Optional
agentA list of tools available to the agent, for TrajectoryEvalChain.
Optional
chainOptional
criteria?: undefined | (CriteriaLike & (Record<string, string> | Criteria))The criteria to use for the evaluator.
Optional
distanceThe distance metric to use for comparing the embeddings.
Optional
embedding?: undefined | EmbeddingsInterfaceThe embedding objects to vectorize the outputs.
Optional
llm?: undefined | BaseLanguageModelInterface<any, BaseLanguageModelCallOptions> & BaseLanguageModel<any, BaseLanguageModelCallOptions>The criteria to use for the evaluator.
The language model to use for the evaluator.
The configuration for the evaluator.
const evalConfig = {
evaluators: [Criteria("helpfulness")],
};
@example
```ts
const evalConfig = {
evaluators: [
Criteria({
"isCompliant": "Does the submission comply with the requirements of XYZ"
})
],
const evalConfig = {
evaluators: [{
evaluatorType: "criteria",
criteria: "helpfulness"
formatEvaluatorInputs: ...
}]
};
const evalConfig = {
evaluators: [{
evaluatorType: "criteria",
criteria: { "isCompliant": "Does the submission comply with the requirements of XYZ" },
formatEvaluatorInputs: ...
}]
Configuration to load a "CriteriaEvalChain" evaluator, which prompts an LLM to determine whether the model's prediction complies with the provided criteria.