Browser-based Models with TensorFlow.js

Introduction to TensorFlow.js is the first module in the Specialization. You'll learn how to train and run machine learning models in any browser using TensorFlow.js. The module also covers techniques for handling data in the browser and culminates in a computer vision project that recognizes and classifies objects from a webcam.

Image Classification In the Browser is the second module, focusing on creating a Convolutional Net with JavaScript, visualizing the training process, and utilizing tf.tidy() to save memory. The module also includes building an MNIST Classifier and a Fashion MNIST Classifier.

The third module, Converting Models to JSON Format, delves into pre-trained TensorFlow.js models, toxic classification, MobileNet, and converting models to JavaScript. This module also covers image classification using MobileNet and a linear model.

Transfer Learning with Pre-Trained Models, the final module, explores retraining the MobileNet model, capturing and training the network with captured data, and performing inference. It culminates in the application of the knowledge gained through building a Rock Paper Scissors project.