Set MLflow Tracking to only track in your Azure Machine Learning workspace

Question

Hi Team,

As per below link

https://learn.microsoft.com/en-us/azure/machine-learning/how-to-use-mlflow-azure-databricks?tabs=custom

And section: Set MLflow Tracking to only track in your Azure Machine Learning workspace

We can set ML Flow Tracking URI to Azure ML in Databricks but i am getting below error upon setting and trying to create experiement

UnsupportedModelRegistryStoreURIException: Model registry functionality is unavailable; got unsupported URI 'azureml://eastus.api.azureml.ms/mlflow/v1.0/subscriptions/60732f07-e07d-4492-b2cc-e43155932aca/resourceGroups/RG-BHOGA/providers/Microsoft.MachineLearningServices/workspaces/[REDACTED]' for model registry data storage. Supported URI schemes are: ['', 'file', 'databricks', 'http', 'https', 'postgresql', 'mysql', 'sqlite', 'mssql']. See https://www.mlflow.org/docs/latest/tracking.html#storage for how to run an MLflow server against one of the supported backend storage locations.

Could you please advise what has to be done?

Answer 1

@bhogasena reddy kalakata Thanks for the question. Can you please share the notebook that you are trying. Here is the sample for ml flow tracking.

Answer 2

@bhogasena reddy kalakata , you probably forgot to install the plugin azureml-mlflow. This is required to interpret the URI and that's why you get the unsupported URI error.