Front-end project implementing handwritten digit recognition based on tensorflow.js - a library that allows you to develop Machine Learning Models in JavaScript and use Machine Learning directly in the browser or in Node.js
This project uses predefined high accuracy data model (freely available on github)
The data model is generated with python script and is based on MNIST dataset that is a dataset of 60,000 small square 28×28 pixel grayscale images of handwritten single digits between 0 and 9
The data model is converted to be used by tensorflow.js
After that is loaded and used by tensorflow.js
