MCPcopy
hub / github.com/aiming-lab/AutoResearchClaw / main

Function main

experiments/arc_bench/scripts/run_baseline.py:269–288  ·  view source on GitHub ↗
()

Source from the content-addressed store, hash-verified

267
268
269def main() -> int:
270 ap = argparse.ArgumentParser(description="Run an ARC-Bench baseline cell")
271 ap.add_argument("--framework", required=True,
272 choices=list(ADAPTERS.keys()))
273 ap.add_argument("--topic", help="topic id, e.g. T01")
274 ap.add_argument("--all", action="store_true")
275 ap.add_argument("--dry-run", action="store_true")
276 args = ap.parse_args()
277
278 if args.all:
279 topic_ids = [t["id"] for t in _load_topics()]
280 elif args.topic:
281 topic_ids = [args.topic]
282 else:
283 ap.error("--topic or --all required")
284
285 RESULTS_ROOT.mkdir(parents=True, exist_ok=True)
286 for tid in topic_ids:
287 run_one(args.framework, tid, dry_run=args.dry_run)
288 return 0
289
290
291if __name__ == "__main__":

Callers 1

run_baseline.pyFile · 0.70

Calls 2

_load_topicsFunction · 0.70
run_oneFunction · 0.70

Tested by

no test coverage detected