Quick Start
Evaluator Types
| Evaluator | Measures |
|---|---|
AccuracyEvaluator | Correctness vs expected |
CriteriaEvaluator | Multiple custom criteria |
PerformanceEvaluator | Speed and efficiency |
Judge | LLM-as-judge scoring |
Best Practices
Test on diverse examples
Test on diverse examples
Use varied test cases to get accurate evaluation.
Iterate based on scores
Iterate based on scores
Low scores indicate where to improve prompts or tools.

