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

Add XTR Support #27

Merged
merged 6 commits into from
Jun 11, 2024
Merged

Add XTR Support #27

merged 6 commits into from
Jun 11, 2024

Conversation

mtbarta
Copy link
Contributor

@mtbarta mtbarta commented Jun 11, 2024 •
edited
Loading

This PR adds preliminary support for running XTR models in LintDB.

  • Tokenizer runs sentencepiece and adds an eos token for XTR.
  • Inverted Lists change to index codes per token for XTR. Therefore, XTR makes one less database call to score.
  • ProductEncoder handles quantization, with the help of new inverted list scanners and distance tables.

This touches a lot of code to help find better abstractions.

closes #11

@mtbarta mtbarta merged commit f9b3364 into main Jun 11, 2024
1 check passed
@mtbarta mtbarta deleted the feat/api-changes branch June 11, 2024 05:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

Support XTR
1 participant
@mtbarta