Function read_problems

eval/utils/data.py:11–12 · view source on GitHub ↗

(evalset_file: str = HUMAN_EVAL)

Source from the content-addressed store, hash-verified

9
10
11	def read_problems(evalset_file: str = HUMAN_EVAL) -> Dict[str, Dict]:
12	return {task["task_id"]: task for task in stream_jsonl(evalset_file)}
13
14
15	def stream_jsonl(filename: str) -> Iterable[Dict]:

evaluate_human_eval.pyFile · 0.90

evaluate_functional_correctnessFunction · 0.85

stream_jsonlFunction · 0.85

no test coverage detected