Issue with Qdrant Vector Store

[hodgesz]

I can successfully create indexes in Qdrant with LlamaIndex. However, when I try to query these indexes I run into problems.

Here is my code:

`
embeddings = HuggingFaceInstructEmbeddings(model_name="hkunlp/instructor-xl",
model_kwargs={"device": "mps"})

client = qdrant_client.QdrantClient(url='http://localhost:6333', prefer_grpc=True)
vector_store = Qdrant(client=client,
collection_name=qdrant_index,
embeddings=embeddings,
content_payload_key='text')

llm = ChatOpenAI(model_name=gpt3_5_turbo_model_name,
temperature=openai_temperature)

index = VectorStoreIndex.from_vector_store(vector_store=vector_store)
retriever = VectorIndexRetriever(
index=index,
vector_store_query_mode="default",
similarity_top_k=6
)

`

The code fails trying to retrieve the index from Qdrant vector store.

AttributeError: 'Qdrant' object has no attribute 'stores_text'
Traceback:
File "/opt/homebrew/Caskroom/miniforge/base/envs/langchain_dev/lib/python3.11/site-packages/streamlit/runtime/scriptrunner/script_runner.py", line 565, in _run_script
exec(code, module.dict)
File "/Users/hodgesz/PycharmProjects/WorkForge/workforge_ai_workshop/code/llamaindex_chat_app.py", line 131, in
index = VectorStoreIndex.from_vector_store(vector_store=vector_store)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/homebrew/Caskroom/miniforge/base/envs/langchain_dev/lib/python3.11/site-packages/llama_index/indices/vector_store/base.py", line 65, in from_vector_store
if not vector_store.stores_text:
^^^^^^^^^^^^^^^^^^^^^^^^

The only similar issue in the forum was here (#6686) but it appears to be different and based on not already successfully creating the index in the vector store.

Any idea?

[dosubot[bot]]

🤖

Hello,

The error you're encountering is likely due to trying to access the 'stores_text' attribute from a 'Qdrant' object, while this attribute is actually defined in the 'QdrantVectorStore' class.

In your code, you're creating an instance of 'QdrantVectorStore' and passing it to the 'from_vector_store' method of 'VectorStoreIndex'. However, the error message suggests that the 'from_vector_store' method is trying to access the 'stores_text' attribute from a 'Qdrant' object, not from a 'QdrantVectorStore' object.

Here's the correct flow:

from llama_index import SimpleDirectoryReader
from llama_index import VectorStoreIndex

documents = SimpleDirectoryReader('test_library', recursive=True).load_data()

# initial creation of index
index = VectorStoreIndex.from_documents(documents)

# after creation, you can connect/reinitialize like so
loaded_index = VectorStoreIndex.from_vector_store(index.vector_store)

For persisting the index, you can use the persist method of the storage_context attribute of your VectorStoreIndex instance. Here's how you can do it:

# Persist the index
index.storage_context.persist('saved_index.json')

This will save the index to a JSON file named 'saved_index.json'. If you want to save it to a different directory, you can provide the directory path as the argument to the persist method.

Please ensure that you're accessing the 'stores_text' attribute from an instance of the 'QdrantVectorStore' class, not from a 'Qdrant' object.

Let me know if this helps or if you have any other questions.

Best,
Dosu

Sources

• [Bug]: Cannot serialize VectorStoreIndex object
• [Bug]: from_vector_store function not working?
• llama_index/vector_stores/qdrant.py

---

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.