inclusion->subsetof

llama_stack/providers/impls/meta_reference/scoring/scorer/subset_of_scorer.py

@@ -15,13 +15,13 @@ from llama_stack.providers.impls.meta_reference.scoring.scorer.common import (
 )
 
 
-class InclusionScorer(BaseScorer):
+class SubsetOfScorer(BaseScorer):
     """
     A scorer that assigns a score of 1.0 if the expected string is included in the generated string, and 0.0 otherwise.
     """
 
     scoring_function_def = ScoringFunctionDef(
-        identifier="inclusion",
+        identifier="subset_of",
         description="Returns 1.0 if the expected is included in generated, 0.0 otherwise.",
         parameters=[],
         return_type=NumberType(),

llama_stack/providers/impls/meta_reference/scoring/scoring.py

@@ -17,15 +17,15 @@ from llama_stack.providers.impls.meta_reference.scoring.scorer.equality_scorer i
     EqualityScorer,
 )
 
-from llama_stack.providers.impls.meta_reference.scoring.scorer.inclusion_scorer import (
-    InclusionScorer,
+from llama_stack.providers.impls.meta_reference.scoring.scorer.subset_of_scorer import (
+    SubsetOfScorer,
 )
 
 from .config import MetaReferenceScoringConfig
 
 SUPPORTED_SCORERS = [
     EqualityScorer,
-    InclusionScorer,
+    SubsetOfScorer,
 ]
 
 SCORER_REGISTRY = {x.scoring_function_def.identifier: x for x in SUPPORTED_SCORERS}

llama_stack/providers/tests/eval/test_eval.py

@@ -65,7 +65,7 @@ async def test_eval(eval_settings):
             model="Llama3.1-8B-Instruct",
             sampling_params=SamplingParams(),
         ),
-        scoring_functions=["inclusion"],
+        scoring_functions=["subset_of"],
     )
     assert response.job_id == "0"
     job_status = await eval_impl.job_status(response.job_id)
@@ -76,4 +76,4 @@ async def test_eval(eval_settings):
 
     assert eval_response is not None
     assert len(eval_response.generations) == 5
-    assert "inclusion" in eval_response.scores
+    assert "subset_of" in eval_response.scores