[FR] Configure MLFlow logging level using an environment variable

[Flametaa]

Willingness to contribute

Yes. I can contribute this feature independently.

Proposal Summary

I would like to be able to configure default MLFlow logging level in scripts using an environment variable, rather than fetching the mlflow logger and setting the level using python as defined in https://mlflow.org/docs/latest/python_api/index.html#log-levels.

Motivation

What is the use case for this feature?

I want to inject this configuration by default in all MLFlow scripts so that the users will have a default configuration that is defined by the admins.

Why is this use case valuable to support for MLflow users in general?

Centralizing configuration in environment variables will make it easier to enforce standardization and to easily configure MLFlow features.

Why is this use case valuable to support for your project(s) or organization?

This will allow us to define the MLFlow logging level in all projects by injecting the variables by default in MLFlow scripts

Why is it currently difficult to achieve this use case?

Users currently have to define the logging level manually in every script as defined in https://mlflow.org/docs/latest/python_api/index.html#log-levels. The logging level defaults to info which logs a lot of unnecessary information when we are executing the script in a non interactive environment.

Details

No response

What component(s) does this bug affect?

• [ ] area/artifacts: Artifact stores and artifact logging
• [ ] area/build: Build and test infrastructure for MLflow
• [ ] area/deployments: MLflow Deployments client APIs, server, and third-party Deployments integrations
• [ ] area/docs: MLflow documentation pages
• [ ] area/examples: Example code
• [ ] area/model-registry: Model Registry service, APIs, and the fluent client calls for Model Registry
• [ ] area/models: MLmodel format, model serialization/deserialization, flavors
• [ ] area/recipes: Recipes, Recipe APIs, Recipe configs, Recipe Templates
• [ ] area/projects: MLproject format, project running backends
• [ ] area/scoring: MLflow Model server, model deployment tools, Spark UDFs
• [ ] area/server-infra: MLflow Tracking server backend
• [X] area/tracking: Tracking Service, tracking client APIs, autologging

What interface(s) does this bug affect?

• [ ] area/uiux: Front-end, user experience, plotting, JavaScript, JavaScript dev server
• [ ] area/docker: Docker use across MLflow's components, such as MLflow Projects and MLflow Models
• [ ] area/sqlalchemy: Use of SQLAlchemy in the Tracking Service or Model Registry
• [ ] area/windows: Windows support

What language(s) does this bug affect?

• [ ] language/r: R APIs and clients
• [ ] language/java: Java APIs and clients
• [ ] language/new: Proposals for new client languages

What integration(s) does this bug affect?

• [ ] integrations/azure: Azure and Azure ML integrations
• [ ] integrations/sagemaker: SageMaker integrations
• [ ] integrations/databricks: Databricks integrations

[B-Step62]

Hi @Flametaa, the proposal makes sense to me! Would you be interested in opening a PR for this feature? Here are some pointers to get started:

• Contribution guide: https://github.com/mlflow/mlflow/blob/master/CONTRIBUTING.md
• A file that declares all environment variables used by MLflow as a reference: https://github.com/mlflow/mlflow/blob/master/mlflow/environment_variables.py#L326

[github-actions[bot]]

@mlflow/mlflow-team Please assign a maintainer and start triaging this issue.

[karthikkurella]

@B-Step62 , I can start working on implementation and raise a PR for the design above. Please take a look. Thank you.

[steveepreston]

+1