mirror of
https://github.com/open-thought/reasoning-gym.git
synced 2026-04-19 12:58:07 +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
061282e373
commit
5d7fbac0ad
106 changed files with 403 additions and 507 deletions
|
|
@ -303,11 +303,11 @@ Reply as a JSON-parsable list of moves which result in any of the jugs being fil
|
|||
danswer = json.loads(answer)
|
||||
valid, _ = verify_solution(entry["metadata"]["puzzle"], danswer)
|
||||
if not valid:
|
||||
return 0.01
|
||||
return 0.01 # json parsable
|
||||
else:
|
||||
return 1.0 # Yay
|
||||
except Exception as e:
|
||||
return 0.01
|
||||
return 0.0
|
||||
|
||||
|
||||
register_dataset("jugs", JugsDataset, JugsConfig)
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue