Profile driven evals#8
Merged
Merged
Conversation
Introduces reusable evaluation profiles with metrics, weights, and criteria; auto-routing adapter registry; composite scoring with quality gates; evaluation policies for CI/CD; CLI entry point; and built-in profiles. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add BatchResult schema and evaluate_profile_batch() for multi-case evaluation from data files - Add context_precision_with_reference to the metric registry - Specify floating-point tolerance for weight validation - Document score normalization assumption (0.0-1.0 range) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
10-task TDD plan covering: exceptions, metric registry, output schemas, built-in profiles, ProfileManager (loading/validation/scoring), framework integration, CLI, and final validation. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Implements the complete profile-driven evaluation system with: - Metric registry with priority-based auto-routing to RAGAS/DeepEval adapters - ProfileManager for YAML profile loading, validation, policy resolution, scoring, and batch evaluation - MetricEvaluation, ProfileResult, and BatchResult Pydantic schemas - Built-in profiles (rag_default, strict, smoke_test) - Click CLI with evaluate, profiles list/show/validate commands - Custom exceptions (ProfileValidationError, UnknownMetricError, ProfileNotFoundError) - 67 new tests (220 total), all passing with zero lint errors Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
…or handling - Add resolve_metric_candidates() and use it in _run_metrics() to fall back to next-priority adapter when the primary fails - Return deepcopy from get_profile() to prevent callers from mutating internal state - Catch yaml.YAMLError in CLI validate command for clean error output - Add tests for all three fixes (5 new tests, 225 total) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
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.
No description provided.