phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-03 19:57:35 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

chore: use openai_chat_completion for llm as a judge scoring

Browse source
This commit is contained in:
Matthew Farrellee 2025-10-01 09:35:20 -04:00
parent ca47d90926
commit 8f9fa0f1fd
1 changed files with 8 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

15
llama_stack/providers/inline/scoring/llm_as_judge/scoring_fn/llm_as_judge_scoring_fn.py
View file

@ -6,7 +6,7 @@
import re
from typing import Any
from llama_stack.apis.inference import Inference, UserMessage
from llama_stack.apis.inference import Inference
from llama_stack.apis.scoring import ScoringResultRow
from llama_stack.apis.scoring_functions import ScoringFnParams
from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
@ -55,15 +55,16 @@ class LlmAsJudgeScoringFn(RegisteredBaseScoringFn):
generated_answer=generated_answer,
)
judge_response = await self.inference_api.chat_completion(
model_id=fn_def.params.judge_model,
judge_response = await self.inference_api.openai_chat_completion(
model=fn_def.params.judge_model,
messages=[
UserMessage(
content=judge_input_msg,
),
{
"role": "user",
"content": judge_input_msg,
}
],
)
content = judge_response.completion_message.content
content = judge_response.choices[0].message.content
rating_regexes = fn_def.params.judge_score_regexes
judge_rating = None