langchain chroma: directly use embedding without re-embedding again

[RickyGunawan09]

Hi guys,
Is there a way to use the embedding that has been stored in Chromadb into a retriever without using from_text or from_document to repeat the embedding process so that it can increase the speed of the langchain process?

so the program flow is, I will save all the embeddings into chromadb. The data is then taken only some data using "get" and then it will be made into a RetrievalQA chain.

When the data from "get" is obtained and used as a retrieval, the data is re-embedded either using from_text or from_document.
Is there a way to directly use embeddings that have been saved in Chromadb?

Thank You

[dosubot[bot]]

🤖

Hi,

Based on the current version of LangChain (v0.0.287) and the provided context, it appears that LangChain does not currently support the direct use of embeddings from Chromadb without re-embedding. The embedding process is typically done using from_text or from_document methods.

However, I understand your concern about the efficiency of the process. It would indeed be more efficient if the embeddings could be directly used from Chromadb without the need for re-embedding.

Unfortunately, I couldn't find any existing feature or method in the LangChain framework that allows for this. The current implementation seems to require the re-embedding process.

I recommend keeping an eye on the repository for any updates or new features that might address this issue in the future. If you have any other questions or need further clarification, feel free to ask.

Best Regards,
Dosu

Sources

• libs/langchain/langchain/embeddings/init.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.