nearest.js

/**
 * FastText.js
 * @author Loreto Parisi (loretoparisi at gmail dot com)
 * @copyright Copyright (c) 2017-2019 Loreto Parisi
*/

"use strict";

(function () {

    var MODELS_ROOT = __dirname + '/models';

    var Util = require('../lib/util');
    var FastText = require('../lib/index');
    var ft = new FastText({
        predict: {
            mostlikely: 10,
            normalize: true
        },
        loadModel: MODELS_ROOT + '/sms_model_w2v.bin' // must specifiy filename and ext
    });

    function getRandomWord(arr) {
        var word = '';
        while (word.length <= 3) {
            word = Util.randomElement(Util.randomElement(arr).split(/[\s.]/));
        }
        return word;
    }

    var samples = [
        "You have WON a guaranteed �1000 cash or a �2000 prize. To claim yr prize call our customer service representative on 08714712379 between 10am-7pm Cost 10p",
        "Sounds better than my evening im just doing my costume. Im not sure what time i finish tomorrow but i will txt you at the end"
    ];


    // load unsupervised model
    ft.loadnn()
        .then(labels => {
            // find Nearest Neighbor words
            var word = getRandomWord(samples);
            console.log("find Nearest Neighbor of \"%s\"", word);
            return ft.nn(word)
        })
        .then(labels => {
            console.log(JSON.stringify(labels, null, 2));
            var word = getRandomWord(samples);
            console.log("find Nearest Neighbor of \"%s\"", word);
            return ft.nn(word);
        })
        .then(labels => {
            console.log(JSON.stringify(labels, null, 2));
            ft.unload();
        })
        .then(done => {
            console.log("model unloaded.");
        })
        .catch(error => {
            console.error("predict error", error);
        });

}).call(this);