Feat: semantic insights generator structure #138
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
leonidb
changed the title
leonidb
force-pushed
the
feat/409-semantic-insights-generator
branch
from
5424e66 to
c6c7663
Compare
yotam319-sparkbeyond
approved these changes
Jan 1, 2026
| test_dataset_with_strategy, | ||
| ) | ||
|
|
||
| __all__ = ['problem', 'real_llm_with_spec', 'test_dataset_with_strategy'] |
Contributor
There was a problem hiding this comment.
are we sure we want to share ['problem', 'real_llm_with_spec', 'test_dataset_with_strategy'] between all generators?
Collaborator
Author
There was a problem hiding this comment.
No, we'll definitely want a separate, non conversation tabular data, I just didn't want to decide on it in this PR, so this is a placeholder, for now
Contributor
There was a problem hiding this comment.
then define an placeholder ['problem', 'real_llm_with_spec', 'test_dataset_with_strategy'] in 'semantic_insights_generator/test_e2e.py' and remove this conftest.py.
| generation_model: LLMWithSpec | ||
| repair_model: LLMWithSpec | ||
| seed: int | None = None | ||
| validator: FeatureValidator = field(factory=LawAndOrderValidator) |
Contributor
There was a problem hiding this comment.
should be a list
Create minimal SemanticInsightsGenerator module with validation integration: - SemanticInsightsGenerator: Main orchestrator implementing FeatureGenerator - BasicFeatureGenerator: Core generation logic with validate_and_retry loop - LLMSqlFeatureCorrector: LLM-based SQL repair stub (TODO implementation) - GeneratedFeatureSpec: Minimal schema for LLM-generated features - Configurable validator and retry budgets with module defaults - Uses official validation infrastructure (LawAndOrderValidator, validate_and_retry) - Streaming feature generation (yields immediately after validation) - Seed parameter for reproducible generation
- Created nested test folder: tests/agentune/analyze/feature/gen/semantic_insights_generator/ - test_e2e.py: Validates generator instantiation and API contract - Expects empty results until LLM generation is implemented - refactor: Move InsightfulTextGenerator tests into a nested folder structure, matching the SemanticInsightsGenerator test structure
leonidb
force-pushed
the
feat/409-semantic-insights-generator
branch
from
bfdc6ce to
98f91f7
Compare
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.
What does this PR do?
Sets up the SemanticInsightsGenerator module structure with validation framework integration. LLM generation logic is stubbed out and ready for implementation.
Changes
Related Issues
Closes SparkBeyond/ao-core#113