-
Notifications
You must be signed in to change notification settings - Fork 114
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.
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
Add Benchmarking Config for Faiss SQfp16 #1575
Add Benchmarking Config for Faiss SQfp16 #1575
Conversation
Signed-off-by: Naveen Tatikonda <[email protected]>
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #1575 +/- ##
============================================
- Coverage 85.15% 84.96% -0.20%
+ Complexity 1371 1366 -5
============================================
Files 172 172
Lines 5566 5566
Branches 546 546
============================================
- Hits 4740 4729 -11
- Misses 596 605 +9
- Partials 230 232 +2 ☔ View full report in Codecov by Sentry. |
@naveentatikonda Lets not add the SQ benchmarks in perf tool and instead put this in OSB. We have migrated our benchmarks to OSB repo. Lets use that repo to add new benchmarks. cc: @vamshin |
Sure I will add it in OSB-workloads. But, shall we also add it here just in case if we want to run tests using perf-tool. Also, if I'm not wrong as of now the OSB supports vector search with no train procedure only. |
Yes it supports only no-train procedures. But SQFp-16 is no train workload right. So we should be able to support it with OSB. |
I would not recommend. |
We don't need training for hnsw. But, for ivf we need training. But, that makes sense that we want to migrate to OSB-workloads. Will add it there. Thanks! |
Description
Add configuration to run benchmarking test for Faiss SQfp16.
Check List
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.