Skip to content

Remove duplicate Completeness class definition and refactor builtin scorers a bit#19064

Merged
xsh310 merged 3 commits intomlflow:masterfrom
xsh310:stack/built-in-session-level-judge-refactor
Nov 26, 2025
Merged

Remove duplicate Completeness class definition and refactor builtin scorers a bit#19064
xsh310 merged 3 commits intomlflow:masterfrom
xsh310:stack/built-in-session-level-judge-refactor

Conversation

@xsh310
Copy link
Collaborator

@xsh310 xsh310 commented Nov 26, 2025 •
edited
Loading

🥞 Stacked PR

Use this link to review incremental changes.

What changes are proposed in this pull request?

Refactor the built-in scorers a bit.

  • Remove the duplicate implementation of Completeness Judge that I merged accidentally
  • Create BuiltInSessionLevelScorer and extract common logic from UserFrustration and ConversationCompleteness

How is this PR tested?

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

Test Plan

Passes Built-in Scorer unit tests:

python3 -m pytest tests/genai/scorers/test_builtin_scorers.py

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.

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

Components

  • area/tracking: Tracking Service, tracking client APIs, autologging
  • area/models: MLmodel format, model serialization/deserialization, flavors
  • area/model-registry: Model Registry service, APIs, and the fluent client calls for Model Registry
  • area/scoring: MLflow Model server, model deployment tools, Spark UDFs
  • area/evaluation: MLflow model evaluation features, evaluation metrics, and evaluation workflows
  • area/gateway: MLflow AI Gateway client APIs, server, and third-party integrations
  • area/prompts: MLflow prompt engineering features, prompt templates, and prompt management
  • area/tracing: MLflow Tracing features, tracing APIs, and LLM tracing functionality
  • area/projects: MLproject format, project running backends
  • area/uiux: Front-end, user experience, plotting, JavaScript, JavaScript dev server
  • area/build: Build and test infrastructure for MLflow
  • area/docs: MLflow documentation pages

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)

@xsh310
Remove duplicate Completeness class definition
95a5115 
Signed-off-by: Xiang Shen <xshen.shc@gmail.com>
@xsh310
Refactor session-level scorers with BuiltInSessionLevelScorer base class
3f5309d 
Signed-off-by: Xiang Shen <xshen.shc@gmail.com>
xsh310
xsh310 commented Nov 26, 2025
View reviewed changes
mlflow/genai/scorers/builtin_scorers.py

@experimental(version="3.7.0")
@format_docstring(_MODEL_API_DOC)
class Completeness(BuiltInScorer):
Copy link
Collaborator Author

@xsh310 xsh310 Nov 26, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think I merged a duplicate definition for Completeness due to some git stack weirdness. Let's remove this!

xsh310
xsh310 commented Nov 26, 2025
View reviewed changes
mlflow/genai/scorers/builtin_scorers.py Outdated
"""

name: str
model: str | None = None
Copy link
Collaborator Author

@xsh310 xsh310 Nov 26, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Moving this model field to BuiltInScorer base class instead of defining it for every child class

Copy link
Collaborator

@dbczumar dbczumar Nov 26, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not every BuiltInScorer should be a Judge (this is something we should fix as a followup) - we may want to add built-in code-based scorers later. Accordingly, not every built-in scorer should have a model either. I'd rather add a subclass BuiltInJudge that inherits from Judge and has a model attribute, but we can do that later; for now, can we leave the model property on each scorer?

cc @AveshCSingh @B-Step62

@github-actions
Copy link
Contributor

github-actions bot commented Nov 26, 2025 •
edited
Loading

Documentation preview for e07e8ce is available at:

More info
  • Ignore this comment if this PR does not change the documentation.
  • The preview is updated when a new commit is pushed to this PR.
  • This comment was created by this workflow run.
  • The documentation was built by this workflow run.

@github-actions github-actions bot added v3.6.1 area/evaluation MLflow Evaluation rn/none List under Small Changes in Changelogs. labels Nov 26, 2025
@xsh310 xsh310 changed the title Remove duplicate Completeness class definition Remove duplicate Completeness class definition and refactor builtin scorers a bit Nov 26, 2025
dbczumar
dbczumar reviewed Nov 26, 2025
View reviewed changes
mlflow/genai/scorers/builtin_scorers.py
instructions=self.instructions,
model=self.model,
description=self.description,
feedback_value_type=Literal["yes", "no"],
Copy link
Collaborator

@dbczumar dbczumar Nov 26, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

How well does this render in the MLflow UI? I'd ideally like to see red / green highlighting depending on the value. Does this accomplish that?

Copy link
Collaborator Author

@xsh310 xsh310 Nov 26, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good question. I haven't looked into the UI yet. Let me create a ticket and investigate this!

dbczumar
dbczumar reviewed Nov 26, 2025
View reviewed changes
mlflow/genai/scorers/builtin_scorers.py
model=self.model,
description=self.description,
feedback_value_type=Literal[
"no_frustration", "frustration_resolved", "frustration_not_resolved"
Copy link
Collaborator

@dbczumar dbczumar Nov 26, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@xsh310 can we make this render nicely in the MLflow UI so that frustration_not_resolved is red and no_frustration / frustration_resolved are green?

dbczumar
dbczumar approved these changes Nov 26, 2025
View reviewed changes
Copy link
Collaborator

@dbczumar dbczumar left a comment •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM once comments are addressed (the ones about UI highlighting can be addressed as a follow-up, but let's try to prioritize it)

@xsh310
Update the api inventory after refactoring the builtin scorer
e07e8ce 
Signed-off-by: Xiang Shen <xshen.shc@gmail.com>
@xsh310 xsh310 force-pushed the stack/built-in-session-level-judge-refactor branch from b701871 to e07e8ce Compare November 26, 2025 22:52
@xsh310
Copy link
Collaborator Author

xsh310 commented Nov 26, 2025

Updated the PR to revert the model attribute change

@xsh310 xsh310 enabled auto-merge November 26, 2025 22:59
@xsh310 xsh310 added this pull request to the merge queue Nov 26, 2025
Merged via the queue into mlflow:master with commit 618c663 Nov 26, 2025
51 checks passed
@xsh310 xsh310 deleted the stack/built-in-session-level-judge-refactor branch November 26, 2025 23:32
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@dbczumar dbczumar dbczumar approved these changes

@alkispoly-db alkispoly-db Awaiting requested review from alkispoly-db

@AveshCSingh AveshCSingh Awaiting requested review from AveshCSingh

@smoorjani smoorjani Awaiting requested review from smoorjani

@serena-ruan serena-ruan Awaiting requested review from serena-ruan

@BenWilson2 BenWilson2 Awaiting requested review from BenWilson2

@harupy harupy Awaiting requested review from harupy

Assignees

No one assigned

Labels

area/evaluation MLflow Evaluation rn/none List under Small Changes in Changelogs. v3.6.1

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@xsh310 @dbczumar