Create mlflow run for DSPy compile

[TomeHirata]

🛠 DevTools 🛠

Install mlflow from this PR

pip install git+https://github.com/mlflow/mlflow.git@refs/pull/14949/merge

For Databricks, use the following command:

%sh
OPTIONS=$(if pip freeze | grep -q 'git+https://github.com/mlflow/mlflow.git'; then echo '--force-reinstall --no-deps'; fi)
pip install $OPTIONS git+https://github.com/mlflow/mlflow.git@refs/pull/14949/merge#subdirectory=skinny

Related Issues/PRs

N/A

What changes are proposed in this pull request?

This PR introduces new options to DSPy autologging:

• log_compiles: create a run for each optimization and log program params
• log_models: log models when an optimizer is called

How is this PR tested?

• Existing unit/integration tests
• New unit/integration tests
• Manual tests

Does this PR require documentation update?

• No. You can skip the rest of this section.
• Yes. I've updated:
  • Examples
  • API references
  • Instructions

Release Notes

Is this a user-facing change?

• No. You can skip the rest of this section.
• Yes. Give a description of this change to be included in the release notes for MLflow users.

When mlflow.dspy.autolog is called with log_compiles, a mlflow run is created and program params are logged.

What component(s), interfaces, languages, and integrations does this PR affect?

Components

• 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
• area/tracking: Tracking Service, tracking client APIs, autologging

Interface

• 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

Language

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

Integrations

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

How should the PR be classified in the release notes? Choose one:

• rn/none - No description will be included. The PR will be mentioned only by the PR number in the "Small Bugfixes and Documentation Updates" section
• rn/breaking-change - The PR will be mentioned in the "Breaking Changes" section
• rn/feature - A new user-facing feature worth mentioning in the release notes
• rn/bug-fix - A user-facing bug fix worth mentioning in the release notes
• rn/documentation - A user-facing documentation change worth mentioning in the release notes

Should this PR be included in the next patch release?

Yes should be selected for bug fixes, documentation updates, and other small changes. No should be selected for new features and larger changes. If you're unsure about the release classification of this PR, leave this unchecked to let the maintainers decide.

What is a minor/patch release?

• Minor release: a release that increments the second part of the version number (e.g., 1.2.0 -> 1.3.0).
  Bug fixes, doc updates and new features usually go into minor releases.
• Patch release: a release that increments the third part of the version number (e.g., 1.2.0 -> 1.2.1).
  Bug fixes and doc updates usually go into patch releases.

• Yes (this PR will be cherry-picked and included in the next patch release)
• No (this PR will be included in the next minor release)

[github-actions]

Documentation preview for f9e2270 will be available when this CircleCI job
completes successfully. You may encounter a {"message":"not found"} error when reloading
a page. If so, add /index.html to the URL.

• Top page
• Changed pages (⚠️ only MDX file changes are detected ⚠️)

More info

• Ignore this comment if this PR does not change the documentation.
• It takes a few minutes for the preview to be available.
• The preview is updated when a new commit is pushed to this PR.
• This comment was created by https://github.com/mlflow/mlflow/actions/runs/13803540302.

[BenWilson2]

Looks good!

[chenmoneygithub]

for me to understand - why do we need a second level of protection? I think we are controlling through autolog()

[TomeHirata]

We changed the bahavior of autolog so that we include the callback even when log_trace is off. Therefore, we need to add this protection.

[chenmoneygithub]

gotcha, correct me if I am wrong - my read is you want to allow users to only track compile() call, in which scenario we will only log to mlflow at the Evaluator.__call__ hook?

[TomeHirata]

Yes, the evaluation callback will have additional logging for optimization progression when called within compile

[chenmoneygithub]

Seems this is not used in this PR? Could you help me understand the role of this variable? Are we going to use it to determine if evaluate should be put under a nested mlflow run?

[TomeHirata]

Yeah, this attribute will be used to determine if the stack is in the compile process or not. We will have some custom logic in the evaluation when it's inside a compile.