(self, query, reference_answer, generated_answer)
| 69 | return score, passing, judge |
| 70 | |
| 71 | def evaluate(self, query, reference_answer, generated_answer): |
| 72 | system_prompt = self.system_template.format( |
| 73 | query=query, |
| 74 | reference_answer=reference_answer, |
| 75 | generated_answer=generated_answer |
| 76 | ) |
| 77 | score, passing, judge = self.llm_eval(query, reference_answer, generated_answer) |
| 78 | result = dict( |
| 79 | score=score, |
| 80 | passing=passing, |
| 81 | judge=judge |
| 82 | ) |
| 83 | return result |
| 84 | |
| 85 | |
| 86 |