ScoreString Class — langchain Architecture

Architecture documentation for the ScoreString class in config.py from the langchain codebase.

Class python

Entity Profile

Dependency Diagram

graph TD
  47293a9b_cea7_3396_d2ae_5ae17921131c["ScoreString"]
  5528a09f_3682_a8b0_958b_cd9a898b6e3c["SingleKeyEvalConfig"]
  47293a9b_cea7_3396_d2ae_5ae17921131c -->|extends| 5528a09f_3682_a8b0_958b_cd9a898b6e3c
  ba7f9415_541a_d99f_0400_a688a10212b0["config.py"]
  47293a9b_cea7_3396_d2ae_5ae17921131c -->|defined in| ba7f9415_541a_d99f_0400_a688a10212b0

Relationship Graph

Source Code

libs/langchain/langchain_classic/smith/evaluation/config.py lines 247–268

    class ScoreString(SingleKeyEvalConfig):
        """Configuration for a score string evaluator.

        This is like the criteria evaluator but it is configured by
        default to return a score on the scale from 1-10.

        It is recommended to normalize these scores
        by setting `normalize_by` to 10.

        Attributes:
            criteria: The criteria to evaluate.
            llm: The language model to use for the evaluation chain.
            normalize_by: If you want to normalize the score, the denominator to use.
                If not provided, the score will be between 1 and 10.
            prompt: The prompt template to use for evaluation.
        """

        evaluator_type: EvaluatorType = EvaluatorType.SCORE_STRING
        criteria: CRITERIA_TYPE | None = None
        llm: BaseLanguageModel | None = None
        normalize_by: float | None = None
        prompt: BasePromptTemplate | None = None