Skip to content

wanghai-tech / hikoon-ACR

Go to a project
Toggle navigation
Toggle navigation pinning
Commit 06794812 067948120de164a61de09d3141d3c106299c68ff by cnb.bofCdSsphPA
Options

Attach runnable command templates to the extraction plan 

Constraint: The Phase-1 PostgreSQL plan needed to become immediately actionable without pretending the workers already exist
Rejected: Keep the plan as ordering-only metadata | It still leaves the next session to reconstruct command wiring by hand
Confidence: high
Scope-risk: narrow
Directive: Keep future worker implementations compatible with the env-var contract emitted by the planner report
Tested: /usr/local/miniconda3/bin/python scripts/plan_phase1_extraction_jobs_live.py --dsn 'postgres://d2:d2pass@127.0.0.1:5432/d2' --schema acr_test --job-status pending --output data/pgvector_eval/music20/phase1_extraction_plan_report.json; /usr/local/miniconda3/bin/python -m py_compile scripts/plan_phase1_extraction_jobs_live.py; git diff --check -- acr-engine/scripts/plan_phase1_extraction_jobs_live.py acr-engine/data/pgvector_eval/music20/phase1_extraction_plan_report.json docs/model-feature-registry-bootstrap.md docs/postgres_db_schema_samples.md docs/session-handoff.md docs/CHANGELOG.md
Not-tested: Real worker binaries at workers/run_chromaprint_job.py and workers/run_embedding_job.py do not exist yet
1 parent f13caa3e
Inline Side-by-side
Showing 6 changed files with 102 additions and 5 deletions
...@@ -49,6 +49,10 @@ ...@@ -49,6 +49,10 @@
49 "run feature extraction for chromaprint v1", 49 "run feature extraction for chromaprint v1",
50 "write to audio_fingerprint", 50 "write to audio_fingerprint",
51 "target scope: reference_set:phase1_hot_reference_v1" 51 "target scope: reference_set:phase1_hot_reference_v1"
52 ],
53 "command_suggestions": [
54 "EXTRACTION_JOB_ID=1 FEATURE_SET_ID=2 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test OUTPUT_TARGET=audio_fingerprint \\\npython workers/run_chromaprint_job.py",
55 "EXTRACTION_JOB_ID=1 FEATURE_SET_ID=2 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test \\\npython workers/mark_job_status.py --status running"
52 ] 56 ]
53 }, 57 },
54 { 58 {
...@@ -90,6 +94,10 @@ ...@@ -90,6 +94,10 @@
90 "run feature extraction for mert v1-95m", 94 "run feature extraction for mert v1-95m",
91 "write to audio_embedding + audio_embedding_vector_768", 95 "write to audio_embedding + audio_embedding_vector_768",
92 "target scope: reference_set:phase1_hot_reference_v1" 96 "target scope: reference_set:phase1_hot_reference_v1"
97 ],
98 "command_suggestions": [
99 "EXTRACTION_JOB_ID=2 FEATURE_SET_ID=3 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=mert MODEL_VERSION=v1-95m VECTOR_TABLE=audio_embedding_vector_768 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py",
100 "EXTRACTION_JOB_ID=2 FEATURE_SET_ID=3 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test \\\npython workers/mark_job_status.py --status running"
93 ] 101 ]
94 }, 102 },
95 { 103 {
...@@ -131,6 +139,10 @@ ...@@ -131,6 +139,10 @@
131 "run feature extraction for mert v1-95m", 139 "run feature extraction for mert v1-95m",
132 "write to audio_embedding + audio_embedding_vector_768", 140 "write to audio_embedding + audio_embedding_vector_768",
133 "target scope: reference_set:phase1_hot_reference_v1" 141 "target scope: reference_set:phase1_hot_reference_v1"
142 ],
143 "command_suggestions": [
144 "EXTRACTION_JOB_ID=3 FEATURE_SET_ID=4 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=mert MODEL_VERSION=v1-95m VECTOR_TABLE=audio_embedding_vector_768 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py",
145 "EXTRACTION_JOB_ID=3 FEATURE_SET_ID=4 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test \\\npython workers/mark_job_status.py --status running"
134 ] 146 ]
135 }, 147 },
136 { 148 {
...@@ -172,6 +184,10 @@ ...@@ -172,6 +184,10 @@
172 "run feature extraction for muq large-msd-iter", 184 "run feature extraction for muq large-msd-iter",
173 "write to audio_embedding + audio_embedding_vector_768", 185 "write to audio_embedding + audio_embedding_vector_768",
174 "target scope: reference_set:phase1_hot_reference_v1" 186 "target scope: reference_set:phase1_hot_reference_v1"
187 ],
188 "command_suggestions": [
189 "EXTRACTION_JOB_ID=4 FEATURE_SET_ID=5 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=muq MODEL_VERSION=large-msd-iter VECTOR_TABLE=audio_embedding_vector_768 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py",
190 "EXTRACTION_JOB_ID=4 FEATURE_SET_ID=5 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test \\\npython workers/mark_job_status.py --status running"
175 ] 191 ]
176 }, 192 },
177 { 193 {
...@@ -213,6 +229,10 @@ ...@@ -213,6 +229,10 @@
213 "run feature extraction for ecapa acr-baseline-v1", 229 "run feature extraction for ecapa acr-baseline-v1",
214 "write to audio_embedding + audio_embedding_vector_192", 230 "write to audio_embedding + audio_embedding_vector_192",
215 "target scope: reference_set:phase1_hot_reference_v1" 231 "target scope: reference_set:phase1_hot_reference_v1"
232 ],
233 "command_suggestions": [
234 "EXTRACTION_JOB_ID=5 FEATURE_SET_ID=6 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=ecapa MODEL_VERSION=acr-baseline-v1 VECTOR_TABLE=audio_embedding_vector_192 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py",
235 "EXTRACTION_JOB_ID=5 FEATURE_SET_ID=6 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test \\\npython workers/mark_job_status.py --status running"
216 ] 236 ]
217 } 237 }
218 ], 238 ],
...@@ -257,6 +277,10 @@ ...@@ -257,6 +277,10 @@
257 "run feature extraction for chromaprint v1", 277 "run feature extraction for chromaprint v1",
258 "write to audio_fingerprint", 278 "write to audio_fingerprint",
259 "target scope: reference_set:phase1_hot_reference_v1" 279 "target scope: reference_set:phase1_hot_reference_v1"
280 ],
281 "command_suggestions": [
282 "EXTRACTION_JOB_ID=1 FEATURE_SET_ID=2 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test OUTPUT_TARGET=audio_fingerprint \\\npython workers/run_chromaprint_job.py",
283 "EXTRACTION_JOB_ID=1 FEATURE_SET_ID=2 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test \\\npython workers/mark_job_status.py --status running"
260 ] 284 ]
261 } 285 }
262 ], 286 ],
...@@ -300,6 +324,10 @@ ...@@ -300,6 +324,10 @@
300 "run feature extraction for mert v1-95m", 324 "run feature extraction for mert v1-95m",
301 "write to audio_embedding + audio_embedding_vector_768", 325 "write to audio_embedding + audio_embedding_vector_768",
302 "target scope: reference_set:phase1_hot_reference_v1" 326 "target scope: reference_set:phase1_hot_reference_v1"
327 ],
328 "command_suggestions": [
329 "EXTRACTION_JOB_ID=2 FEATURE_SET_ID=3 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=mert MODEL_VERSION=v1-95m VECTOR_TABLE=audio_embedding_vector_768 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py",
330 "EXTRACTION_JOB_ID=2 FEATURE_SET_ID=3 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test \\\npython workers/mark_job_status.py --status running"
303 ] 331 ]
304 }, 332 },
305 { 333 {
...@@ -341,6 +369,10 @@ ...@@ -341,6 +369,10 @@
341 "run feature extraction for mert v1-95m", 369 "run feature extraction for mert v1-95m",
342 "write to audio_embedding + audio_embedding_vector_768", 370 "write to audio_embedding + audio_embedding_vector_768",
343 "target scope: reference_set:phase1_hot_reference_v1" 371 "target scope: reference_set:phase1_hot_reference_v1"
372 ],
373 "command_suggestions": [
374 "EXTRACTION_JOB_ID=3 FEATURE_SET_ID=4 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=mert MODEL_VERSION=v1-95m VECTOR_TABLE=audio_embedding_vector_768 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py",
375 "EXTRACTION_JOB_ID=3 FEATURE_SET_ID=4 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test \\\npython workers/mark_job_status.py --status running"
344 ] 376 ]
345 }, 377 },
346 { 378 {
...@@ -382,6 +414,10 @@ ...@@ -382,6 +414,10 @@
382 "run feature extraction for muq large-msd-iter", 414 "run feature extraction for muq large-msd-iter",
383 "write to audio_embedding + audio_embedding_vector_768", 415 "write to audio_embedding + audio_embedding_vector_768",
384 "target scope: reference_set:phase1_hot_reference_v1" 416 "target scope: reference_set:phase1_hot_reference_v1"
417 ],
418 "command_suggestions": [
419 "EXTRACTION_JOB_ID=4 FEATURE_SET_ID=5 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=muq MODEL_VERSION=large-msd-iter VECTOR_TABLE=audio_embedding_vector_768 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py",
420 "EXTRACTION_JOB_ID=4 FEATURE_SET_ID=5 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test \\\npython workers/mark_job_status.py --status running"
385 ] 421 ]
386 }, 422 },
387 { 423 {
...@@ -423,6 +459,10 @@ ...@@ -423,6 +459,10 @@
423 "run feature extraction for ecapa acr-baseline-v1", 459 "run feature extraction for ecapa acr-baseline-v1",
424 "write to audio_embedding + audio_embedding_vector_192", 460 "write to audio_embedding + audio_embedding_vector_192",
425 "target scope: reference_set:phase1_hot_reference_v1" 461 "target scope: reference_set:phase1_hot_reference_v1"
462 ],
463 "command_suggestions": [
464 "EXTRACTION_JOB_ID=5 FEATURE_SET_ID=6 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=ecapa MODEL_VERSION=acr-baseline-v1 VECTOR_TABLE=audio_embedding_vector_192 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py",
465 "EXTRACTION_JOB_ID=5 FEATURE_SET_ID=6 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test \\\npython workers/mark_job_status.py --status running"
426 ] 466 ]
427 } 467 }
428 ] 468 ]
...@@ -436,7 +476,8 @@ ...@@ -436,7 +476,8 @@
436 "feature_name": "fingerprint_asset", 476 "feature_name": "fingerprint_asset",
437 "window_sec": 5.0, 477 "window_sec": 5.0,
438 "hop_sec": 2.5, 478 "hop_sec": 2.5,
439 "physical_target": "audio_fingerprint" 479 "physical_target": "audio_fingerprint",
480 "primary_command": "EXTRACTION_JOB_ID=1 FEATURE_SET_ID=2 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test OUTPUT_TARGET=audio_fingerprint \\\npython workers/run_chromaprint_job.py"
440 }, 481 },
441 { 482 {
442 "order": 2, 483 "order": 2,
...@@ -446,7 +487,8 @@ ...@@ -446,7 +487,8 @@
446 "feature_name": "semantic_embedding", 487 "feature_name": "semantic_embedding",
447 "window_sec": 5.0, 488 "window_sec": 5.0,
448 "hop_sec": 2.5, 489 "hop_sec": 2.5,
449 "physical_target": "audio_embedding" 490 "physical_target": "audio_embedding",
491 "primary_command": "EXTRACTION_JOB_ID=2 FEATURE_SET_ID=3 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=mert MODEL_VERSION=v1-95m VECTOR_TABLE=audio_embedding_vector_768 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py"
450 }, 492 },
451 { 493 {
452 "order": 3, 494 "order": 3,
...@@ -456,7 +498,8 @@ ...@@ -456,7 +498,8 @@
456 "feature_name": "semantic_embedding", 498 "feature_name": "semantic_embedding",
457 "window_sec": 10.0, 499 "window_sec": 10.0,
458 "hop_sec": 5.0, 500 "hop_sec": 5.0,
459 "physical_target": "audio_embedding" 501 "physical_target": "audio_embedding",
502 "primary_command": "EXTRACTION_JOB_ID=3 FEATURE_SET_ID=4 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=mert MODEL_VERSION=v1-95m VECTOR_TABLE=audio_embedding_vector_768 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py"
460 }, 503 },
461 { 504 {
462 "order": 4, 505 "order": 4,
...@@ -466,7 +509,8 @@ ...@@ -466,7 +509,8 @@
466 "feature_name": "semantic_embedding", 509 "feature_name": "semantic_embedding",
467 "window_sec": 5.0, 510 "window_sec": 5.0,
468 "hop_sec": 2.5, 511 "hop_sec": 2.5,
469 "physical_target": "audio_embedding" 512 "physical_target": "audio_embedding",
513 "primary_command": "EXTRACTION_JOB_ID=4 FEATURE_SET_ID=5 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=muq MODEL_VERSION=large-msd-iter VECTOR_TABLE=audio_embedding_vector_768 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py"
470 }, 514 },
471 { 515 {
472 "order": 5, 516 "order": 5,
...@@ -476,7 +520,8 @@ ...@@ -476,7 +520,8 @@
476 "feature_name": "semantic_embedding", 520 "feature_name": "semantic_embedding",
477 "window_sec": 5.0, 521 "window_sec": 5.0,
478 "hop_sec": 2.5, 522 "hop_sec": 2.5,
479 "physical_target": "audio_embedding" 523 "physical_target": "audio_embedding",
524 "primary_command": "EXTRACTION_JOB_ID=5 FEATURE_SET_ID=6 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=ecapa MODEL_VERSION=acr-baseline-v1 VECTOR_TABLE=audio_embedding_vector_192 OUTPUT_TARGET=audio_embedding \\\npython workers/run_embedding_job.py"
480 } 525 }
481 ] 526 ]
482 } 527 }
...\ No newline at end of file ...\ No newline at end of file
......
...@@ -25,6 +25,26 @@ def parse_target_scope(target_scope: str) -> dict[str, Any]: ...@@ -25,6 +25,26 @@ def parse_target_scope(target_scope: str) -> dict[str, Any]:
25 return {'scope_type': 'unknown', 'scope_value': target_scope} 25 return {'scope_type': 'unknown', 'scope_value': target_scope}
26 26
27 27
28 def build_command_suggestions(job: dict[str, Any], schema: str) -> list[str]:
29 base_env = f"EXTRACTION_JOB_ID={job['extraction_job_id']} FEATURE_SET_ID={job['feature_set_id']} TARGET_SCOPE='{job['target_scope']}' PG_SCHEMA={schema}"
30 commands = []
31 if job['lane'] == 'exact':
32 commands.append(
33 base_env
34 + " OUTPUT_TARGET=audio_fingerprint \\\npython workers/run_chromaprint_job.py"
35 )
36 else:
37 commands.append(
38 base_env
39 + f" MODEL_NAME={job['model_name']} MODEL_VERSION={job['model_version']} VECTOR_TABLE={job['vector_table']} OUTPUT_TARGET={job['physical_target']} \\\npython workers/run_embedding_job.py"
40 )
41 commands.append(
42 base_env
43 + " \\\npython workers/mark_job_status.py --status running"
44 )
45 return commands
46
47
28 def main() -> None: 48 def main() -> None:
29 ap = argparse.ArgumentParser() 49 ap = argparse.ArgumentParser()
30 ap.add_argument('--dsn', required=True) 50 ap.add_argument('--dsn', required=True)
...@@ -112,6 +132,7 @@ def main() -> None: ...@@ -112,6 +132,7 @@ def main() -> None:
112 f"target scope: {row[2]}", 132 f"target scope: {row[2]}",
113 ], 133 ],
114 } 134 }
135 item['command_suggestions'] = build_command_suggestions(item, args.schema)
115 jobs.append(item) 136 jobs.append(item)
116 by_lane.setdefault(lane, []).append(item) 137 by_lane.setdefault(lane, []).append(item)
117 138
...@@ -139,6 +160,7 @@ def main() -> None: ...@@ -139,6 +160,7 @@ def main() -> None:
139 'window_sec': job['window_sec'], 160 'window_sec': job['window_sec'],
140 'hop_sec': job['hop_sec'], 161 'hop_sec': job['hop_sec'],
141 'physical_target': job['physical_target'], 162 'physical_target': job['physical_target'],
163 'primary_command': job['command_suggestions'][0],
142 } 164 }
143 for idx, job in enumerate(jobs) 165 for idx, job in enumerate(jobs)
144 ], 166 ],
......
1 ## 2026-06-04 1 ## 2026-06-04
2 2
3 - 更新 `plan_phase1_extraction_jobs_live.py` 与 `phase1_extraction_plan_report.json`，把 Phase-1 execution plan 从“仅有排序计划”推进到“附带 `command_suggestions / primary_command` 的可复制执行命令模板”。
3 - 新增 `acr-engine/scripts/plan_phase1_extraction_jobs_live.py` 与 `acr-engine/data/pgvector_eval/music20/phase1_extraction_plan_report.json`，支持从 PostgreSQL 的 `feature_extraction_job` 真实读取 pending jobs，并联表生成按 lane / priority 排序的 Phase-1 execution plan。 4 - 新增 `acr-engine/scripts/plan_phase1_extraction_jobs_live.py` 与 `acr-engine/data/pgvector_eval/music20/phase1_extraction_plan_report.json`，支持从 PostgreSQL 的 `feature_extraction_job` 真实读取 pending jobs，并联表生成按 lane / priority 排序的 Phase-1 execution plan。
4 - 新增 `acr-engine/scripts/bootstrap_phase1_extraction_jobs_live.py` 与 `acr-engine/data/pgvector_eval/music20/phase1_extraction_jobs_report.json`，把 Phase-1 的 `feature_extraction_job` 初始化做成可直接连 PostgreSQL 的 live 脚本，并已在 `acr_test` schema 真实创建 5 条 pending jobs。 5 - 新增 `acr-engine/scripts/bootstrap_phase1_extraction_jobs_live.py` 与 `acr-engine/data/pgvector_eval/music20/phase1_extraction_jobs_report.json`，把 Phase-1 的 `feature_extraction_job` 初始化做成可直接连 PostgreSQL 的 live 脚本，并已在 `acr_test` schema 真实创建 5 条 pending jobs。
5 - 补充 `phase1_registry_bootstrap_idempotency_report.json` 与文档说明，验证 `bootstrap_phase1_model_registry_live.py` 在 `acr_test` schema 上连续执行两次后表计数保持稳定，证明 Phase-1 registry bootstrap 具备可重复执行的幂等性。 6 - 补充 `phase1_registry_bootstrap_idempotency_report.json` 与文档说明，验证 `bootstrap_phase1_model_registry_live.py` 在 `acr_test` schema 上连续执行两次后表计数保持稳定，证明 Phase-1 registry bootstrap 具备可重复执行的幂等性。
......
...@@ -397,3 +397,25 @@ cd /workspace/acr-engine ...@@ -397,3 +397,25 @@ cd /workspace/acr-engine
397 结论： 397 结论：
398 398
399 > 现在 PostgreSQL 里已经不仅能描述“有哪些 job”，还可以直接生成**按执行顺序排好的抽特征计划**。 399 > 现在 PostgreSQL 里已经不仅能描述“有哪些 job”，还可以直接生成**按执行顺序排好的抽特征计划**。
400
401 ### 10.3 ready-to-run command suggestions（已补齐）
402
403 本轮又进一步把 planner 升级为：**每条 job 都生成 command suggestion**。
404
405 示例：
406
407 #### exact lane
408
409 ```bash
410 EXTRACTION_JOB_ID=1 FEATURE_SET_ID=2 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test OUTPUT_TARGET=audio_fingerprint \
411 python workers/run_chromaprint_job.py
412 ```
413
414 #### semantic lane
415
416 ```bash
417 EXTRACTION_JOB_ID=2 FEATURE_SET_ID=3 TARGET_SCOPE='reference_set:phase1_hot_reference_v1' PG_SCHEMA=acr_test MODEL_NAME=mert MODEL_VERSION=v1-95m VECTOR_TABLE=audio_embedding_vector_768 OUTPUT_TARGET=audio_embedding \
418 python workers/run_embedding_job.py
419 ```
420
421 这意味着下个 session 不需要先手工拼环境变量和 job 绑定关系，而可以直接从 planner 报告里复制命令模板。
......
...@@ -430,6 +430,12 @@ flowchart LR ...@@ -430,6 +430,12 @@ flowchart LR
430 对应 live 报告： 430 对应 live 报告：
431 - `acr-engine/data/pgvector_eval/music20/phase1_extraction_plan_report.json` 431 - `acr-engine/data/pgvector_eval/music20/phase1_extraction_plan_report.json`
432 432
433 本轮补充后，plan 里还会真实给出：
434 - `command_suggestions`
435 - `primary_command`
436
437 也就是从 PostgreSQL 的 pending jobs 已经可以直接走到“可复制的执行命令模板”。
438
433 ### 路线 1：继续做 PostgreSQL 工程化 439 ### 路线 1：继续做 PostgreSQL 工程化
434 440
435 1. 把 `live_pgvector_music20_eval.py` 泛化成： 441 1. 把 `live_pgvector_music20_eval.py` 泛化成：
......
...@@ -185,6 +185,7 @@ sed -n '1,320p' acr-engine/sql/acr_pg_schema_v2.sql ...@@ -185,6 +185,7 @@ sed -n '1,320p' acr-engine/sql/acr_pg_schema_v2.sql
185 - Phase-1 registry bootstrap 已有幂等性证据：同 schema 连续执行两次后，`model_registry=5 / feature_set_registry=6 / reference_set_registry=2` 保持不变 185 - Phase-1 registry bootstrap 已有幂等性证据：同 schema 连续执行两次后，`model_registry=5 / feature_set_registry=6 / reference_set_registry=2` 保持不变
186 - PostgreSQL `acr_test` schema 上已真实创建 5 条 `feature_extraction_job`，后续 MERT / MuQ 接入可直接从 pending jobs 启动 186 - PostgreSQL `acr_test` schema 上已真实创建 5 条 `feature_extraction_job`，后续 MERT / MuQ 接入可直接从 pending jobs 启动
187 - PostgreSQL `acr_test` schema 上已真实生成 Phase-1 extraction execution plan，当前顺序是 `chromaprint -> mert -> mert-long -> muq -> ecapa` 187 - PostgreSQL `acr_test` schema 上已真实生成 Phase-1 extraction execution plan，当前顺序是 `chromaprint -> mert -> mert-long -> muq -> ecapa`
188 - extraction plan 报告里已包含 `command_suggestions / primary_command`，下次可直接从 plan 抄 worker 命令模板
188 189
189 ### 未验证 / 仍是缺口 190 ### 未验证 / 仍是缺口
190 - **未实际跑 MERT / MuQ encoder-only 特征抽取** 191 - **未实际跑 MERT / MuQ encoder-only 特征抽取**
......