adaptive-rag / src /eval /__init__.py
mcikalmerdeka's picture
replace Ragas evaluation script with DeepEval implementation for end-to-end evaluation. Update documentation to reflect changes in evaluation strategy and metrics. Remove deprecated Ragas script.
b2667ea
Raw
History Blame Contribute Delete
360 Bytes
"""Evaluation framework.
- ``golden.jsonl`` — hand-curated Q&A pairs spanning all 5 strategies.
- ``run_routing_eval.py`` — router-only accuracy + per-strategy breakdown.
- ``run_deepeval.py`` — full pipeline through DeepEval (faithfulness, answer
relevancy, contextual precision) + HTML report.
- ``reports/`` — generated outputs (gitignored).
"""