Run Mlflow using the metadata CLI

In this section, we provide guides and references to use the Mlflow connector.

Configure and schedule Mlflow metadata and profiler workflows from the OpenMetadata UI:

OpenMetadata 0.12 or later

To deploy OpenMetadata, check the Deployment guides.

To run the Ingestion via the UI you'll need to use the OpenMetadata Ingestion Container, which comes shipped with custom Airflow plugins to handle the workflow deployment.

To run the Mlflow ingestion, you will need to install:

All connectors are defined as JSON Schemas. Here you can find the structure to create a connection to Mlflow.

In order to create and run a Metadata Ingestion workflow, we will follow the steps to create a YAML configuration able to connect to the source, process the Entities if needed, and reach the OpenMetadata server.

The workflow is modeled around the following JSON Schema

This is a sample config for Mlflow:

trackingUri: Mlflow Experiment tracking URI. E.g., http://localhost:5000

registryUri: Mlflow Model registry backend. E.g., mysql+pymysql://mlflow:password@localhost:3307/experiments

The sourceConfig is defined here:

markDeletedMlModels: Set the Mark Deleted Ml Models toggle to flag ml models as soft-deleted if they are not present anymore in the source system.

To send the metadata to OpenMetadata, it needs to be specified as type: metadata-rest.

The main property here is the openMetadataServerConfig, where you can define the host and security provider of your OpenMetadata installation.

For a simple, local installation using our docker containers, this looks like:

filename.yaml

We support different security providers. You can find their definitions here.

  • JWT tokens will allow your clients to authenticate against the OpenMetadata server. To enable JWT Tokens, you will get more details here.
  • You can refer to the JWT Troubleshooting section link for any issues in your JWT configuration. If you need information on configuring the ingestion with other security providers in your bots, you can follow this doc link.

First, we will need to save the YAML file. Afterward, and with all requirements installed, we can run:

Note that from connector to connector, this recipe will always be the same. By updating the YAML configuration, you will be able to extract metadata from different sources.