Fix: Add mlflowconfig to eval base model#5745
Open
mollyheamazon wants to merge 2 commits intoaws:masterfrom
Open
Fix: Add mlflowconfig to eval base model#5745mollyheamazon wants to merge 2 commits intoaws:masterfrom
mollyheamazon wants to merge 2 commits intoaws:masterfrom
Conversation
mollyheamazon
temporarily deployed
to
auto-approve
GitHub Actions
Inactive
mollyheamazon
temporarily deployed
to
auto-approve
GitHub Actions
Inactive
mollyheamazon
temporarily deployed
to
auto-approve
GitHub Actions
Inactive
mollyheamazon
changed the title
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Fix MLflow tracking support for base model only evaluations.
The three
_BASE_MODEL_ONLYpipeline templates (DETERMINISTIC, CUSTOM_SCORER, LLMAJ) were missing theMlflowConfigblock that their fine-tuned counterparts already had. The auto-resolve infrastructure inbase_evaluator.pywas already wiring upmlflow_resource_arnfor all evaluator types, but the resolved ARN was silently dropped because the templates had noMlflowConfigblock to render it into.Changes:
LLMAJ_TEMPLATE_BASE_MODEL_ONLY, DETERMINISTIC_TEMPLATE_BASE_MODEL_ONLY, and CUSTOM_SCORER_TEMPLATE_BASE_MODEL_ONLYinpipeline_templates.pytest_pipeline_templates.pyto assertMlflowConfigis present (previously incorrectly asserted absent)mlflow_resource_arnintest_benchmark_evaluator.pyinteg test for base model only evaluation to validate the fix end-to-endBy submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.