Tejas Shah
Tejas Shah
Lucene introduced [MultiLeafCollector](https://github.com/apache/lucene/blob/main/lucene/core/src/java/org/apache/lucene/search/knn/MultiLeafKnnCollector.java) to optimize concurrent segment search. While this cannot be used for faiss due to JNI layer, this can possibly be leveraged for exact search. This can further...
The implementation aims at pruning search space based on context awareness when context-aware-grouping mapper is presents. It does a best attempt extraction of grouping criteria from the query, if no...
**What do you want to do?** - [x] Request a change to existing documentation - [x] Add new documentation - [ ] Report a technical problem with the documentation -...
### Description Fixes integer overflow for efficient filtering ### Related Issues Resolves #[2901](https://github.com/opensearch-project/k-NN/issues/2901)] ### Check List - [x] New functionality includes testing. - [x] Commits are signed per the DCO...
When using efficient filtering, estimated distance computations are compared with max distance computation threshold to determine whether to use exact search or approximate nearest neighbor search. The Max distance computation...