neural-search
https://github.com/opensearch-project/neural-search
Java
Plugin that adds dense neural retrieval into the OpenSearch ecosytem
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Java not yet supported0 Subscribers
Add a CodeTriage badge to neural-search
Help out
- Issues
- [BUG] Hybrid query reports hits.total.relation "eq" when a shard's hit count lands exactly on track_total_hits
- [BUG] Hybrid query: explain mode changes returned _score values (and hit order for z_score)
- [BUG] Hybrid query wrapped in boosting or nested is not rejected and runs ignoring normalization pipeline
- [AUTO] Increment version to 2.19.8-SNAPSHOT
- [AUTO] Increment version to 3.11.0-SNAPSHOT
- [BUG] Under concurrent segment search in classic hybrid query ranking depends on segment layout, one-shard fetch is inflated, rescore runs per slice
- [FEATURE] Exact (brute-force) k-NN query for semantic fields
- [BUG] Batch semantic highlighting misses inner_hits targets under boosting or script_score when a compound query sits in between
- [Hybrid Query] Run a hybrid query that no search pipeline normalizes in fusion mode with the default techniques
- [PROPOSAL] Hybrid query: allow `sort: [{"_score": "desc"}, {"<field>": ...}]` so `search_after` can page relevance-ordered results
- Docs
- Java not yet supported