Machine learning experiments and models Git integration and deployment pipelines (Preview)

The Machine learning experiments and models integrate with the lifecycle management capabilities in Microsoft Fabric, providing a standardized collaboration between all development team members throughout the product's life. Lifecycle management facilitates an effective product versioning and release process by continuously delivering features and bug fixes into multiple environments. To learn more, see What is lifecycle management in Microsoft Fabric?.

Machine learning experiments and models Git integration

Machine learning (ML) experiments and models contain both metadata and data. ML experiments contain runs while ML models contains model versions. From a development workflow perspective, Notebooks might reference an ML experiment or an ML model.

As a principle, data is not stored in Git—only artifact metadata is tracked. By default, ML experiments and models are managed through the Git sync/update process, but experiment runs and model versions aren't tracked or versioned in Git and their data is preserved in workspace storage. Lineage between notebooks, experiments, and models is inherited from the Git-connected workspace.

Git representation

The following information is serialized and tracked in a Git connected workspace for machine learning experiment and models:

• Display name.
• Version.
• Logical guid. The tracked logical guid is an automatically generated cross-workspace identifier representing an item and its source control representation.
• Dependencies. Lineage between notebooks, experiments, and models are preserved across Git-connected workspaces, maintaining clear traceability among related artifacts.

Git integration capabilities

The following capabilities are available:

• Serialize ML experiment and model artifact metadata into a Git-tracked JSON representation.
• Support multiple workspaces linked to the same Git branch, enabling tracked metadata to sync across workspaces.
• Allow updates to be applied directly or controlled via pull requests to manage changes between upstream and downstream workspaces/branches.
• Track renames of experiments and models in Git to preserve identity across workspaces.
• No actions are taken on experiment runs or model versions; their data is preserved in workspace storage and isn't stored or overwritten by Git.

Machine learning experiments and models in deployment pipelines

Machine learning (ML) experiments and models are supported in Microsoft Fabric lifecycle management deployment pipelines. It enables environment segmentation best-practices.

ML experiments and models deployment pipelines integration capabilities:

• Support for deploying ML experiments and models across development, test, and production workspaces.
• Deployments synchronize only artifact metadata; experiment runs and model versions (their data) are preserved and aren't overwritten.
• Renames of experiments and models are propagated across workspaces when included in a deployment pipeline.
• Lineage between notebooks, experiments, and models is maintained across workspaces during pipeline deployments, preserving traceability between related artifacts.

Related content

• What is lifecycle management in Microsoft Fabric?
• Tutorial: Lifecycle management in Fabric
• Introduction to Git integration
• Introduction to deployment pipelines