Sunday, March 7, 2021

TensorFlow.js Blueprint App Step by Step

I describe how to prepare data that comes from API into TensorFlow.js Dataset structure, how to shuffle, normalize, one-hot-encode, and batch the data. Next, I go on model training and explain why fitDataset is recommended way to train a neural network in TensorFlow.js. At last, I show how to do inference and print the results. This app is built with React, but the same code can be reused with any JS toolkit/framework.

 

No comments: