Skip to content

Conversation

@bamader
Copy link
Collaborator

@bamader bamader commented Oct 20, 2025

Description

This PR introduces approximate neighbor search to our performance metrics notebook. It also creates a script that uses a relative grid-search to optimize the parameters for HNSW evaluation.

Related Issues

Closes #106

@codecov-commenter
Copy link

codecov-commenter commented Oct 20, 2025

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.87%. Comparing base (977e2d6) to head (eaed1d7).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files
@@           Coverage Diff           @@
##             main     #119   +/-   ##
=======================================
  Coverage   96.87%   96.87%           
=======================================
  Files           2        2           
  Lines          32       32           
=======================================
  Hits           31       31           
  Misses          1        1           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

Copy link
Collaborator

@m-goggins m-goggins left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good overall! Left a couple of questions and small nits about documentation. I think the notes about the HNSW parameters could live in the scripts for now or in a big documentation file about our approach. Maybe just leave in the scripts for now and add the larger documentation later?

@bamader
Copy link
Collaborator Author

bamader commented Oct 23, 2025

I also uploaded a few util scripts with the PR that might come in handy for one-off stuff in preparation for Azure (e.g., computing the HNSW index for a particular set of embeddings, or converting pickle-saved embedding Tensors into CPU format)

@bamader bamader merged commit 7f25d8e into main Oct 23, 2025
5 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Investigate and Implement Approximate Nearest Neighbor Search

4 participants