SingleKeyEvalConfig Class — langchain Architecture

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

Class python

Entity Profile

Dependency Diagram

graph TD
  5528a09f_3682_a8b0_958b_cd9a898b6e3c["SingleKeyEvalConfig"]
  3f727cf4_38d7_5e57_1492_800220e6e9f9["EvalConfig"]
  5528a09f_3682_a8b0_958b_cd9a898b6e3c -->|extends| 3f727cf4_38d7_5e57_1492_800220e6e9f9
  ba7f9415_541a_d99f_0400_a688a10212b0["config.py"]
  5528a09f_3682_a8b0_958b_cd9a898b6e3c -->|defined in| ba7f9415_541a_d99f_0400_a688a10212b0
  682cf20d_5338_c7c4_4ecb_fe694cc89fa5["get_kwargs()"]
  5528a09f_3682_a8b0_958b_cd9a898b6e3c -->|method| 682cf20d_5338_c7c4_4ecb_fe694cc89fa5

Relationship Graph

Source Code

libs/langchain/langchain_classic/smith/evaluation/config.py lines 57–77

class SingleKeyEvalConfig(EvalConfig):
    """Configuration for a run evaluator that only requires a single key."""

    reference_key: str | None = None
    """The key in the dataset run to use as the reference string.
    If not provided, we will attempt to infer automatically."""
    prediction_key: str | None = None
    """The key from the traced run's outputs dictionary to use to
    represent the prediction. If not provided, it will be inferred
    automatically."""
    input_key: str | None = None
    """The key from the traced run's inputs dictionary to use to represent the
    input. If not provided, it will be inferred automatically."""

    @override
    def get_kwargs(self) -> dict[str, Any]:
        kwargs = super().get_kwargs()
        # Filer out the keys that are not needed for the evaluator.
        for key in ["reference_key", "prediction_key", "input_key"]:
            kwargs.pop(key, None)
        return kwargs