Handwritten digit recognition with CNNs by TensorFlow

Saitama University
Thu, Apr 13, 5:45 PM (JST)

About this event

What you will build: You will make a webpage that uses TensorFlow.js to train a model in the browser. Given a black and white image of a particular size it will classify which digit appears in the image. The steps involved are:

- Load the data.

- Define the architecture of the model.

- Train the model and monitor its performance as it trains.

- Evaluate the trained model by making some predictions.

What you'll learn:

- TensorFlow.js syntax for creating convolutional models using the TensorFlow.js Layers API.

- Formulating classification tasks in TensorFlow.js

- How to monitor in-browser training using the tfjs-vis library.

What you'll need

- A recent version of Chrome or another modern browser that supports ES6 modules.

- A text editor, either running locally on your machine or on the web via something like Codepen or Glitch.

- Knowledge of HTML, CSS, JavaScript, and Chrome DevTools (or your preferred browsers devtools).

- A high-level conceptual understanding of Neural Networks. If you need an introduction or refresher, consider watching this video by 3blue1brown or this video on Deep Learning in Javascript by Ashi Krishnan.