Dot Product Support Similarity Metric for IndexIVFFlatPanorama
#4732
+644
−440
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
AlSchlo
changed the title
IndexIVFFlatPanorama
AlSchlo
force-pushed
the
dot-product-panorama
branch
from
99ff3ce to
d355aa0
Compare
Contributor
|
@mnorris11 has imported this pull request. If you are a Meta employee, you can view this in D89670228. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
As title, will tackle
IVFFlat&HNSWin subsequent PRs.Stacks upon this refactoring PR.
(Not necessarily dependent but makes my workflow a bit easier).
Did some refactoring to keep the code clean. Will do further test code refactoring first before supporting other indexes.
Testing Plan
Adapted existing tests to run with both distance metrics. However this proved tricky due to floating point imprecision, so I had to loose up some constraints at some places.
Note
The diffs of this PR are horrible, but that is mostly due to the fact that each Python line gets indented...
If you open each file prev-new next to each other, you will see it is very similar.