{"commit":{"body":[{"label":"Why","text":"다채널 처리 저하에서 불필요한 준비 비용을 줄이고 Pose batch 수용 범위를 확인한다."},{"label":"What","text":"pool4·추론용mux576·제외Frame 직렬화 생략·경량 generation fence·collector 최소 제거를 구현했다."},{"label":"Impact","text":"Pose4x4/5ms 반영 후4채널은 결과1.70~1.85FPS로미달. 임시등록13개정리,기존3개Medium설정/running복원. 품질은사용자평가대기다."},{"label":"Verify","text":"source/image 각67tests PASS. batch4 shape/finite PASS,실제Tritonbatch1~3관측. 8/12/16채널미실행, decodedqueue2보류."},{"label":"Rollback","text":"source exact diff 및private image/config/model CAS 원장으로부분복원. 기존engine과image보존. 무관한서비스변경없음."}],"title":"refactor(runtime): Pose batch 후보와 경량 처리 경로 반영"},"created_at":"2026-09-16T04:19:06+00:00","expires_at":"2026-09-17T04:19:06+00:00","job_id":"contextra-pose-batch-20260916-124403","paths":[{"change":"modified","content_sha256":"e0b7df76245cdfad9cf89402039ae82e87a4cd79afaa2ea04e5c38c40b8615ee","path":"app/frame_ingress.py"},{"change":"modified","content_sha256":"e5b6eb34ffb24b1f34352446dbaf34d4cab80f8f2ef258c6433f4e5967f3afca","path":"app/medium_preprocess.py"},{"change":"modified","content_sha256":"e37143ede9ab28e0d0bf84934b616dcb2006b1e49dd266ecf8bfe7b4546fc880","path":"app/multi_pipeline.py"},{"change":"modified","content_sha256":"376177ee6959d1937c7eff88e139f75dd39f40f558c95725947242a5e566ca9b","path":"app/source_adapters/rtsp.py"},{"change":"modified","content_sha256":"83113b7ad10cb238451e9683b7872b3ddd920443aec2b96fec5718e2fd2e4a11","path":"app/stream_lifecycle.py"},{"change":"modified","content_sha256":"21ecae9e89fb70e2da67b95c1c1982aa8bb533f81121d1df4110871329d37496","path":"app/stream_manager.py"},{"change":"modified","content_sha256":"4ea144d38924076d7fc9748fb7693f29faca77fbad49e1b6d23345b7786564c0","path":"docs/profiles/ltb1021/topics/runtime-lifecycle/README.ko.md"},{"change":"added","content_sha256":"46da314d4720e257f8580bb145fc83d6b443ecf08e38fe5378241502dc2097a2","path":"docs/profiles/ltb1021/topics/runtime-lifecycle/decision/20260916T124403+0900-decision-pose-batch.md"},{"change":"added","content_sha256":"2bb64dc6c384827322aa7b09d820b9326db98907eaf6cc5178d2dda966f46f14","path":"docs/profiles/ltb1021/topics/runtime-lifecycle/runs/20260916T124403+0900-run-pose-batch.md"},{"change":"added","content_sha256":"4895e98185b1f721bf825ac6a370bab12ada57b1d47c473d8b9de81746eda35b","path":"docs/profiles/ltb1021/topics/runtime-lifecycle/runs/20260916T130958+0900-run-pose-batch-runtime.md"},{"change":"modified","content_sha256":"08b6570d0bbe834cdd7a6971bb37d5ed1c32994a20584ec1b46e193f8b5ffde5","path":"tests/test_medium_preprocess.py"},{"change":"added","content_sha256":"6ddb7dbfe2e60b9a856ff1addf6e3d27d6272249539856e1ce248aa6054f7269","path":"tests/test_pose_batch_path.py"},{"change":"modified","content_sha256":"c38e9965707ffa3ae0ced51c419d783bff654eddc8f2c1e8594121a0901d7c2d","path":"tests/test_rtsp_source_adapter.py"}],"plan_hash":"bd16ef972513f24e96eccb749a9e44364050ac504363ffdbdd9ef261fc586ec1","profile":"ltb1021","push":{"branch":"task/runtime-lifecycle-refactor","mode":"upstream","remote":"origin","remote_identity":"sha256:3a9ce41319a2ebea","upstream_head":"c869cc4acf6e7bc26062af0235528e381f1ba3f0","upstream_ref":"origin/task/runtime-lifecycle-refactor"},"repository":{"baseline_head":"c869cc4acf6e7bc26062af0235528e381f1ba3f0","branch":"task/runtime-lifecycle-refactor","worktree":"/workspace/contextra-analytics-runtime-lifecycle-refactor"},"schema_version":"ltb1021.codex_hub.git_closeout_manifest.v1","session_alias":"work-session","verify":["Source/image67 targeted tests each PASS; git diff --check PASS; four-channel below target; original channels restored."]}