If you follow machine learning, you've probably heard of TensorFlow. Originally developed by researchers and engineers working on the Google Brain team for numerical computation using data flow graphs, it has historically been limited to desktop machines and cloud servers. At least, until March of this year!
TensorFlow.js takes what you know and love about TensorFlow and moves it to the browser (and Node.js), allowing you to run, train, and develop ML models anywhere.
In this presentation, we'll cover the basic terminology and concepts TensorFlow is built upon. After we've got our bearings, we'll walk through examples using pre-trained ML models and ultimately develop our own ML model in the browser!
Josh is on the data team at UnitedMasters, providing music artists with premium distribution services and facilitating unique partnerships between artists and the world’s biggest brands. Josh frequently experiments with the intersection of technology and the arts when he isn't climbing, running, or spending time with his wife.