November 2019
Intermediate to advanced
296 pages
7h 52m
English
Since TensorFlow.js is written in TypeScript, it can be naturally integrated with existing web technologies. This means that a machine learning model that's been written with TensorFlow.js can be run in a web browser without any modifications having to be made. This ensures that we can make use of the rich user interface that the web browser provides. Actually, TensorFlow.js itself has its own set of utilities so that we can import the Document Object Model (DOM) element of an image as a tensor, such as a canvas. This makes it easy for us to run the application with the data that's available on the internet. The teachable machine is a good example of the power of integrating with the web. It uses ...
Read now
Unlock full access