Function createUserFrictionEvaluator

createUserFrictionEvaluator<
    RecordType extends
        Record<string, unknown> = UserFrictionEvaluationRecord,
>(
    args: UserFrictionEvaluatorArgs<RecordType>,
): ClassificationEvaluator<RecordType>
Creates a user friction evaluator function.

This function returns an evaluator that detects expressed user friction with an assistant's preceding behavior, such as corrections, retries, frustration, and challenges to unrequested or unexplained actions.
Type Parameters
- RecordType extends Record<string, unknown> = UserFrictionEvaluationRecord
Parameters
- args: UserFrictionEvaluatorArgs<RecordType>
  The arguments for creating the user friction evaluator.
  - Optionalchoices?: ClassificationChoicesMap
  - OptionalinputMapping?: ObjectMapping<RecordType>
    
    The mapping of the input to evaluate to the shape that the evaluator expects
  - model: LanguageModel
  - Optionalname?: string
  - OptionaloptimizationDirection?: OptimizationDirection
  - OptionalpromptTemplate?: PromptTemplate
  - Optionaltelemetry?: TelemetryConfig
Returns ClassificationEvaluator<RecordType>
An evaluator function that takes a UserFrictionEvaluationRecord and returns a classification result indicating whether the latest user message expresses friction or no friction.
Example
```
const evaluator = createUserFrictionEvaluator({ model: openai("gpt-4o-mini") });
const result = await evaluator.evaluate({
  conversation: "User: Show recent orders.\nAssistant: Here are last month's orders.",
  userMessage: "No, I asked for this week.",
});
console.log(result.label); // "friction"
```
- Defined in llm/createUserFrictionEvaluator.ts:57