RunEvaluator#

class langsmith.evaluation.evaluator.RunEvaluator[source]#

Evaluator interface class.

Methods

__init__()

aevaluate_run(run[, example, evaluator_run_id])

Evaluate an example asynchronously.

evaluate_run(run[, example, evaluator_run_id])

Evaluate an example.
__init__()#
async aevaluate_run(
run: Run,
example: Example | None = None,
evaluator_run_id: UUID | None = None,
) → EvaluationResult | EvaluationResults[source]#

Evaluate an example asynchronously.

Parameters:

  • run (Run)

  • example (Example | None)

  • evaluator_run_id (UUID | None)

Return type:

EvaluationResult | EvaluationResults

abstractmethod evaluate_run(
run: Run,
example: Example | None = None,
evaluator_run_id: UUID | None = None,
) → EvaluationResult | EvaluationResults[source]#

Evaluate an example.

Parameters:

  • run (Run)

  • example (Example | None)

  • evaluator_run_id (UUID | None)

Return type:

EvaluationResult | EvaluationResults