Open Source Multi Modal Rag with chromadb using OpenCLIP embeddings

Поділитися
Вставка
  • Опубліковано 14 жов 2024

КОМЕНТАРІ • 8

  • @echohive
    @echohive  4 місяці тому

    Code files and coding walkthrough for this project available at my 1000x MasterClass: www.patreon.com/posts/1000x-dev-103326330
    Search 200+ echohive videos and code download links:www.echohive.live/
    Auto Streamer: www.autostreamer.live/
    Fastapi course: www.patreon.com/posts/learn-fastapi-26-95041684
    Chat with us on Discord: discord.gg/PPxTP3Cs3G
    Follow on twitter(X) : twitter.com/hive_echo

  • @micbab-vg2mu
    @micbab-vg2mu 4 місяці тому

    Great project - I will test it:)

    • @echohive
      @echohive  4 місяці тому

      Thank you 🙏 feel free to let me know what you think.

  • @tonywhite4476
    @tonywhite4476 4 місяці тому

    👏👏

    • @echohive
      @echohive  4 місяці тому

      Thank you 🙏

  • @tonywhite4476
    @tonywhite4476 4 місяці тому

    Is there a way to give it camera access so we can take a picture and load it as well?

    • @echohive
      @echohive  4 місяці тому +1

      Yeah that should be possible. Like taking a snapshot and storing it to the documents folder and rerunning the script. Or you can let is check the documents folder frequently during the script or before each query.

    • @tonywhite4476
      @tonywhite4476 4 місяці тому

      @@echohive I have a streamlit app that uses gpt-4o and has a camera feature that allows me to take a snapshot but you have to inquire about the image at once and it's not saved in memeory. This works better bc I can store multiple images before inquiring and they're saved to a database. I think checking frequently would be a little over-kill but works fine for your reasearch papers app where it's dynamically updated. I believe it works there bc papers are upload sporadically but with this, you actually know when data is laoded. So, if it had a UI I would just collect the images/data to the documents folder and have an update button which will rerun the script in the background. I travel a lot so i take a ton of pictures from the uber/lyft and it would be great to be able to find out more about them without Googling. I can think of tons of use cases.