victor-db

Crates.iovictor-db
lib.rsvictor-db
version0.1.3
sourcesrc
created_at2023-09-14 02:02:38.786151
updated_at2023-09-14 16:13:28.430947
descriptionA browser-optimized vector database
homepage
repositoryhttps://github.com/not-pizza/victor/
max_upload_size
id972136
size9,230,632
Andre Popovitch (anchpop)

documentation

README

Victor

Web-optimized vector database (written in Rust).

Features

  1. Rust API (using native filesystem, or a transient in-memory filesystem)

  2. Web API (Using the Private Origin File System)

  3. Very efficient vector storage format

    1. For a vector with 1536 dimensions, our representation consumes 1.5 KB, while naively encoding with JSON would consume 20.6 KB.
  4. PCA for vector compression when storage space is low

JS Example

Installation

npm install victor-db

Usage

import { Db } from "victor";

const db = await Db.new();

const content = "My content!";
const tags = ["these", "are", "tags"];
const embedding = new Float64Array(/* your embedding here */);

// write to victor
await db.insert(content, embedding, tags);

// read the 10 closest results from victor that are tagged with "tags"
// (only 1 will be returned because we only inserted one embedding)
const result = await db.search(embedding, ["tags"], 10);
assert(result[0].content == content);

// clear database
await db.clear();

See www/ for a more complete example, including fetching embeddings from OpenAI.

Rust Example

Installation

cargo add victor-db

Usage

use std::path::PathBuf;

use victor_db::native::Db;

let _ = std::fs::create_dir("./victor_test_data");
let mut victor = Db::new(PathBuf::from("./victor_test_data"));

victor.clear_db().await.unwrap();

victor
    .write(
        "Test Vector 1",
        vec![1.0, 0.0, 0.0],
        vec!["Test".to_string()],
    )
    .await;
victor
    .write(
        "Test Vector 2",
        vec![0.0, 1.0, 0.0],
        vec!["Test".to_string()],
    )
    .await;

// read the 10 closest results from victor that are tagged with "tags"
// (only 2 will be returned because we only inserted two embeddings)
let nearest = victor
   .find_nearest_neighbors(vec![0.9, 0.0, 0.0], vec!["Test".to_string()], 10)
   .await
   .first()
   .unwrap()
   .content
   .clone();
assert_eq!(nearest, "Test Vector 1".to_string());

This example is also in the /examples directory. If you've cloned this repository, you can run it with cargo run --example native_filesystem.

Hacking

  1. Victor is written in Rust, and compiled to wasm with wasm-pack.

    Install wasm pack with cargo install wasm-pack or npm i -g wasm-pack (https://rustwasm.github.io/wasm-pack/installer/)

  2. Build Victor with wasm-pack build

  3. Set up the example project, which is in www/.

    If you use nvm, you can just run cd www/ && nvm use

    Then, npm i.

  4. From www/, start the example project with npm run start.

Architecture

Relevant code at src/packed_vector.rs.

Packed vector storage explanation


File structure explanation

Us

Sam Hall

Andre Popovitch

Commit count: 54

cargo fmt