Hi Team ,

It should be the same no? You just evaluate the outputs

Yep , I also thought the same but I do get an error. Also when I run the ragas example of llama index I get an error when the following statement is intercepted :
from ragas.metrics import (
faithfulness,
answer_relevancy,
context_precision,
context_recall,
)
on the lines of :
“cannot import name pydanticoutputparser from langchain_core.output_parsers”

Seems like an issue with langchain and ragas then 😅

@Logan M do we have any example to refer to integrate with ragas from LlamaIndex

I do not -- I just know whatever is on the ragas documentation.

I can’t make that thing work , were you able to run that example successfully , and help would be really helpful !