Welcome to “A Beginner’s Guide to Machine Learning in JavaScript”! In this series, I’ll teach the concepts behind machine learning using the ml5.js library. #machinelearning #javascript #ml5 #p5js.

In this video, I use the “pre-trained” MobileNet model to classify the content of an image. #machinelearning #mobilenet #imageclassification #ml5 #p5js.

In this video, I discuss image classification with MobileNet using real-time video. #machinelearning #mobilenet #imageclassification #ml5 #p5js #webcam.

In this video, I cover object detection in ml5.js with the COCO-SSD pre-trained model.

In this video, I discuss the process behind “transfer learning” with ml5’s feature extractor. #imageclassification #ml5 #p5js.

In this video, I use ml5’s feature extractor to train a machine learning image classifier with my own images. #machinelearning #classification #ml5 #p5js.

In this video, I use ml5’s feature extractor and re-train the MobileNet model to output a “regression” controlling a slider. #machinelearning #regression #ml5 #p5js.

In this video, I discuss how to save and load an image classification model (trained using the “feature extractor”) in ml5.js.

This video covers “KNN Classification.” In this first part, I begin the process of building a “teachable machine”-style image classifier with ml5.js.

This video series covers “KNN Classification.” In this second part, I train the KNN image classifier using the ml5.js library.

In this last part, I demonstrate how to save and load the KNN Classification model. (The code linked below will work with 0.2.1 of ml5 and will be updated when the new 0.2.2 release comes out.)

This video covers how to train a neural network machine learning model with real-time interactive data in ml5.js. The example demonstrated uses the mouse as input and performs classification (the assigned label is a musical note).

This video takes the sketch from the previous video on training your own ml5.js neural network, and shows how to save the data collected in a JSON file which can be loaded again later.

Here I show how you to save a model that has been trained in ml5.js. You can then load (aka “deploy”) that model later (“pre-trained model”) in another sketch!

In this video I cover how to train a neural network to perform a “regression” task (rather than classification). The result is a continuous numerical output (frequency value) instead of a categorical one (specific note).

In this video I cover pose estimation: finding the keypoints of person’s pose and skeleton using the pre-trained machine learning model PoseNet (in JavaScript with p5.js and ml5.js).

This tutorial combines PoseNet and ml5.neuralNetwork(). I use the output of the pre-trained model (the “pose” itself) as the inputs to an ml5.js neural network classifier.

This tutorial builds on the previous video combining PoseNet and ml5.neuralNetwork(). Once again, the output of the pre-trained model (the “pose” itself) is the input to an ml5.js neural network. However, this time the final output is a regression (3 continuous values) instead of classification.

This tutorial builds on ml5.neuralNetwork() videos examining raw pixels as inputs to a neural network. This sets the stage for a discussion on convolutional neural networks.

This video covers the architecture of a Convolutional Neural Network, focusing on the concept of “filters”.

Part 2 of this video on Convolutional Neural Networks covers on the concept of “max pooling.”

Welcome back! In this video, I update the previous classification example (with pixels) and incorporate a convolutional neural network in ml5.js!

Quick, draw! In this video, I demonstrate how to use the DoodleNet pre-trained machine learning model to classify drawings of cats, strawberries, and other common doodles in JavaScript (with p5.js and ml5.js).