pgvector.git / summary / log / commit / refs

commit    f1983ce6722417e1f056f83ea2faf0d89cd6c71b
Author:   Andrew Kane <andrew@ankane.org>
Date:     Mon Mar 06 01:34:49 2023 +0000

    Added section on querying - #64 [skip ci]


README.md | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/README.md b/README.md index 71a71f7a..c8a3a1cd 100644 --- a/README.md +++ b/README.md @@ -145,6 +145,22 @@ To index many different values of `category_id`, consider [partitioning](https:/ CREATE TABLE items (embedding vector(3), category_id int) PARTITION BY LIST(category_id); ``` +## Querying + +Use a `SELECT` clause to get the distance + +```sql +SELECT embedding <-> '[3,1,2]' AS distance FROM items; +``` + +Use a `WHERE` clause to get rows within a certain distance + +```sql +SELECT * FROM items WHERE embedding <-> '[3,1,2]' < 100; +``` + +Note: Combine with `ORDER BY` and `LIMIT` to use an index + ## Performance To speed up queries without an index, increase `max_parallel_workers_per_gather`. [parent: 0a3615b82f1d]