phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-11 11:50:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

input query optional input for braintrust scorer

Browse source
This commit is contained in:
Xi Yan 2024-10-14 21:17:16 -07:00
parent 7b5895003a
commit 3c29108b6e
4 changed files with 8 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

5
llama_stack/providers/impls/meta_reference/evals/tasks/run_scoring_task.py
View file

@ -33,11 +33,15 @@ class RunScoringTask(BaseTask):
for x in dataset:
expected_answer = x.data["expected_answer"]
generated_answer = x.data["generated_answer"]
input_query = None
if "input_query" in x.data:
input_query = x.data["input_query"]
scorer_inputs.append(
ScorerInputSample(
expected_answer=expected_answer,
generated_answer=generated_answer,
input_query=input_query,
)
)
@ -74,7 +78,6 @@ class RunScoringTask(BaseTask):
)
scorer_results = scorer.score(postprocessed)
cprint(scorer_results, "magenta")
eval_result = scorer.aggregate_results(scorer_results)
return eval_result