include ranges rather than sampled values in difficulty metadata dicts (#387)

* update difficulty metadata for logic datasets

* update difficulty metadata for graph datasets

* update difficulty metadata for geometry datasets

* update difficulty metadata for games datasets

* update difficulty metadata for cognition datasets

* update difficulty metadata for arithmetic datasets

* update difficulty metadata for arc datasets

* update difficulty metadata for algorithmic datasets

* update difficulty metadata for algebra datasets

* use tuples

* update tests

* update tests
This commit is contained in:
Oliver Stanley 2025-03-20 09:27:03 +00:00 committed by GitHub
parent b69c35818a
commit 7475a20700
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
80 changed files with 304 additions and 126 deletions

View file

@ -117,7 +117,14 @@ class RectangleCountDataset(ProceduralDataset):
return {
"question": QUESTION_TEMPLATE.format(puzzle=puzzle),
"answer": str(answer),
"metadata": {"puzzle": puzzle, "solution": answer, "difficulty": {"max_rectangles": target}},
"metadata": {
"puzzle": puzzle,
"solution": answer,
"num_rectangles": target,
"difficulty": {
"max_rectangles": self.config.max_rectangles,
},
},
}
def score_answer(self, answer: Optional[str], entry: dict[str, Any]) -> float: