Introduction to Machine Learning with TensorFlow.js
A compact and gentle introduction to Machine Learning in JavaScript. Learn by building 4 applications from scratch, no previous experience required.
👀 Read online 📕 Get the bookDescription
Many exciting things are happening with AI, from which, until recently, JavaScript developers were largely shut out. However, things are changing, now if you can do npm install @tensorflow/tfjs
you can do AI.
This absolute beginner course takes someone with no knowledge of Machine Learning and teaches them the basics.
The course will teach how to use the popular TensorFlow.JS library, a complete
You’ll learn:

What are Neural Networks and how is it related to Machine Learning?

What is TensorFlow.js and how to use it?

The essential mathematics.

How to build and train a neural network to solve regression and classification tasks.

How to find, convert, load and use pretrained models.
All will be explained in baby steps with a mix of 50% lecture and 50%
Taught by Asim Hussain,
Program outline/syllabus

Introduction
We will cover an overview of Machine Learning and Neural Networks as well as a history of TensorFlow and TensorFlow.js.

Project 1: Using a pretrained model
In this first project you will learn how to use a pretrained model and build your first
AIpowered application. 
TensorFlow
Now we will dig deeper into TensorFlow itself we’ll cover what Tensors are, how to create them with TensorFlow.js and how to perform basic mathematical calculations using Tensors.

Optimization
In this section we will explain the core function of TensorFlow and what makes up the field of Machine Learning, Optimization. We will learn what a loss function is and how to use TensorFlow.js to optimise some values based on the loss function.

Project 2: Linear & Polynomial regression
In this lecture we will cover what regression is and when would you use it, why we start with regression and how to build your first regression model.

Project 3: Recognizing handwritten numbers
In this section we will start using the layers API from TensorFlow.js and build a much more sophisticated application that recognizes handwritten digits. We will then use the same problem and solve it using a variety of different ML algorithms.

Project 4: Transfer Learning
Finally we bring all that knowledge together into Transfer Learning, we will take a pretrained model and train a new model on top of it. Transfer Learning is one of the fastest and least computationally intensive ways to make use of Machine Learning in JavaScript.