AttributeError: 'VectorStoreRetriever' object has no attribute 'embeddings'

[thainv02]

Checked other resources

• I added a very descriptive title to this issue.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.
• I am sure that this is a bug in LangChain rather than my code.
• The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).

Example Code

  from langchain.schema.document import Document
  from langchain_core.vectorstores import VectorStoreRetriever
  from langchain.retrievers import WikipediaRetriever
  from typing import List
  class RerankRetriever(VectorStoreRetriever):
      vectorstore: VectorStoreRetriever
      def get_relevant_documents(self, query: str) -> List[Document]:
          docs = self.vectorstore.get_relevant_documents(query=query)
          candidates = [doc.page_content for doc in docs]
          queries = [query]*len(candidates)
          features = tokenizer_rerank(queries, candidates,  padding=True, truncation=True, return_tensors="pt").to(device)
          with torch.no_grad():
              scores = model_rerank(**features).logits
              values, indices = torch.sum(scores, dim=1).sort()
              # relevant_docs = docs[indices[0]]
          return [docs[indices[0]],docs[indices[1]]]
  class RerankWikiRetriever(VectorStoreRetriever):
      vectorstore: WikipediaRetriever
      def get_relevant_documents(self, query: str) -> List[Document]:
          docs = self.vectorstore.get_relevant_documents(query=query)
          candidates = [doc.page_content for doc in docs]
          queries = [query]*len(candidates)
          features = tokenizer_rerank(queries, candidates,  padding=True, truncation=True, return_tensors="pt").to(device)
          with torch.no_grad():
              scores = model_rerank(**features).logits
              values, indices = torch.sum(scores, dim=1).sort()
              # relevant_docs = docs[indices[0]]
          return [docs[indices[0]],docs[indices[1]]]
    
    
    
      from langchain.retrievers import WikipediaRetriever
      from langchain.vectorstores import Qdrant
      from langchain.llms import HuggingFacePipeline
      from qdrant_client import QdrantClient
      from langchain.prompts import PromptTemplate
      from langchain.embeddings import HuggingFaceInferenceAPIEmbeddings
      from langchain.chains import RetrievalQA,MultiRetrievalQAChain
      from langchain.llms import VLLM
      from langchain.llms import HuggingFaceHub
      
      class LLMServe:
          def __init__(self) -> None:
            self.embeddings = self.load_embeddings()
            self.current_source = "wiki"
            self.retriever = self.load_retriever(retriever_name = self.current_source,embeddings=self.embeddings)
            self.pipe = self.load_model_pipeline(max_new_tokens=300)
            self.prompt = self.load_prompt_template()
            self.rag_pipeline = self.load_rag_pipeline(llm=self.pipe,
                                                  retriever=self.retriever,
                                                  prompt=self.prompt)
          def load_embeddings(self):
            embeddings = HuggingFaceInferenceAPIEmbeddings(
                model_name=EMBEDDINGS_MODEL_NAME,
                api_key = HUGGINGFACE_API_KEY,
                #model_kwargs = {'device': "auto"}
            )
            return embeddings
      
  
  
      def load_retriever(self,retriever_name,embeddings):
        retriever=None
        if retriever_name == "wiki":
          retriever = RerankWikiRetriever(vectorstore = WikipediaRetriever(lang="vi",
                                         doc_content_chars_max=800,top_k_results=15))
        else:
          client = QdrantClient(
              url=QDRANT_URL,api_key=QDRANT_API_KEY, prefer_grpc=False
          )
          db = Qdrant(client=client,
                      embeddings=embeddings,
                      collection_name=QDRANT_COLLECTION_NAME)
  
          retriever = RerankRetriever(vectorstore = db.as_retriever(search_kwargs={"k":15}))
  
        return retriever
  
      def load_model_pipeline(self,max_new_tokens=100):
        llm = VLLM(
            model=GENERATE_MODEL_NAME,
            trust_remote_code=True,  # mandatory for hf models
            max_new_tokens=max_new_tokens,
              # temperature=1.0,
              # top_k=50,
              # top_p=0.9,
            top_k=10,
            top_p=0.95,
            temperature=0.4,
            dtype="half",
            vllm_kwargs={"quantization": "awq"}
        )
        return llm
  
      def load_prompt_template(self):
        # query_template = "Bạn là một trợ lý của trường Đại học Nguyễn Tất Thành. Hãy trả lời câu hỏi sau dựa trên ngữ cảnh, nếu ngữ cảnh không cung cấp câu trả lời hoặc không chắc chắn hãy trả lời 'Tôi không biết thông tin này, tuy nhiên đoạn thông tin dưới phần tham khảo có thể có câu trả lời cho bạn!' đừng cố tạo ra câu trả lời không có trong ngữ cảnh.\nNgữ cảnh: {context} \nCâu hỏi: {question}\nTrả lời: "
        # query_template = "Tham khảo ngữ cảnh:{context}\n\n### Câu hỏi:{question}\n\n### Trả lời:"
        query_template = "Bạn là một chatbot thông minh trả lời câu hỏi dựa trên ngữ cảnh (context).\n\n### Context:{context} \n\n### Human: {question}\n\n### Assistant:"
        prompt = PromptTemplate(template=query_template,
                          input_variables= ["context","question"])
        return prompt
  
      def load_rag_pipeline(self,llm,retriever,prompt):
        rag_pipeline = RetrievalQA.from_chain_type(
        llm=llm, chain_type='stuff',
        retriever=retriever,
        chain_type_kwargs={
        "prompt": prompt
        },
        return_source_documents=True)
        return rag_pipeline
  
      def rag(self,source):
        if source == self.current_source:
          return self.rag_pipeline
        else:
          self.retriever = self.load_retriever(retriever_name=source,embeddings=self.embeddings)
          self.rag_pipeline = self.load_rag_pipeline(llm=self.pipe,
                                        retriever=self.retriever,
                                        prompt=self.prompt)
          self.current_source = source
          return self.rag_pipeline
          
          app = LLMServe()
          
          from typing import Union
          from fastapi.middleware.cors import CORSMiddleware
          from fastapi.responses import JSONResponse
          from fastapi.encoders import jsonable_encoder
          from fastapi import FastAPI
          origins = ["*"]
          app_api = FastAPI()
          app_api.add_middleware(
              CORSMiddleware,
              allow_origins=origins,
              allow_credentials=True,
              allow_methods=["*"],
              allow_headers=["*"],
          )
          
          @app_api.get("/")
          def read_root():
              return "API RAG"
          
          @app_api.get("/rag/{source}")
          async def read_item(source: str, q: str | None = None):
              if q:
                  data = app.rag(source=source)(q)
                  sources = []
                  for docs in data["source_documents"]:
                      sources.append(docs.to_json()["kwargs"])
                  res = {
                      "result" : data["result"],
                      "source_documents":sources
                  }
                  return JSONResponse(content=jsonable_encoder(res))
              return None
              
                            
          import nest_asyncio
          from pyngrok import ngrok
          import uvicorn
          ngrok.set_auth_token(NGROK_TOKEN)
          ngrok_tunnel = ngrok.connect(8000,domain=NGROK_STATIC_DOMAIN)
          print('Public URL:', ngrok_tunnel.public_url)
          nest_asyncio.apply()
          uvicorn.run(app_api, port=8000)

Error Message and Stack Trace (if applicable)

AttributeError: 'VectorStoreRetriever' object has no attribute 'embeddings'

Description

I encountered this error while trying to create a server on Google Colab to run my application. It successfully generated a public URL, but when I made a request to that URL, I encountered the error mentioned above. This is my first time asking questions like this, so if I missed any details, please let me know. Thank you very much.

System Info

Public URL: https://natural-glider-willingly.ngrok-free.app/
INFO: Started server process [218]
INFO: Waiting for application startup.
INFO: Application startup complete.
INFO: Uvicorn running on http://127.0.0.1:8000/ (Press CTRL+C to quit)
INFO: 42.114.173.251:0 - "GET /rag/HUMG_db_v1?q=%C4%91i%E1%BB%81u%20ki%E1%BB%87n%20x%C3%A9t%20h%E1%BB%8Dc%20b%E1%BB%95ng HTTP/1.1" 500 Internal Server Error
:40: LangChainDeprecationWarning: The class Qdrant was deprecated in LangChain 0.0.37 and will be removed in 1.0. An updated version of the class exists in the :class:~langchain-qdrant package and should be used instead. To use it run pip install -U :class:~langchain-qdrant and import as from :class:~langchain_qdrant import Qdrant``.
db = Qdrant(client=client,
:23: LangChainDeprecationWarning: The method Chain.__call__ was deprecated in langchain 0.1.0 and will be removed in 1.0. Use :meth:`~invoke` instead.
data = app.rag(source=source)(q)
ERROR: Exception in ASGI application
Traceback (most recent call last):
File "/usr/local/lib/python3.10/dist-packages/uvicorn/protocols/http/httptools_impl.py", line 401, in run_asgi
result = await app( # type: ignore[func-returns-value]
File "/usr/local/lib/python3.10/dist-packages/uvicorn/middleware/proxy_headers.py", line 60, in call
return await self.app(scope, receive, send)
File "/usr/local/lib/python3.10/dist-packages/fastapi/applications.py", line 1054, in call
await super().call(scope, receive, send)
File "/usr/local/lib/python3.10/dist-packages/starlette/applications.py", line 113, in call
await self.middleware_stack(scope, receive, send)
File "/usr/local/lib/python3.10/dist-packages/starlette/middleware/errors.py", line 187, in call
raise exc
File "/usr/local/lib/python3.10/dist-packages/starlette/middleware/errors.py", line 165, in call
await self.app(scope, receive, _send)
File "/usr/local/lib/python3.10/dist-packages/starlette/middleware/cors.py", line 85, in call
await self.app(scope, receive, send)
File "/usr/local/lib/python3.10/dist-packages/starlette/middleware/exceptions.py", line 62, in call
await wrap_app_handling_exceptions(self.app, conn)(scope, receive, send)
File "/usr/local/lib/python3.10/dist-packages/starlette/_exception_handler.py", line 62, in wrapped_app
raise exc
File "/usr/local/lib/python3.10/dist-packages/starlette/_exception_handler.py", line 51, in wrapped_app
await app(scope, receive, sender)
File "/usr/local/lib/python3.10/dist-packages/starlette/routing.py", line 715, in call
await self.middleware_stack(scope, receive, send)
File "/usr/local/lib/python3.10/dist-packages/starlette/routing.py", line 735, in app
await route.handle(scope, receive, send)
File "/usr/local/lib/python3.10/dist-packages/starlette/routing.py", line 288, in handle
await self.app(scope, receive, send)
File "/usr/local/lib/python3.10/dist-packages/starlette/routing.py", line 76, in app
await wrap_app_handling_exceptions(app, request)(scope, receive, send)
File "/usr/local/lib/python3.10/dist-packages/starlette/_exception_handler.py", line 62, in wrapped_app
raise exc
File "/usr/local/lib/python3.10/dist-packages/starlette/_exception_handler.py", line 51, in wrapped_app
await app(scope, receive, sender)
File "/usr/local/lib/python3.10/dist-packages/starlette/routing.py", line 73, in app
response = await f(request)
File "/usr/local/lib/python3.10/dist-packages/fastapi/routing.py", line 301, in app
raw_response = await run_endpoint_function(
File "/usr/local/lib/python3.10/dist-packages/fastapi/routing.py", line 212, in run_endpoint_function
return await dependant.call(**values)
File "", line 23, in read_item
data = app.rag(source=source)(q)
File "/usr/local/lib/python3.10/dist-packages/langchain_core/_api/deprecation.py", line 182, in warning_emitting_wrapper
return wrapped(*args, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/langchain/chains/base.py", line 389, in call
return self.invoke(
File "/usr/local/lib/python3.10/dist-packages/langchain/chains/base.py", line 170, in invoke
raise e
File "/usr/local/lib/python3.10/dist-packages/langchain/chains/base.py", line 160, in invoke
self._call(inputs, run_manager=run_manager)
File "/usr/local/lib/python3.10/dist-packages/langchain/chains/retrieval_qa/base.py", line 151, in _call
docs = self._get_docs(question, run_manager=_run_manager)
File "/usr/local/lib/python3.10/dist-packages/langchain/chains/retrieval_qa/base.py", line 271, in _get_docs
return self.retriever.invoke(
File "/usr/local/lib/python3.10/dist-packages/langchain_core/retrievers.py", line 227, in invoke
**self._get_ls_params(**kwargs),
File "/usr/local/lib/python3.10/dist-packages/langchain_core/vectorstores/base.py", line 1063, in _get_ls_params
if self.vectorstore.embeddings:
File "/usr/local/lib/python3.10/dist-packages/pydantic/main.py", line 856, in getattr
raise AttributeError(f'{type(self).name!r} object has no attribute {item!r}')
AttributeError: 'VectorStoreRetriever' object has no attribute 'embeddings'

[thainv02]

help me

[kodychik]

Hi can I work on this issue?
Thanks

[thainv02]

Hi can I work on this issue? Thanks

Looking forward to receiving your help