LLM assisted rules generation #577
Open
Conversation
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
souravg-db2
requested review from
tombonfert
and removed request for
a team
|
❌ 359/360 passed, 3 flaky, 1 failed, 1 skipped, 3h13m6s total ❌ test_e2e_workflow: pyspark.errors.exceptions.connect.SparkConnectGrpcException: () BAD_REQUEST: session_id is no longer usable. Generate a new session_id by detaching and reattaching the compute and then try again [sessionId=6c16fe1b-7232-4f41-9f9b-d997bf6a6890, reason=INACTIVITY_TIMEOUT]. (requestId=8537a2da-e41e-4ecd-a0b5-693b24c0c70c) (12m28.318s)Flaky tests:
Running from acceptance #2669 |
There was a problem hiding this comment.
Pull Request Overview
This PR introduces LLM-assisted data quality rules generation capabilities to the DQX framework. The implementation leverages DSPy for LLM orchestration and includes comprehensive training examples and validation mechanisms.
- Adds core LLM functionality with DSPy integration for automated data quality rule generation
- Implements training dataset and validation framework for optimizing rule generation accuracy
- Provides utility functions for schema metadata extraction and function documentation
Reviewed Changes
Copilot reviewed 6 out of 6 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
| src/databricks/labs/dqx/llm/llm_core.py | Core LLM implementation with DSPy module, validation logic, and compiler configuration |
| src/databricks/labs/dqx/llm/llm_engine.py | High-level interface for generating business rules using LLM |
| src/databricks/labs/dqx/llm/utils.py | Utility functions for schema extraction, training data loading, and function documentation |
| src/databricks/labs/dqx/llm/resources/training_examples.yml | Training dataset with business scenarios and expected quality rules |
| tests/unit/test_llm_utils.py | Unit tests for LLM utility functions and training data validation |
| pyproject.toml | Adds dspy dependency for LLM functionality |
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
mwojtyczka
requested review from
mwojtyczka
and removed request for
tombonfert
mwojtyczka
reviewed
Sep 16, 2025
mwojtyczka
reviewed
Sep 16, 2025
mwojtyczka
reviewed
Sep 16, 2025
mwojtyczka
reviewed
Sep 16, 2025
| @@ -0,0 +1,93 @@ | |||
| - name: "product_code_not_null_or_empty" | |||
There was a problem hiding this comment.
this should be extended to contain more examples
mwojtyczka
requested changes
Sep 16, 2025
| logger.warning(f"✗ Rules validation errors: {validation_status.errors}") | ||
|
|
||
| # Content similarity score (30%) | ||
| similarity_score = _calculate_rule_similarity(expected_rules, actual_rules) |
There was a problem hiding this comment.
this is not needed: DQEngineCore.validate_checks(actual_rules) validates the generated rule is correct. There should be no need to compare against expected. It would also be problematic to maintain because a lot of arguments can be optional and providing examples of all possible combinations won't be possible
|
|
||
| # Json parsing score (20%) | ||
| try: | ||
| actual_rules = json.loads(actual) |
There was a problem hiding this comment.
this does not seem to be necessary since the forward method in DQRuleGeneration is making sure a valid json is outputed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Changes
Linked issues
Resolves #..
Tests