Home

Awesome

annoy-node

Build Status

Node bindings for Annoy, an efficient Approximate Nearest Neighbors implementation written in C++.

Version 4.0.0 requires Node 14 or Node 16 and does not yet work on 18.

Status: Tests pass, including one that loads 3 million vectors, but API coverage is not complete. Run on OS X and Linux with Node 8, 10, and 12. Not tried on Windows yet. Support for Node 6.3 and 4.6 ended at version 2.0.1 of this package.

All of the Python API methods are implemented. The names are camel cased, JavaScript-style.

There are a few minor differences in behavior:

Installation

On Linux, if you don't already have Python 2.7 and g++ 4.8, you need to install them. Here's how you do it on Ubuntu:

(sudo) apt-get install python2.7
(sudo) apt-get install g++-4.8
npm config set python /path/to/executable/python2.7

Then, symlink g++ somewhere it can be found:

ln -s /usr/bin/g++-4.8 /usr/local/bin/g++

On OS X, they should already be there.

Then:

npm install annoy

Usage

var Annoy = require('annoy');
var annoyIndex1 = new Annoy(10, 'Angular');

annoyIndex1.addItem(0, [-5.0, -4.5, -3.2, -2.8, -2.1, -1.5, -0.34, 0, 3.7, 6]);
annoyIndex1.addItem(1, [5.0, 4.5, 3.2, 2.8, 2.1, 1.5, 0.34, 0, -3.7, -6]);
annoyIndex1.addItem(2, [0, 0, 0, 0, 0, -1, -1, -0.2, 0.1, 0.8]);
annoyIndex1.build();
annoyIndex1.save(annoyPath);

read();

function read() {
  var annoyIndex2 = new Annoy(10, 'Angular');

  if (annoyIndex2.load(annoyPath)) {
    var v1 = annoyIndex2.getItem(0);
    var v2 = annoyIndex2.getItem(1);
    console.log('Gotten vectors:', v1, v2);

      for (var i = 0; i < v1.length; ++i) {
        sum.push(v1[i] + v2[i]);
      }

      var neighbors = annoyIndex2.getNNsByVector(sum, 10, -1, false);
      console.log('Nearest neighbors to sum', neighbors);

      var neighborsAndDistances = annoyIndex2.getNNsByVector(sum, 10, -1, true);
      console.log('Nearest neighbors to sum with distances', neighborsAndDistances);
  }
}

Development

npm install -g node-gyp
node-gyp rebuild

Run eslint . and make prettier before committing.

Tests

Run tests with make test.

You can also run tests individually:

There is also a big-test target that is not a dependency of the test target. It loads about 3 million 300-dimension vectors. It takes about six minutes to run on good-for-2016 hardware. Before you can run it, you need to download GoogleNews-vectors-negative300.bin to tests/data.

Then, you can run make tests/data/GoogleNews-vectors-negative300.json, which takes a while, and gets the test data ready for the big test. (See comment about running that in the Makefile.) Then, make big-test.

Contributors

Thanks to:

License

The MIT License (MIT)

Copyright (c) 2016 Jim Kang

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.