explodinggradients / ragas

Evaluation framework for your Retrieval Augmented Generation (RAG) pipelines
https://docs.ragas.io
Apache License 2.0
6.79k stars 669 forks source link

why?can you help me? #537

Closed liuruit closed 4 months ago

liuruit commented 8 months ago

为什么我选择tongyi模型,它返回不出来最后评估的结果,只有openai才可以吗 下面是我在colab上面跑的代码出的错误

evaluating with [faithfulness]
  0%|          | 0/1 [00:00<?, ?it/s]ERROR:dashscope:Unsupported atom data type: <class 'langchain_core.prompts.chat.ChatPromptTemplate'>
  0%|          | 0/1 [00:00<?, ?it/s]
---------------------------------------------------------------------------
UnsupportedDataType                       Traceback (most recent call last)
[<ipython-input-81-5fa2ac208107>](https://localhost:8080/#) in <cell line: 3>()
      1 from ragas.llama_index import evaluate
      2 
----> 3 result = evaluate(query_engine, metrics, eval_questions, eval_answers)

25 frames
[/usr/local/lib/python3.10/dist-packages/dashscope/io/input_output.py](https://localhost:8080/#) in resolve_input(input, is_encode_binary, custom_type_resolver)
    109         return custom_type_resolver[type(input)](input)
    110     else:
--> 111         raise UnsupportedDataType('Unsupported atom data type: %s' %
    112                                   type(input))

UnsupportedDataType: Unsupported atom data type: <class 'langchain_core.prompts.chat.ChatPromptTemplate'>
jjmachan commented 8 months ago

which version of Ragas are you using @liuruit ? llamaindex is not support for v0.1 at the moment sadly - you can track #557 for updates

will get this fixed for you as fast as we can Liuruit :)