open-thought/reasoning-gym
1
0
Fork 0
mirror of https://github.com/open-thought/reasoning-gym.git synced 2026-04-22 16:49:06 +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

16
reasoning_gym/graphs/quantum_lock.py
View file

@ -169,21 +169,15 @@ Buttons:
The function awards 1.0 for a correct answer and less otherwise.
"""
if answer == None:
if not isinstance(answer, str):
return 0.0
# Get correct solution from metadata
correct_solution = entry["metadata"].get("solution_path", [])
# Normalize both answers
def normalize_seq(seq):
"""Handle both string and list inputs by converting to string first"""
# Convert sequence to string representation if it's a list
input_str = "".join(seq) if isinstance(seq, list) else str(seq or "")
return [c.upper() for c in re.findall(r"[A-C]", input_str.upper())]
def normalize_seq(seq: str) -> list[str]:
return [c.upper() for c in re.findall(r"[A-C]", seq.upper())]
user_sequence = normalize_seq(answer)
target_sequence = normalize_seq("".join(correct_solution))
target_sequence = normalize_seq(entry["answer"])
# Exact sequence match required
if user_sequence == target_sequence:
@ -196,7 +190,7 @@ Buttons:
return 1.0 # Different answer, but qually correct
return 0.5 # Alternative scoring - you're correct, but not optimal
return 0.1
return 0.0
def simulate_sequence(self, metadata: dict, sequence: list[str]) -> int:
"""Simulate button presses to verify solutions"""