ScorerData
The result from a single scorer for one example.
Each ScoringResult contains a list of ScorerData – one per scorer
that was run. Inspect value for the scorer’s output.
Attributes
PropType
name?str
Scorer name (e.g. "faithfulness").
Type
strvalue?Optional[Union[str, float]]
The scorer result. Binary scorers use "Yes" or "No"; categorical scorers use their selected category.
Type
Optional[Union[str, float]]Default
Nonescore_type?Optional[str]
Result type returned by the scorer.
Type
Optional[str]Default
Noneminimum_score_range?float
Lower bound of the scoring scale.
Type
floatDefault
0maximum_score_range?float
Upper bound of the scoring scale.
Type
floatDefault
1evaluation_model?Optional[str]
The LLM used to produce the score.
Type
Optional[str]Default
Noneerror?Optional[str]
Error message if scoring failed.
Type
Optional[str]Default
Noneadditional_metadata?Dict[str, Any]
Extra metadata from the scorer.
Type
Dict[str, Any]Default
field(default_factory=dict)id?Optional[str]
Unique identifier for this result.
Type
Optional[str]Default
Nonesuccess?Optional[bool]
Outcome of a client-side pass condition for the row this scorer result belongs to. None when no pass condition was evaluated.
Type
Optional[bool]Default
Noneto_dict()
Serialize to a dictionary, omitting None fields.
def to_dict() -> Dict[str, Any]:Returns
Dict[str, Any]