feat(O4): relevance judge with Prometheus metrics

- RelevanceJudge: threshold-based relevance evaluation
- evaluate(): single query-result pair scoring
- evaluate_batch(): batch eval with precision/recall/F1
- Tracks: evals total, relevant/irrelevant, score histogram
- Updates precision/recall/F1 gauges per batch
- 4 unit tests passing
- Metrics: R1-R9 (9 metrics instrumented)
This commit is contained in:
2026-09-13 21:38:27 +09:00
parent 8334910144
commit 853c78bbdb
2 changed files with 162 additions and 0 deletions
+1
View File
@@ -2,6 +2,7 @@ pub mod endpoints;
pub mod handlers;
pub mod http_server;
pub mod metrics;
pub mod relevance_judge;
pub mod query;
pub mod auth;
pub mod ingest_worker;