Model Registry

• Model Registry
  • Prerequisites
  • Overview
  • Testing from the web interface
  • Testing from notebooks
  • Testing from pipelines
  • Troubleshooting
  • Quick reference
  • Appendix: useful commands
  • What to do next
• Model Registry scenarios
  • Overview
  • Prerequisites
  • Scenario 1. Add a model to Model Registry from a notebook
  • Scenario 2. Error on model name duplication
  • Scenario 3. Input data validation during model registration
  • Scenario 4. Model versioning from a notebook using the SDK in a virtual environment
  • Scenario 5. Happy path: register a model and version from MLOps Studio Pipelines
  • Scenario 6. Rerun by cloning the same pipeline run
  • Scenario 7. Incorrect Model Registry endpoint in a pipeline
  • Scenario 8. Register multiple model versions in the Model Registry from a pipeline
  • Scenario 9. Happy path: register a model and version directly in the Model Registry UI, then add a new version
  • Scenario 10. Error when attempting to add a model that already exists
  • What to do next
• Serve a model from Model Registry with KServe
  • Overview
  • Prerequisites
  • Namespace variable
  • What this test verifies
  • Step 0. Optional cleanup of stale pods
  • Step 1. Create the inference payload
  • Step 2. Deploy the InferenceService
  • Step 3. Wait for readiness and collect diagnostics
  • Step 4. Public test and SSO versus DNS detection
  • Step 5. DNS workaround with –resolve
  • Step 6. Direct-to-pod test to bypass ingress and authentication
  • Cleanup
  • What to do next