pgvector-rust
pgvector support for Rust
Supports Rust-Postgres, SQLx, and Diesel
Getting Started
Follow the instructions for your database library:
Rust-Postgres
Add this line to your application’s Cargo.toml
under [dependencies]
:
pgvector = { version = "0.1", features = ["postgres"] }
Create a vector from a Vec
let vec = pgvector::Vector::from(vec![1.0, 2.0, 3.0]);
Insert a vector
client.execute("INSERT INTO table (column) VALUES ($1)", &[&vec])?;
Get the nearest neighbor
let row = client.query_one("SELECT * FROM table ORDER BY column <-> $1 LIMIT 1", &[&vec])?;
Retrieve a vector
let row = client.query_one("SELECT column FROM table LIMIT 1", &[])?;
let vec: pgvector::Vector = row.get(0);
Use Option
if the value could be NULL
let res: Option<pgvector::Vector> = row.get(0);
SQLx
Add this line to your application’s Cargo.toml
under [dependencies]
:
pgvector = { version = "0.1", features = ["sqlx"] }
Create a vector from a Vec
let vec = pgvector::Vector::from(vec![1.0, 2.0, 3.0]);
Insert a vector
sqlx::query("INSERT INTO table (column) VALUES ($1)").bind(vec).execute(&pool).await?;
Get the nearest neighbors
let rows = sqlx::query("SELECT * FROM table ORDER BY column <-> $1 LIMIT 1")
.bind(vec).fetch_all(&pool).await?;
Retrieve a vector
let row = sqlx::query("SELECT column FROM table LIMIT 1").fetch_one(&pool).await?;
let vec: pgvector::Vector = row.try_get("column")?;
Diesel
Add this line to your application’s Cargo.toml
under [dependencies]
:
pgvector = { version = "0.1", features = ["diesel"] }
And add this line to your application’s diesel.toml
under [print_schema]
:
import_types = ["diesel::sql_types::*", "pgvector::sql_types::*"]
Create a migration
diesel migration generate create_vector_extension
with up.sql
:
CREATE EXTENSION vector
and down.sql
:
DROP EXTENSION vector
Run the migration
diesel migration run
You can now use the vector
type in future migrations
CREATE TABLE items (
factors VECTOR(3)
)
For models, use:
pub struct Item {
pub factors: Option<pgvector::Vector>
}
Create a vector from a Vec
let factors = pgvector::Vector::from(vec![1.0, 2.0, 3.0]);
Insert a vector
let new_item = Item {
factors: Some(factors)
};
diesel::insert_into(items::table)
.values(&new_item)
.get_result::- (&conn)?;
Get the nearest neighbors
use pgvector::VectorExpressionMethods;
let neighbors = items::table
.order(items::factors.l2_distance(factors))
.limit(5)
.load::- (&conn)?;
Also supports max_inner_product
and cosine_distance
Add an approximate index in a migration
CREATE INDEX my_index ON items USING ivfflat (factors vector_l2_ops)
Use vector_ip_ops
for inner product and vector_cosine_ops
for cosine distance
Reference
Convert a vector to a Vec
let f32_vec = vec.to_vec();
History
View the changelog
Contributing
Everyone is encouraged to help improve this project. Here are a few ways you can help:
- Report bugs
- Fix bugs and submit pull requests
- Write, clarify, or fix documentation
- Suggest or add new features
To get started with development:
git clone https://github.com/ankane/pgvector-rust.git
cd pgvector-rust
cargo test --all-features