open-thought/reasoning-gym
1
0
Fork 0
mirror of https://github.com/open-thought/reasoning-gym.git synced 2026-04-19 12:58:07 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Minor question template & score_answer improvements (#261)

Browse source 
* 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:
Andreas Köpf 2025-03-04 21:55:09 +01:00 • committed by GitHub
parent 061282e373
commit 5d7fbac0ad
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
106 changed files with 403 additions and 507 deletions
Download patch file Download diff file Expand all files Collapse all files

10
reasoning_gym/arithmetic/decimal_arithmetic.py
View file

@ -178,7 +178,7 @@ class DecimalArithmeticDataset(ProceduralDataset):
+ problem_str
)
return {"question": problem_str, "answer": answer, "metadata": {}}
return {"question": problem_str, "answer": str(answer), "metadata": {}}
def score_answer(self, answer: Optional[str], entry: dict[str, Any]) -> float:
"""
@ -189,12 +189,12 @@ class DecimalArithmeticDataset(ProceduralDataset):
Returns:
float: 1.0 if the user's answer is within tolerance; otherwise, 0.01.
"""
if answer is None:
if not isinstance(answer, str):
return 0.0
try:
user_ans: Decimal = Decimal(answer)
correct_ans: Decimal = entry["answer"]
correct_ans: Decimal = Decimal(entry["answer"])
# Determine tolerance based on the desired precision.
precision: int = self.config.max_num_decimal_places
@ -202,9 +202,9 @@ class DecimalArithmeticDataset(ProceduralDataset):
if abs(user_ans - correct_ans) <= tol:
return 1.0
except Exception:
return 0.01
pass
return 0.01
return 0.0
# Register the dataset with the factory.