nat.plugins.eval.rag_evaluator.register#

Attributes#

logger

Classes#

RagasMetricConfig

RAGAS metrics configuration.

RagasEvaluatorConfig

Evaluation using RAGAS metrics.

Functions#

register_ragas_evaluator(config, builder)

Module Contents#

logger#
class RagasMetricConfig(/, **data: Any)#

Bases: pydantic.BaseModel

RAGAS metrics configuration.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

skip: bool = False#
kwargs: dict | None = None#
class RagasEvaluatorConfig#

Bases: nat.data_models.evaluator.EvaluatorLLMConfig

Evaluation using RAGAS metrics.

metric: str | dict[str, RagasMetricConfig] = None#
input_obj_field: str | None = None#
classmethod validate_metric(values)#

Ensures metric is either a string or a single-item dictionary.

property metric_name: str#

Returns the single metric name.

property metric_config: RagasMetricConfig#

Returns the metric configuration (or a default if only a string is provided).

async register_ragas_evaluator(
config: RagasEvaluatorConfig,
builder: nat.builder.builder.EvalBuilder,
)#