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

Is binary vector data available for use at this point? #1691

Open
MORIMOTO520212 opened this issue Dec 13, 2024 · 1 comment
Open

Is binary vector data available for use at this point? #1691

MORIMOTO520212 opened this issue Dec 13, 2024 · 1 comment

Comments

@MORIMOTO520212
Copy link

Is binary vector data available for use at this point?
I configured the binary vector settings in the r2r.toml file, but upon checking the database, it seems to include regular float-type vectors as well. Which type of vector data is used during retrieval?
Finally, is it possible to remove the regular float-type vectors from the database?
@emrgnt-cmplxty
Copy link
Contributor

Great questions - the binary vector implementation is still a bit experimental.

You would need to make these changes by hand at the moment. If you would like to make a PR to make it more configurable s.t. we can drop the non-quantized vectors, then we would greatly welcome this contribution.

Otherwise, I will take note of this and circle back when we have the bandwidth. How have you found the binary performance? Are you combining with a reranker?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@MORIMOTO520212 @emrgnt-cmplxty