
Inventory notebooks, pipelines, linked services, and Spark jobs from Synapse
List all Synapse workspaces in your Azure subscription
Complete end-to-end migration with discovery, transformation, and provisioning
Transform Synapse notebooks to Fabric format (mssparkutils → notebookutils)
Format to return (ipynb or fabricGitSource)
ID of the item to retrieve
Filter by item type (optional)
Statement type (sql, spark, etc.)
Microsoft Fabric lakehouse ID
Optional default lakehouse name
Livy session ID
List of metrics to analyze
→ notebookutils
ID of the notebook to execute
NYC taxi trip data analysis with sample dataset
Optional parameters to pass to the notebook
SQL or KQL query to execute
Type of report (performance, usage, health, summary)
Comprehensive sales data analysis with sample dataset
Optional session configuration
Livy session ID
Spark Job Definition ID
Statement ID
Template type (blank, salesanalysis, nyctaxianalysis, dataexploration, machinelearning, custom)
Time range for metrics (1h, 24h, 7d, 30d)
Microsoft Fabric workspace ID