[Question]: Using DocArrayHnswVectorStore to persist

[ricoyudog]

Question Validation

• I have searched both the documentation and discord for an answer.

Question

Hi teams, I am tring follow the save local document to save the index which created by
`

from llama_index import GPTVectorStoreIndex, SimpleDirectoryReader, Document
from llama_index.vector_stores import DocArrayHnswVectorStore
vector_store = DocArrayHnswVectorStore(work_dir=)
storage_context = StorageContext.from_defaults(vector_store=vector_store)
index = GPTVectorStoreIndex.from_documents(documents, storage_context=storage_context,show_progress=True)

index.storage_context.persist(persist_dir=)

, But when I go through the local file, there is no vector_stores file. So it cant not be load by

 load_index_from_storage

The director only have:
docstore.json
graph_store.json
index_store.json

May I have help from you.
Thank you very much!

[dosubot[bot]]

🤖

Hi there,

From your code, it seems like you're trying to persist the index created by GPTVectorStoreIndex using DocArrayHnswVectorStore. However, you're not seeing the expected vector_stores file in your local directory.

Based on the similar issues I found in the LlamaIndex repository, it seems like when persisting to a remote vector store, everything is actually persisted on the remote vector store itself. Therefore, you might not see a vector_stores file in your local directory.

Here's how you can reconnect to your remote vector store:

vector_store = DocArrayHnswVectorStore(work_dir=your_work_dir)
storage_context = StorageContext.from_defaults(vector_store=vector_store)
index = GPTVectorStoreIndex([], storage_context=storage_context)

Please replace your_work_dir with the directory where your vector store is located.

Also, it's important to note that you don't need to call persist() or load_index_from_storage() when using a vector store integration like DocArrayHnswVectorStore.

If you're still having trouble, could you please provide more information about your setup? Specifically, it would be helpful to know more about your work_dir and persist_dir parameters, and whether you're using a local or remote vector store.

I hope this helps! Let me know if you have any other questions.

Best,
Dosu

Sources

• [Bug]: Cannot load StorageContext from persisted storage context (Chroma)
• storage_context.persist() does not creates vector_index.json file
• llama_index/vector_stores/simple.py
• tests/indices/vector_store/test_simple.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.