WHy am i getting this error- {
"detail": "Error creating vector store: Error code: 400 - {'error': {'message': "'$.input' is invalid. Please check the API reference: https://platform.openai.com/docs/api-reference.\", 'type': 'invalid_request_error', 'param': None, 'code': None}}"
}

Hey everyone, has anyone used S3DirectoryReader? I am only getting encrypted files if i use it. Please help

Hey Logan. What would be the optimal way for ingesting multiple documents into nodes that are stored in S3 bucket? Currently i am using S3.list_objects and reading the files then converting them into text. Also, i am manually creating nodes for every page of every pdf. Can you help me understand whether this approach is good for multiple hundred pdfs lets say?

I want to make a quick RAG using create-llama but it seems to not index the local folder that I provide to it during the setup phase. Am i missing out on something?

Also, isnt there a better way to quickly know what are the changes for importing modules? It gets annoying after some time :/

@kapa.ai i want to use llmsherpa with S3 objects V2. The flow should be reading the documents from S3 buckets and then parsing it using LLMsherpa. help me with the code for it

@kapa.ai how do i specify my choice of llm in the following
vector_store = PineconeVectorStore(pinecone_index=pinecone_index, add_sparse_vector=True)
pinecone_vectorstore_index = VectorStoreIndex.from_vector_store(vector_store = vector_store)

print('case id', request.case_id)

retriever = pinecone_vectorstore_index.as_retriever(
# similarity_top_k = 5,
# metadata_filters={"case_id": request.case_id}
vector_store_kwargs={"filter": {"case_id": request.case_id}}
)
response_synthesizer = get_response_synthesizer(response_mode="refine")

query_engine = RetrieverQueryEngine(
retriever=retriever,
response_synthesizer=response_synthesizer
)
query_engine.update_prompts(
{"response_synthesizer:text_qa_template": custom_prompt}
)
response = query_engine.query(request.query)

return {"response": response.response}