[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Bring your own databsae #272

Open
C4Phoenix opened this issue Sep 13, 2024 · 1 comment
Open

Bring your own databsae #272

C4Phoenix opened this issue Sep 13, 2024 · 1 comment
Labels
documentation Improvements or additions to documentation Kalosm Related to the Kalosm library

Comments

@C4Phoenix
Copy link
C4Phoenix commented Sep 13, 2024 •
edited
Loading

Specific Demand request

Issue: 'First party database integration' #134 talks about the desire too support the ability for users too implement their own database. I really like this project and would love too use and contribute too it. Being able (or seeing how) too substitute a different DB is what would help me tremendously.

Implement Suggestion

  • more use of documented traits for the document store vector store etc.
  • I was planning on using tantivy but maybe I can implement it somewhere here under an optional feature against a trait.
  • make Surrealdb more optional. (I currently chose against using it because it did not support binary storage)

other

Let me know if your planning on supporting something like this and how I would be able too help on implementation/documentation.
@ealmloff ealmloff added documentation Improvements or additions to documentation Kalosm Related to the Kalosm library labels Sep 14, 2024
@ealmloff
Copy link
Collaborator

A guide in the docs or tantivy integration for full text search would both be great additions. You can use VectorDB to do vector search on top of your database. First you need to get an embedding for your text. You can do that with Bert::embed. Every vector you insert into the vector index with add_embedding gets a unique id. You need to provide a mapping between that id and the doc id in your database then you can do vector search with get_closest, get the doc id with your mapping and use that doc id to get the final document

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
documentation Improvements or additions to documentation Kalosm Related to the Kalosm library
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@C4Phoenix @ealmloff