Nocturnal Parakeet Monitor


    1.2.2 • Public • Published

    Teachable Machine Node



    Teachable Machine Node empowers you to load any image classification model trained with Google's Teachable Machine tool in a Node.Js project.

    Teachable Machine makes AI easy for everyone, by offering a fast and fun way to train a real TensorFlow.js Machine Learning Models without any coding required. You can train the computer to recognize images, sounds, & poses, using your camera or your own dataset. Check The Awesome Teachable Machine List full of useful resources and amazing projects to gain some cool ideas.

    For now, Teachable Machine Node holds support only for image models, but we won't stop here. Check out the Roadmap of what comes next!


    • Install using npm
    npm install @sashido/teachablemachine-node
    • Install using yarn
    yarn add @sashido/teachablemachine-node


    Create you own Model with Teachable Machine

    1. Gathering samples is the fundamental first step to your Teachable Machine Model. Use your camera to collect data or upload some preselected images.

    2. Train your Teachable Machine Image Model.

    Check the Advanced option for further insights on the model performance and accuracy. Once certain it returns valid results, hit the Export option.

    1. Make sure that you select Tensorflow.js format when exporting. That way your model will be uploaded (for free) and you will receive an access URL.

    1. Pass the URL to the teachablemachine-node to load the model. Next, let it have the image URL and call classify to get the predictions.

    Play around with our pre-trained 'Is It A Dog?' model.

    SashiDo's team is full of notorious animal lovers and no wonder the sample model was trained to recognize dog pics from other images. 😊 We've collected a dataset of more than 2000 images of dogs, cats, horses, other animals, people and everyday objects and uploaded them into two different classes with Teachable Machine.

    A big shoutout to as a great platform for free photos. Check it out in case you're just starting and do not have raw data yet.

    You can load our Is_It_A_Dog model using this URL and following the example below:



    Here's a quick example of how to load the model in your project.

    const TeachableMachine = require("@sashido/teachablemachine-node");
    const model = new TeachableMachine({
      modelUrl: ""
      imageUrl: "",
    }).then((predictions) => {
      console.log("Predictions:", predictions);
    }).catch((e) => {
      console.log("ERROR", e);


    And the alternative for all Express fans.

    const express = require("express");
    const TeachableMachine = require("@sashido/teachablemachine-node");
    const model = new TeachableMachine({
      modelUrl: ""
    const app = express();
    const port = 3000;
    app.get("/image/classify", async (req, res) => {
      const { url } = req.query;
      return model.classify({
        imageUrl: url,
      }).then((predictions) => {
        return res.json(predictions);
      }).catch((e) => {
        res.status(500).send("Something went wrong!")
    app.listen(port, () => {
      console.log(`Example app listening at http://localhost:${port}`);


    In the long run, we will add more options, so you can train and load all kinds of Teachable Machine Models.

    1. Add support for Pose Models.

    2. Add support for Audio Models.

    3. Add support for Gifs.

    4. Add support for Videos.

    We would love to have your opinion which's the one you would like to see supported first. Don't be shy and drop us a line at


    Contributors of any kind are welcome. Share your awesome improvements in a pull request and join our mission to make Machine Learning more affordable & accessible!


    Copyright © 2020, CloudStrap AD. See LICENSE for further details.


    npm i @sashido/teachablemachine-node

    DownloadsWeekly Downloads






    Unpacked Size

    22 kB

    Total Files


    Last publish


    • sashidoteam
    • pivanov