Skip to content

Latest commit

 

History

History
321 lines (240 loc) · 9.72 KB

README_CN.md

File metadata and controls

321 lines (240 loc) · 9.72 KB

Hora

[Homepage] [Document] [Examples]

Hora Search Everywhere!

Hora 完全基于 Rust🦀 实现,事实证明,Rust 确实非常非常快,完全可以媲美 C++ ,且Hora使用 SIMD进行了加速,速度非常快⚡️⚡️⚡️,具体速度可以参考下面的 benchmark.

Hora, 日语为 「ほら」,读法像 [hōlə] ,意思是 Wow, You see! , Look at that! 。 这个名字的灵感来自日本著名歌曲 「小さな恋のうた」

Demos

👩 Face-Match [online demo], have a try!

🍷 Dream wine comments search [online demo], have a try!

Features

  • Performant ⚡️

    • SIMD-Accelerated (packed_simd)
    • Stable algorithm implementation
    • Multiple threads design
  • Supports Multiple Languages ☄️

    • Python
    • Javascript
    • Java
    • Go (WIP)
    • Ruby (WIP)
    • Swift (WIP)
    • R (WIP)
    • Julia (WIP)
    • Can also be used as a service
  • Supports Multiple Indexes 🚀

    • Hierarchical Navigable Small World Graph Index (HNSWIndex) (details)
    • Satellite System Graph (SSGIndex) (details)
    • Product Quantization Inverted File(PQIVFIndex) (details)
    • Random Projection Tree(RPTIndex) (LSH, WIP)
    • BruteForce (BruteForceIndex) (naive implementation with SIMD)
  • Portable 💼

    • Supports WebAssembly
    • Supports Windows, Linux and OS X
    • Supports IOS and Android (WIP)
    • Supports no_std (WIP, partial)
    • No heavy dependencies, such as BLAS
  • Reliability 🔒

    • Rust compiler secures all code
    • Memory managed by Rust for all language libraries such as Python's
    • Broad testing coverage
  • Supports Multiple Distances 🧮

    • Dot Product Distance
      • equation
    • Euclidean Distance
      • equation
    • Manhattan Distance
      • equation
    • Cosine Similarity
      • equation
  • Productive

    • Well documented
    • Elegant, simple and easy to learn API

Installation

Rust

in Cargo.toml

[dependencies]
hora = "0.1.1"

Python

$ pip install horapy

Javascript (WebAssembly)

$ npm i horajs

Building from source

$ git clone https://github.com/hora-search/hora
$ cargo build

Benchmarks

by aws t2.medium (CPU: Intel(R) Xeon(R) CPU E5-2686 v4 @ 2.30GHz) more information

Examples

Rust example [more info]

use hora::core::ann_index::ANNIndex;
use rand::{thread_rng, Rng};
use rand_distr::{Distribution, Normal};

pub fn demo() {
    let n = 1000;
    let dimension = 64;

    // make sample points
    let mut samples = Vec::with_capacity(n);
    let normal = Normal::new(0.0, 10.0).unwrap();
    for _i in 0..n {
        let mut sample = Vec::with_capacity(dimension);
        for _j in 0..dimension {
            sample.push(normal.sample(&mut rand::thread_rng()));
        }
        samples.push(sample);
    }

    // init index
    let mut index = hora::index::hnsw_idx::HNSWIndex::<f32, usize>::new(
        dimension,
        &hora::index::hnsw_params::HNSWParams::<f32>::default(),
    );
    for (i, sample) in samples.iter().enumerate().take(n) {
        // add point
        index.add(sample, i).unwrap();
    }
    index.build(hora::core::metrics::Metric::Euclidean).unwrap();

    let mut rng = thread_rng();
    let target: usize = rng.gen_range(0..n);
    // 523 has neighbors: [523, 762, 364, 268, 561, 231, 380, 817, 331, 246]
    println!(
        "{:?} has neighbors: {:?}",
        target,
        index.search(&samples[target], 10) // search for k nearest neighbors
    );
}

感谢 @vaaaaanquish 这个完整的纯 Rust 🦀 图片检索 example, 想了解更多可以点击 Pure Rustな近似最近傍探索ライブラリhoraを用いた画像検索を実装する

Python example [more info]

import numpy as np
from horapy import HNSWIndex

dimension = 50
n = 1000

# init index instance
index = HNSWIndex(dimension, "usize")

samples = np.float32(np.random.rand(n, dimension))
for i in range(0, len(samples)):
    # add node
    index.add(np.float32(samples[i]), i)

index.build("euclidean")  # build index

target = np.random.randint(0, n)
# 410 in Hora ANNIndex <HNSWIndexUsize> (dimension: 50, dtype: usize, max_item: 1000000, n_neigh: 32, n_neigh0: 64, ef_build: 20, ef_search: 500, has_deletion: False)
# has neighbors: [410, 736, 65, 36, 631, 83, 111, 254, 990, 161]
print("{} in {} \nhas neighbors: {}".format(
    target, index, index.search(samples[target], 10)))  # search

JavaScript example [more info]

import * as horajs from "horajs";

const demo = () => {
    const dimension = 50;
    var bf_idx = horajs.BruteForceIndexUsize.new(dimension);
    // var hnsw_idx = horajs.HNSWIndexUsize.new(dimension, 1000000, 32, 64, 20, 500, 16, false);
    for (var i = 0; i < 1000; i++) {
        var feature = [];
        for (var j = 0; j < dimension; j++) {
            feature.push(Math.random());
        }
        bf_idx.add(feature, i); // add point 
    }
    bf_idx.build("euclidean"); // build index
    var feature = [];
    for (var j = 0; j < dimension; j++) {
        feature.push(Math.random());
    }
    console.log("bf result", bf_idx.search(feature, 10)); //bf result Uint32Array(10) [704, 113, 358, 835, 408, 379, 117, 414, 808, 826]
}

(async () => {
    await horajs.default();
    await horajs.init_env();
    demo();
})();

Java example [more info]

public void demo() {
    final int dimension = 2;
    final float variance = 2.0f;
    Random fRandom = new Random();

    BruteForceIndex bruteforce_idx = new BruteForceIndex(dimension); // init index instance

    List<float[]> tmp = new ArrayList<>();
    for (int i = 0; i < 5; i++) {
        for (int p = 0; p < 10; p++) {
            float[] features = new float[dimension];
            for (int j = 0; j < dimension; j++) {
                features[j] = getGaussian(fRandom, (float) (i * 10), variance);
            }
            bruteforce_idx.add("bf", features, i * 10 + p); // add point
            tmp.add(features);
          }
    }
    bruteforce_idx.build("bf", "euclidean"); // build index

    int search_index = fRandom.nextInt(tmp.size());
    // nearest neighbor search
    int[] result = bruteforce_idx.search("bf", 10, tmp.get(search_index));
    // [main] INFO com.hora.app.ANNIndexTest  - demo bruteforce_idx[7, 8, 0, 5, 3, 9, 1, 6, 4, 2]
    log.info("demo bruteforce_idx" + Arrays.toString(result));
}

private static float getGaussian(Random fRandom, float aMean, float variance) {
    float r = (float) fRandom.nextGaussian();
    return aMean + r * variance;
}

Roadmap

  • Full test coverage
  • Implement EFANNA algorithm to achieve faster KNN graph building
  • Swift support and iOS/macOS deployment example
  • Support R
  • support mmap

Related Projects and Comparison

  • Faiss, Annoy, ScaNN:

    • Hora 的实现受到这些库的强烈启发。
    • Faiss 更侧重于 GPU 场景,Hora 比 Faiss 更轻(无重度依赖)
    • Hora 期待支持更多的语言,与性能相关的一切都会由 Rust🦀 实现。
    • Annoy 只支持 LSH (Random Projection) 算法。
    • ScaNNFaiss 不太用户友好,(例如缺乏文档)。
    • Hora is ALL IN RUST 🦀.
  • Milvus, Vald, Jina AI

    • MilvusVald 也支持多种语言,但作为服务而不是库
    • Milvus 是建立在一些库上的,比如 Faiss,而 Hora 是一个库,所有算法都是自己实现的

Contribute

We appreciate your help!

我们很高兴您的参与,欢迎任何贡献,包括文档和测试。 您可以在 GitHub 上创建 Pull RequestIssue,我们会尽快审核。

我们使用 GitHub 问题来跟踪建议和错误。

Clone the repo

git clone https://github.com/hora-search/hora

Build

cargo build

Test

cargo test --lib

Try the changes

cd examples
cargo run

License

The entire repository is licensed under the Apache License.