LLM Evaluation with Unit Tests in Mellea.
Message
Example
TestData
validate_examples
validate_examples(cls, v)
TestBasedEval
parts
parts(self) -> list[Component | CBlock]
format_for_llm
format_for_llm(self) -> TemplateRepresentation
set_judge_context
set_judge_context(self, input_text: str, prediction: str, targets_for_input: list[str])
from_json_file
from_json_file(cls, filepath: str) -> list['TestBasedEval']