Learn to build and deploy a machine learning classification model in JavaScript using ml5.js. Follow the step-by-step process of data collection, model training, and prediction for a shape classifier neural network. Generate datasets, load images, create and train neural networks, and test the model with various input methods including mouse-drawn shapes and webcam input. Explore concepts like convolutional neural networks, dataset improvement, and model saving. Gain practical experience in applying machine learning techniques to image classification tasks using popular libraries and tools such as Processing, p5.js, and ml5.js.
Overview
Syllabus
Introduction.
Generating the dataset in Processing.
Loading images in p5.js.
Create a ml5 neural network.
Adding the data.
About training the model.
Test training.
Training with full dataset.
Improving the dataset.
Saved model.
Separate sketch for prediction.
Loading the model.
Testing the model with shapes drawn in p5.js.
Mouse drawn shapes.
Using a webcam and a notebook.
Wrap up.
What's next?.
Taught by
The Coding Train
