PairwiseStringEvaluator#

class langchain.evaluation.schema.PairwiseStringEvaluator[source]#

Compare the output of two models (or two outputs of the same model).

Attributes

requires_input

Whether this evaluator requires an input string.

requires_reference

Whether this evaluator requires a reference label.

Methods

__init__()

aevaluate_string_pairs(*, prediction, ...[, ...])

Asynchronously evaluate the output string pairs.

evaluate_string_pairs(*, prediction, ...[, ...])

Evaluate the output string pairs.

__init__()#
async aevaluate_string_pairs(*, prediction: str, prediction_b: str, reference: str | None = None, input: str | None = None, **kwargs: Any) dict[source]#

Asynchronously evaluate the output string pairs.

Parameters:
  • prediction (str) – The output string from the first model.

  • prediction_b (str) – The output string from the second model.

  • reference (Optional[str], optional) – The expected output / reference string.

  • input (Optional[str], optional) – The input string.

  • kwargs (Any) – Additional keyword arguments, such as callbacks and optional reference strings.

Returns:

A dictionary containing the preference, scores, and/or other information.

Return type:

dict

evaluate_string_pairs(*, prediction: str, prediction_b: str, reference: str | None = None, input: str | None = None, **kwargs: Any) dict[source]#

Evaluate the output string pairs.

Parameters:
  • prediction (str) – The output string from the first model.

  • prediction_b (str) – The output string from the second model.

  • reference (Optional[str], optional) – The expected output / reference string.

  • input (Optional[str], optional) – The input string.

  • kwargs (Any) – Additional keyword arguments, such as callbacks and optional reference strings.

Returns:

A dictionary containing the preference, scores, and/or other information.

Return type:

dict