-
**What is the bug?**
we had an index with 5 shards and we wanted to rindex it to a new index with one shard. approximate search does not show results with exact same vector in the new index.
Happe…
-
### Checklist
- [X] I have searched for [similar issues](https://github.com/isl-org/Open3D/issues).
- [X] For Python issues, I have tested with the [latest development wheel](https://www.open3d.org/d…
-
When I experiment with the HNSW indexing method, kNN search returns less number of results than requested. Is this behavior normal? If normal, how does the algorithm decide how many neighbors to retur…
-
should be faster than calculating the entire matrix
see #32
(creating this issues for documentation purposes)
-
Today, if we ingest a large piece of text into a Knowledge base entry, only the first 512 word pieces are used for creating the embeddings that ELSER uses to match on during semantic search.
This m…
-
### Description
Users new to vector search are sometimes surprised by one major difference relative to lexical/BM25 search — all vectors "match" the query vector. Users are frequently familiar only…
-
Currently most of our the nearest neighbors methods support "int64_t" type for the nearest neighbor indices. The only exception is CAGRA, that returns `uint32_t` indices. For consistency, it would be …
-
In KNN style searches, we should support `rowid in (...)` constraints in queries like so:
```sql
select rowid, distance
from vss_articles
where vss_search(description_embeddings, :query_vector)
…
-
{'error': {'root_cause': [{'type': 'illegal_argument_exception', 'reason': 'unknown setting [index.knn] please check that any required plugins are installed, or check the breaking changes documentatio…
ghost updated
2 years ago
-
Incremental knn search save the priority queue in each round generated by knn search if best-first strategy is adopted, and use priority queue saved in the previous round to search knn in the current …