francescofan

@dosu I finished your step 1, and there is already a directory '.\\ragas\\cache\\zh' here. As to step 2 generate_with_langchain_docs doesn't take a cache_dir parameter, so where can I find settings...

@dosu hey, the JSON files were created by generator.save(evolutions=[simple, reasoning, multi_context,conditional]), how could it be invaild,guy?

@qpwo2468 do you have any solutions?

@dosu 1, Configuration: the bug occurs when no my_run_config is set. 2. Implementation: where is "properly differentiates between generating questions and ground truths for conditional scenarios" 3. Ragas version:0.1.20

@dosu I have tried 1/2/3/4/6 method. But the problem is still there. And how to get more Logging.

@dosu I am in a windows environment. So how to set up export __RAGAS_DEBUG_TRACKING=True

@dosu Have you ever met the problem that you submit the exact 2 same code and the first is ok but the second is always encoutered APIConnectionError(Connection error.)?? just like...

@dosu Hello. I have tried all methods you provide but in vain. The problem was still there. But after I finished fixing another problem ' https://github.com/explodinggradients/ragas/issues/1432', this problem is fixed,too....

@dosu To resolve this, you can try setting the 'strictness' value to 1 when configuring the 'answer_relevancy' metric. This adjustment should align with the current support for 'n = 1'...

@dosu I have modified py file ''Lib\site-packages\ragas\metrics\_answer_relevance.py" like following: ```python class AnswerRelevancy(MetricWithLLM, MetricWithEmbeddings, SingleTurnMetric): ... question_generation: Prompt = field(default_factory=lambda: QUESTION_GEN) # strictness: int = 3 strictness: int = 1 ```...