(model_answer, answer)
| 19 | return INVALID_ANS |
| 20 | |
| 21 | def is_correct(model_answer, answer): |
| 22 | gt_answer = extract_answer_from_output(answer) |
| 23 | assert gt_answer != INVALID_ANS |
| 24 | return model_answer == gt_answer |
| 25 | |
| 26 | def clean_answer(model_pred): |
| 27 | model_pred = model_pred.lower() |
nothing calls this directly
no test coverage detected