Feature/rrf score normalization v2 #1089

martin-gaievski · 2025-01-10T21:24:15Z

Description

[Describe what this change achieves]

Related Issues

Resolves #[Issue number to be closed when this PR is merged]

Check List

New functionality includes testing.
New functionality has been documented.
API changes companion pull request created.
Commits are signed per the DCO using --signoff.
Public documentation issue/PR created.

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.

…874) * initial commit of RRF Signed-off-by: Isaac Johnson <[email protected]> Co-authored-by: Varun Jain <[email protected]> Signed-off-by: Martin Gaievski <[email protected]>

* Initial unit test implementation Signed-off-by: Ryan Bogan <[email protected]> --------- Signed-off-by: Ryan Bogan <[email protected]> Signed-off-by: Martin Gaievski <[email protected]>

* Integrate explainability for hybrid query into RRF processor Signed-off-by: Martin Gaievski <[email protected]>

Signed-off-by: Bo Zhang <[email protected]>

* add impl Signed-off-by: zhichao-aws <[email protected]> * add UT Signed-off-by: zhichao-aws <[email protected]> * rename pruneType; UT Signed-off-by: zhichao-aws <[email protected]> * changelog Signed-off-by: zhichao-aws <[email protected]> * ut Signed-off-by: zhichao-aws <[email protected]> * add it Signed-off-by: zhichao-aws <[email protected]> * change on 2-phase Signed-off-by: zhichao-aws <[email protected]> * UT Signed-off-by: zhichao-aws <[email protected]> * it Signed-off-by: zhichao-aws <[email protected]> * rename Signed-off-by: zhichao-aws <[email protected]> * enhance: more detailed error message Signed-off-by: zhichao-aws <[email protected]> * refactor to prune and split Signed-off-by: zhichao-aws <[email protected]> * changelog Signed-off-by: zhichao-aws <[email protected]> * fix UT cov Signed-off-by: zhichao-aws <[email protected]> * address review comments Signed-off-by: zhichao-aws <[email protected]> * enlarge score diff range Signed-off-by: zhichao-aws <[email protected]> * address comments: check lowScores non null instead of flag Signed-off-by: zhichao-aws <[email protected]> --------- Signed-off-by: zhichao-aws <[email protected]>

Signed-off-by: Bo Zhang <[email protected]>

) Signed-off-by: Bo Zhang <[email protected]>

* Allow empty string for field in field map Signed-off-by: Yizhe Liu <[email protected]> * Allow empty string when validation Signed-off-by: Yizhe Liu <[email protected]> * Add to change log Signed-off-by: Yizhe Liu <[email protected]> * Update CHANGELOG to: Support empty string for fields in text embedding processor Signed-off-by: Yizhe Liu <[email protected]> --------- Signed-off-by: Yizhe Liu <[email protected]>

…nested objects (#1040) * Fix bug where ingestion failed for input document containing list of nested objects Signed-off-by: Yizhe Liu <[email protected]> * Address comments to use better method name/implementation Signed-off-by: Yizhe Liu <[email protected]> * Address comments: modify the test case to have doc with various fields Signed-off-by: Yizhe Liu <[email protected]> --------- Signed-off-by: Yizhe Liu <[email protected]>

Signed-off-by: Martin Gaievski <[email protected]>

…es (#1043) * Fixed mismatch between document source and score fields when sorting is enabled in hybrid query Signed-off-by: Martin Gaievski <[email protected]>

) Signed-off-by: will-hwang <[email protected]>

* add support for builder constructor in neural query builder Signed-off-by: will-hwang <[email protected]> * create custom builder class to enforce valid neural query builder instantiation Signed-off-by: will-hwang <[email protected]> * refactor code to remove duplicate Signed-off-by: will-hwang <[email protected]> * include new constructor in qa packages Signed-off-by: will-hwang <[email protected]> * refactor code to remove unnecessary code Signed-off-by: will-hwang <[email protected]> * fix bug in neural query builder instantiation Signed-off-by: will-hwang <[email protected]> --------- Signed-off-by: will-hwang <[email protected]>

* add hybrid search with rescore IT Signed-off-by: will-hwang <[email protected]> * remove rescore in hybrid search IT Signed-off-by: will-hwang <[email protected]> * remove previous version checks in build file Signed-off-by: will-hwang <[email protected]> * removing version checks only in rolling upgrade tests Signed-off-by: will-hwang <[email protected]> * remove newly added tests in restart test Signed-off-by: will-hwang <[email protected]> * Revert "remove newly added tests in restart test" This reverts commit 0987831. Signed-off-by: will-hwang <[email protected]> --------- Signed-off-by: will-hwang <[email protected]>

…t has dot in field name (#1062) * Fix bug where document embedding fails to be generated due to document has dot in field name Signed-off-by: Yizhe Liu <[email protected]> * Address comments Signed-off-by: Yizhe Liu <[email protected]> --------- Signed-off-by: Yizhe Liu <[email protected]>

…EmbeddingProcessorIT (#1074) Signed-off-by: Yizhe Liu <[email protected]>

* Add reindex integration tests Signed-off-by: Andy Qin <“[email protected]”>

* Fix github CI by adding eclipse dependency in formatting.gradle Signed-off-by: Varun Jain <[email protected]> * Add changelog Signed-off-by: Varun Jain <[email protected]> --------- Signed-off-by: Varun Jain <[email protected]>

Signed-off-by: Bo Zhang <[email protected]>

Signed-off-by: Martin Gaievski <[email protected]>

…874) * initial commit of RRF Signed-off-by: Isaac Johnson <[email protected]> Co-authored-by: Varun Jain <[email protected]> Signed-off-by: Martin Gaievski <[email protected]>

* Initial unit test implementation Signed-off-by: Ryan Bogan <[email protected]> --------- Signed-off-by: Ryan Bogan <[email protected]> Signed-off-by: Martin Gaievski <[email protected]>

* Integrate explainability for hybrid query into RRF processor Signed-off-by: Martin Gaievski <[email protected]>

codecov · 2025-01-10T21:57:25Z

Codecov Report

Attention: Patch coverage is 87.33333% with 19 lines in your changes missing coverage. Please review.

Project coverage is 80.47%. Comparing base (b4cb267) to head (a661ca3).

Files with missing lines	Patch %	Lines
...rocessor/normalization/ScoreNormalizationUtil.java	41.17%	6 Missing and 4 partials ⚠️
...pensearch/neuralsearch/processor/RRFProcessor.java	83.78%	0 Missing and 6 partials ⚠️
...rmalization/MinMaxScoreNormalizationTechnique.java	60.00%	2 Missing ⚠️
...essor/normalization/RRFNormalizationTechnique.java	97.72%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #1089      +/-   ##
============================================
+ Coverage     80.19%   80.47%   +0.28%     
- Complexity     1139     1198      +59     
============================================
  Files            87       93       +6     
  Lines          3953     4077     +124     
  Branches        666      681      +15     
============================================
+ Hits           3170     3281     +111     
- Misses          531      536       +5     
- Partials        252      260       +8

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

Johnsonisaacn and others added 26 commits December 17, 2024 15:49

Reciprocal Rank Fusion (RRF) normalization technique in hybrid query (#…

ece4069

…874) * initial commit of RRF Signed-off-by: Isaac Johnson <[email protected]> Co-authored-by: Varun Jain <[email protected]> Signed-off-by: Martin Gaievski <[email protected]>

Reciprocal Rank Fusion (RRF) normalization technique in hybrid query (#…

582e882

…874) * initial commit of RRF Signed-off-by: Isaac Johnson <[email protected]> Co-authored-by: Varun Jain <[email protected]> Signed-off-by: Martin Gaievski <[email protected]>

Add integration and unit tests for missing RRF coverage (#997)

627fcb4

* Initial unit test implementation Signed-off-by: Ryan Bogan <[email protected]> --------- Signed-off-by: Ryan Bogan <[email protected]> Signed-off-by: Martin Gaievski <[email protected]>

Integrate explainability for hybrid query into RRF processor (#1037)

1d93192

* Integrate explainability for hybrid query into RRF processor Signed-off-by: Martin Gaievski <[email protected]>

Support of new k-NN query parameter expand_nested. (#1013)

ab16432

Signed-off-by: Bo Zhang <[email protected]>

Remove mistakenly added code from HybridSearchIT. (#1032)

8086520

Signed-off-by: Bo Zhang <[email protected]>

Fix random total hits range for ExplanationPayloadProcessorTests. (#1034

931dbe8

) Signed-off-by: Bo Zhang <[email protected]>

Added runtime dependencies for jackson lib (#1055)

f35132d

Signed-off-by: Martin Gaievski <[email protected]>

Fixed document source and score field mismatch in sorted hybrid queri…

692433c

…es (#1043) * Fixed mismatch between document source and score fields when sorting is enabled in hybrid query Signed-off-by: Martin Gaievski <[email protected]>

Optimize ML inference connection retry logic (#1054)

53eeffc

increase cluster node timeout value for time consuming test cases (#1063

6f0382d

) Signed-off-by: will-hwang <[email protected]>

Clean up unused validateFieldName() and use existing methods for Text…

c9c13e4

…EmbeddingProcessorIT (#1074) Signed-off-by: Yizhe Liu <[email protected]>

Add reindex integration tests (#1075)

7b9c3e1

* Add reindex integration tests Signed-off-by: Andy Qin <“[email protected]”>

Correct NeuralQueryBuilder doEquals() and doHashCode(). (#1045)

477a9ec

Signed-off-by: Bo Zhang <[email protected]>

Rename explanation response processor (#1085)

71de351

Signed-off-by: Martin Gaievski <[email protected]>

Reciprocal Rank Fusion (RRF) normalization technique in hybrid query (#…

719a6a3

…874) * initial commit of RRF Signed-off-by: Isaac Johnson <[email protected]> Co-authored-by: Varun Jain <[email protected]> Signed-off-by: Martin Gaievski <[email protected]>

Add integration and unit tests for missing RRF coverage (#997)

36b2e47

* Initial unit test implementation Signed-off-by: Ryan Bogan <[email protected]> --------- Signed-off-by: Ryan Bogan <[email protected]> Signed-off-by: Martin Gaievski <[email protected]>

Integrate explainability for hybrid query into RRF processor (#1037)

d68b296

* Integrate explainability for hybrid query into RRF processor Signed-off-by: Martin Gaievski <[email protected]>

Merge branch 'main' into feature/rrf-score-normalization-v2

a661ca3

martin-gaievski closed this Jan 10, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Feature/rrf score normalization v2 #1089

Feature/rrf score normalization v2 #1089

martin-gaievski commented Jan 10, 2025

codecov bot commented Jan 10, 2025

Feature/rrf score normalization v2 #1089

Feature/rrf score normalization v2 #1089

Conversation

martin-gaievski commented Jan 10, 2025

Description

Related Issues

Check List

codecov bot commented Jan 10, 2025

Codecov Report