mirror of
https://github.com/lancedb/lancedb.git
synced 2026-11-13 22:26:53 +00:00
Jack YeandClaude Sonnet 5
de95ed3f6d
A broad documentation pass across the areas LanceDB Enterprise and Embedded diverge on most: **Data Curation** -- GPU Function execution, Slurm scheduling (and what Slurm is, for readers new to it), blob late materialization as its own page, joins, aggregation/profiling, sampling, near-duplicate detection, chunking, distributed shuffle, and IVF-partition clustering. **Feature Engineering** -- computed columns and materialized views, restructured to lead with the Function-based path and show the SQL form side by side with Python throughout; a single Auto Refresh page instead of duplicating scheduling details on every refresh page; adaptive execution; checkpointing; GPU and Slurm placement for refresh workers; a Training Cache page for the PyTorch data loader's automatic cache benefit. **Security & Governance** -- an Audit Log reference (schema, access, grants), Secrets promoted to an ordinary catalog object with their own SQL DDL and a vending/rotation guide, and privilege-table fixes verified against the authorization source. **Deployment & Maintenance** -- Horizontal Scaling and OpenTelemetry moved into a real Deployment area; Automated Cleanup folded into Intelligent Auto-Maintenance (every other maintenance decision already lived there); RTE Compaction and Advanced Conflict Resolution documented in depth. **Landing page** -- the Enterprise feature comparison table reorganized (several redundant or stale rows removed, several new areas added), and the "why Enterprise" bullets replaced with ones tied to a concrete page (Performance, Scale, Fault tolerance, Multicloud deployment, Simplify operations) instead of vague promises. Plus a full pass addressing lancedb-gatekeeper's review: a real test-safety bug (a quickstart test could silently overwrite and delete a deployment's own `frames` table), a fabricated SQL statement with no backing implementation, an actual dedup-guarantee bug in `take_row_ids`, and dozens of documentation-accuracy fixes verified against source rather than taken on faith -- each either fixed, or left with an explicit note on why it needs a feature or deployment owner's sign-off instead of being guessed at. 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_018PXMCbRH6i5w47TsjeEUMT --------- Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
60 lines
2.5 KiB
Plaintext
60 lines
2.5 KiB
Plaintext
---
|
|
title: "IVF_PARTITION"
|
|
sidebarTitle: "IVF_PARTITION"
|
|
description: "The ivf_partition() scalar function: assign each row to its trained IVF index partition."
|
|
icon: circle-nodes
|
|
keywords: ["ivf_partition", "sql", "vector index", "ivf", "partition", "scalar function", "clustering", "blocking"]
|
|
availability:
|
|
embedded: unavailable
|
|
enterprise: available
|
|
---
|
|
|
|
<CodeGroup>
|
|
```sql SQL icon="database"
|
|
SELECT
|
|
ivf_partition(embedding, selection => 'random', count => 100, seed => 42) AS partition,
|
|
array_agg(id) AS ids
|
|
FROM documents
|
|
GROUP BY partition
|
|
```
|
|
</CodeGroup>
|
|
|
|
`ivf_partition` reads an existing IVF-family vector index's trained
|
|
partitions and assigns each row to the one its vector falls into. The column
|
|
needs an up-to-date, single-segment index before the function runs -- the
|
|
same requirement [`dedupe`](/curation/dedupe#example-data) has.
|
|
|
|
Grouping by its result buckets rows by vector proximity without a full
|
|
pairwise scan -- useful for blocking a large table into comparable chunks,
|
|
such as before a pairwise similarity pass.
|
|
|
|
## Input {#input}
|
|
|
|
`ivf_partition` takes exactly 1, 3, or 4 arguments -- the vector column
|
|
alone (scans every partition), or the vector column plus `selection` and
|
|
`count`, plus `seed` only when `selection` is `'random'`:
|
|
|
|
| Argument | Type | Required | Meaning |
|
|
| --- | --- | --- | --- |
|
|
| vector column | column reference | yes | the indexed vector column to assign |
|
|
| `selection` | `'first'` \| `'random'` | required with `count` | which partitions to sample -- there is no explicit `'all'`; omit `selection` and `count` entirely to scan every partition |
|
|
| `count` | int, > 0 | required with `selection` | number of partitions to sample |
|
|
| `seed` | int, >= 0 | required for `'random'`, refused for `'first'` | random seed |
|
|
|
|
Arguments can be passed positionally, `ivf_partition(embedding, 'random', 100, 42)`,
|
|
or named as shown above. `count` and `seed` are the same scan-scoping
|
|
arguments [`duplicate_pairs`](/sql/dql/duplicate-pairs#sampling) accepts,
|
|
so a sampled partition estimate carries over -- but not `dedupe`, which
|
|
always scans every partition and refuses sampling arguments entirely.
|
|
|
|
## What to read next {#next}
|
|
|
|
<CardGroup cols={2}>
|
|
<Card title="Cluster records by index partition" icon="circle-nodes" href="/curation/clustering">
|
|
Use this function to group and sample records by an existing index's partitions.
|
|
</Card>
|
|
<Card title="DEDUPE" icon="layer-group" href="/sql/dql/dedupe">
|
|
The packaged table function for the deduplication use case, over the same index.
|
|
</Card>
|
|
</CardGroup>
|