Description
Subject: Issue with weaviate.Client in LangChain: ValueError: client should be an instance of weaviate.Client, got <class ‘weaviate.client.WeaviateClient’>
Message:
Hi everyone,
I’ve been working on integrating Weaviate with LangChain to create a retriever for semantic search. However, I’m encountering an issue when trying to use the Weaviate class from LangChain to create a retriever. Here’s the error I’m getting:
ValueError: client should be an instance of weaviate.Client, got <class 'weaviate.client.WeaviateClient'>
Context
I’m using the following code to connect to Weaviate and create a retriever:
import os
from weaviate import connect_to_local
from weaviate.config import AdditionalConfig, Timeout
from langchain.vectorstores import Weaviate
from langchain.embeddings import OpenAIEmbeddings
# Connect to Weaviate
client = connect_to_local(
headers={"X-OpenAI-Api-Key": os.getenv("OPENAI_APIKEY")},
additional_config=AdditionalConfig(
timeout=Timeout(init=10, query=60, insert=120)
)
)
# Initialize embeddings
embeddings = OpenAIEmbeddings(openai_api_key=os.getenv("OPENAI_APIKEY"))
# Create a Weaviate retriever
retriever = Weaviate(
client=client,
index_name="EmailQuery",
text_key=["subject", "beneficiary_email", "manager_response", "topic"],
embedding=embeddings
).as_retriever()
Has anyone else encountered this issue when using LangChain with the latest Weaviate client?
Any additional Information
langchain-community== 0.2.11
langchain-weaviate==0.0.3
weaviate-client== 4.8.1
