Skip to content


Folders and files

Last commit message
Last commit date

Latest commit



6 Commits

Repository files navigation

MNIST Classifier - A WebAssembly BLAS Demo

A handwritten digit classifier demonstrating BLAS routines running in a web browser using WebAssembly.

A multilayer perceptron network is used for the modeling and classification of digits. The pre-training of model weights has been performed ahead of time using Scikit-learn, and the classification of digits runs interactively using JavaScript and WebAssembly.

The MNIST database provides the source data used to train the model. Classifier output is shown using Observable Plot.

Interactive Website

Draw a digit from 0-9 in the box and the classifier will try to label the handwritten digit. The resulting relative probabilities will be shown in a plot on the right.