danvics/haiku.rag
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Record what the multi-database work changes, and what it measured

Browse source 
`--db-name` selects one database, not a subset: it is not repeatable.

The reranker recommendation was one-sided. It reports both measured effects now:
stronger aggregate retrieval across shards, and weaker attribution between
near-identical documents, where fusion keeps twins apart because each database
contributes its own top-ranked result.

Image queries are vector-only and skip the reranker.

The changelog described components of the new feature as fixes to the last
release. They are one Added entry, and the changes a user upgrading does see —
`settings` printing YAML, the document filter paging and searching, `list`
printing only the fields a document has — are listed.
This commit is contained in:
Yiorgis Gozadinos 2026-08-27 17:30:29 +03:00
parent 0dec3d4e63
commit 4d7fdd2d26
No known key found for this signature in database
2 changed files with 45 additions and 24 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

33
CHANGELOG.md
View file

@ -4,25 +4,30 @@
### Added ### Added
- `lancedb.databases` configures a named set of local or remote databases. `search`, - `lancedb.databases` configures a named set of local or remote databases.
`ask` and `analyze` accept a `sources` subset; results, documents and citations `search`, `ask` and `analyze` accept a `sources` subset; results, documents and
include the originating database in `source`. Search results are combined with citations carry the originating database in `source`, and model context names
the configured reranker, or reciprocal rank fusion when reranking is disabled. it as `Collection:` when a search spans more than one. Candidates are combined
- `haiku-rag search`, `ask`, `analyze` and `chat` can cover a configured database with the configured reranker, or reciprocal rank fusion when reranking is
set. Commands that access one database select it with `--db-name NAME` or disabled. Operations that need one database raise `AmbiguousDatabaseError`, an
unknown name raises `UnknownDatabaseError`, and a cited chunk ID retrieved from
or previously cited from more than one selected database raises
`AmbiguousCitationError`.
- `haiku-rag search`, `ask`, `analyze` and `chat` cover a configured set.
Commands that access one database select it with `--db-name NAME` or
`--db PATH`. `--db PATH`.
- Citation resolution rejects a chunk ID retrieved from, or previously cited - `HaikuRAG.aclose()` releases a client whatever it covers. `close()` remains
from, more than one selected database with `AmbiguousCitationError`. limited to clients covering one database.
### Fixed ### Fixed
- `client.chunk()` and `client.embedder` work when the client covers multiple - `haiku-rag settings` prints YAML instead of Python dict reprs.
databases. Operations that require one database raise `AmbiguousDatabaseError`. - The chat document filter pages results, searches the database for the rest, and
- The chat document filter selects by document ID and shows each document's lists the selected separately, instead of mounting a checkbox per document.
database. Selection is by document ID.
- `haiku-rag list` prints only the fields a document has.
- `haiku-rag` prints the message and exits when the configured embedder does not match the database, instead of raising a traceback. - `haiku-rag` prints the message and exits when the configured embedder does not match the database, instead of raising a traceback.
- Capabilities created without a client now honor `lancedb.uri` and - Capabilities created without a client honor `lancedb.uri`.
`lancedb.databases`.
- A `lancedb.uri` without a scheme is treated as a local path. `--db PATH` - A `lancedb.uri` without a scheme is treated as a local path. `--db PATH`
overrides it. overrides it.
- Inspector search results mark truncated previews with an ellipsis. - Inspector search results mark truncated previews with an ellipsis.

36
docs/configuration/storage.md
View file

@ -278,22 +278,38 @@ document in each of them.
#### Ranking #### Ranking
Configure a reranker when searching multiple databases. Reciprocal rank fusion Reciprocal rank fusion compares positions rather than scores, so each database
compares positions rather than scores, so each database contributes top-ranked contributes top-ranked results even when another database has stronger matches. A
results even when another database has stronger matches. A reranker scores the reranker scores the combined candidate set directly, which has been measured to
combined candidate set directly. help aggregate retrieval and to hurt attribution between near-identical
documents.
In a 3,045-query evaluation over a corpus split across three databases, Aggregate retrieval is stronger with a reranker. In a 3,045-query evaluation over
reranking produced retrieval MAP 0.9914, compared with 0.9918 for the same corpus a corpus split across three databases, reranking produced retrieval MAP 0.9914,
in one database. Without a reranker, MAP was 0.6044, compared with 0.9798 in one compared with 0.9918 for the same corpus in one database. Without a reranker, MAP
database. Reranking cost grows with the number of databases because each was 0.6044, compared with 0.9798 in one database. Reranking cost grows with the
contributes candidates. number of databases because each contributes candidates.
A reranker scores the combined candidates with no notion of which database each
came from, so on near-identical text it can pick the wrong database's chunk,
where fusion keeps them apart because each database contributes its own
top-ranked result. In two nine-case acceptance runs over a synthetic corpus
holding one station in two databases under near-identical names, attribution was
weaker with reranking: citing the right database succeeded 5 of 9 and 6 of 9
times with a reranker, against 8 of 9 and 9 of 9 without.
Configure a reranker where retrieval breadth matters, and measure it where
answers have to attribute between documents that read alike.
Without a reranker, consider increasing `search.limit` with the number of Without a reranker, consider increasing `search.limit` with the number of
databases. With three complete rankings and a limit of 5, a database may databases. With three complete rankings and a limit of 5, a database may
contribute only one or two results. A higher limit also sends more results to contribute only one or two results. A higher limit also sends more results to
the caller and model. the caller and model.
Image queries are vector-only and skip the reranker: there is no query text to
score a document against, so candidates keep their vector ranking and fusion
ranks by position.
If any selected database fails to open, the operation fails and identifies that If any selected database fails to open, the operation fails and identifies that
database. database.
@ -319,7 +335,7 @@ available on a multi-database client.
Commands use database sets as follows: Commands use database sets as follows:
- **Set-capable**: `search`, `ask`, `analyze`, and `chat` use the full - **Set-capable**: `search`, `ask`, `analyze`, and `chat` use the full
configured set, or the subset named by `--db-name`. configured set, or the single database selected by `--db-name`.
- **Config-only**: `settings`, `init-config`, and `download-models` do not open - **Config-only**: `settings`, `init-config`, and `download-models` do not open
a database. a database.
- **Single-database**: everything else — document writes, `rebuild`, `vacuum`, - **Single-database**: everything else — document writes, `rebuild`, `vacuum`,