Kishore Nallan comments

Results 569 comments of


                                            Kishore Nallan

facet_by's total_values is returning max_facet_value params value or default value when not present in params

Typesense supports two strategies for efficient faceting, and has some built-in heuristics to pick the right strategy for you by default. In `27.0.rc11` RC build, we have introduced a new...

[Feature Request] Ability to integrate results from custom Semantic Search server e.g. ColbertV2 or RAGatouille

We don't have plans to integrate with external semantic search implementations at the moment. I will leave this issue open as a feature request.

Vector search returning documents that are nowhere near the best match

@honzasterba The `k` parameter is described here: https://typesense.org/docs/0.25.1/api/vector-search.html#nearest-neighbor-vector-search It controls the number of results that are fetched from the vector index. > I can provide the DB files if necessary....

Vector search returning documents that are nowhere near the best match

@honzasterba Sorry, we could not get around to investigating this in more detail. Can you please share your data again? I have emailed you as well.

Vector search: search for multiple embedding-vectors

@iUnknownAdorn How do you want to combine the scores produced from each vector field?

Vector search: search for multiple embedding-vectors

That sounds good, we will add to our roadmap.

[Feature Request] Add custom metadata to a collection

Feature available in `0.26.0.rc48`. You can use the `metadata` key to add additional information to collection schema: ``` curl -k "http://localhost:8108/collections" -X POST -H "Content-Type: application/json" -H "X-TYPESENSE-API-KEY: ${TYPESENSE_API_KEY}" -d...

Kishore Nallan

facet_by's total_values is returning max_facet_value params value or default value when not present in params

[Feature Request] Ability to integrate results from custom Semantic Search server e.g. ColbertV2 or RAGatouille

Vector search returning documents that are nowhere near the best match

Vector search returning documents that are nowhere near the best match

Vector search: search for multiple embedding-vectors

Vector search: search for multiple embedding-vectors

[Feature Request] Add custom metadata to a collection

Facet counts are incorrect when combined with grouping and pinning

Facetting on numerical ranges doesn't seem to work when group_by is used

Make the filter clauses limit configurable.