mirror of
https://github.com/open-thought/reasoning-gym.git
synced 2026-04-23 16:55:05 +00:00
Minor question template & score_answer improvements (#261)
* math prompt improvements * ignore brackets in complex_arithmetic results * improve additional instruction in prompt of polynomial_equations * more strict tests for score_answer in polynomial_equations * simplify special reward handling * fix test_intermediate_integration * fix sokoban dataset * add common dataset score_answer consistency test
This commit is contained in:
parent
bf24999bb0
commit
b2904ccab9
106 changed files with 403 additions and 507 deletions
|
|
@ -401,16 +401,14 @@ class CircuitLogicDataset(ProceduralDataset):
|
|||
}
|
||||
|
||||
def score_answer(self, answer: Optional[str], entry: dict[str, Any]) -> float:
|
||||
if answer is None or len(answer) == 0:
|
||||
return 0.0
|
||||
if isinstance(answer, str) and len(answer) > 0:
|
||||
oracle_answer = entry["answer"]
|
||||
if oracle_answer == answer:
|
||||
return 1.0
|
||||
elif oracle_answer == answer.strip():
|
||||
return len(oracle_answer) / len(answer)
|
||||
|
||||
oracle_answer = entry["answer"]
|
||||
if oracle_answer == answer:
|
||||
return 1.0
|
||||
elif oracle_answer == answer.strip():
|
||||
return len(oracle_answer) / len(answer)
|
||||
|
||||
return 0.01
|
||||
return 0.0
|
||||
|
||||
|
||||
register_dataset("circuit_logic", CircuitLogicDataset, CircuitLogicConfig)
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue