RetrievalQA.from_chain_type showing a validation error of not being able to instantiate abstract

Question

When I am trying to build a RetrievalQA.from_chain_type with my local llm and PineCone VectorDataBase. However, it is not able to create a retriever owing to the error of not being able to instantiate abstract class of BaseRetriever.

The error is a Validation Error as shown below:

ValidationError                           Traceback (most recent call last)
/Users/dhruv/Desktop/Machine_Learning/Projects/Medical_ChatBot_Application/preprocessing/Experiment.ipynb Cell 28 line 1
----> 1 qa=RetrievalQA.from_chain_type(
      2     llm=llm, 
      3     chain_type="stuff", 
      4     retriever=docsearch.as_retriever(search_kwargs={'k': 2}),
      5     return_source_documents=True, 
      6     chain_type_kwargs=chain_type_kwargs)

File ~/anaconda3/envs/mchatbot/lib/python3.9/site-packages/langchain/chains/retrieval_qa/base.py:95, in BaseRetrievalQA.from_chain_type(cls, llm, chain_type, chain_type_kwargs, **kwargs)
     91 _chain_type_kwargs = chain_type_kwargs or {}
     92 combine_documents_chain = load_qa_chain(
     93     llm, chain_type=chain_type, **_chain_type_kwargs
     94 )
---> 95 return cls(combine_documents_chain=combine_documents_chain, **kwargs)

File ~/anaconda3/envs/mchatbot/lib/python3.9/site-packages/langchain/load/serializable.py:74, in Serializable.__init__(self, **kwargs)
     73 def __init__(self, **kwargs: Any) -> None:
---> 74     super().__init__(**kwargs)
     75     self._lc_kwargs = kwargs

File ~/anaconda3/envs/mchatbot/lib/python3.9/site-packages/pydantic/main.py:341, in pydantic.main.BaseModel.__init__()

ValidationError: 1 validation error for RetrievalQA
retriever
  Can't instantiate abstract class BaseRetriever with abstract methods _aget_relevant_documents, _get_relevant_documents (type=type_error)

I tried to create my own class retriever and check the documentation of the PineCone API, however was not able to get the output chain as desired. My docsearch instance is basically a PineCone object and is of the type :

<langchain_pinecone.vectorstores.PineconeVectorStore at 0x147954250>

This is how I created my docsearch object :

from langchain_pinecone import PineconeVectorStore as PC

docsearch = PC.from_texts([t.page_content for t in text_chunks],
        embeddings,
        index_name = index_name)

Kindly let me know how to proceed ahead with this. My options are either this or using ChromaDB as a last resort. However, since I have memory constraints I would prefer to solve this via PineConeDB itself.

@MubashirAhmedSiddiqui yes it was. I mean not this particular code, but I have used the RetrievalQA.from_chain_type with a ChromaDB vector space and it worked fine. — Dhruv Pamneja
– Dhruv Pamneja, Commented Apr 17, 2024 at 1:54

j3ffyang · Accepted Answer · 2024-04-15 16:24:15Z

0

How is chain_type_kwargs defined? From API doc >

https://api.python.langchain.com/en/latest/chains/langchain.chains.retrieval_qa.base.BaseRetrievalQA.html#langchain.chains.retrieval_qa.base.BaseRetrievalQA

chain_type_kwargs (Optional[dict]) –

For example

qa_chain = RetrievalQA.from_chain_type(
        llm=llm,
        retriever=vectordb.as_retriever(),
        chain_type="stuff",
        chain_type_kwargs={"prompt": prompt},
        return_source_documents=True)

answered Apr 15, 2024 at 16:24

j3ffyang

2,52827 silver badges16 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Dhruv Pamneja Over a year ago

The chain_type_kwargs is defined as follows : Prompt = PromptTemplate(template=prompt_template,input_variables=["context", "question"]) chain_type_kwargs = {'prompt':Prompt} It is exactly the same as the example stated above.

Gt001 · Accepted Answer · 2024-07-07 09:40:06Z

0

Try upgrading Langchain and all other packages versions. Langchain version should be over 0.1.0 and after upgrading restart kerner and run again. This helped me, also make sure all the dependencies are compatible with each other.

answered Jul 7, 2024 at 9:40

Gt001

11 bronze badge

1 Comment

Community Over a year ago

Your answer could be improved with additional supporting information. Please edit to add further details, such as citations or documentation, so that others can confirm that your answer is correct. You can find more information on how to write good answers in the help center.

Collectives™ on Stack Overflow

RetrievalQA.from_chain_type showing a validation error of not being able to instantiate abstract

2 Answers 2

1 Comment

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related