Making art out of the learning process of an algorithm

tSNEJS is an implementation of t-SNE machine learning algorithm for visualization. Check out the forked repo https://github.com/karpathy/tsnejs. The main project website also has a live example and more description.

t-SNE is a visualization algorithm that embeds high-dimensional data for visualization in a low-dimensional space of two or three dimensions. Specifically, it models each high-dimensional object by a two- or three-dimensional point in such a way that similar objects are modeled by nearby points and dissimilar objects are modeled by distant points with high probability. If you have some data and you can measure their pairwise differences, t-SNE visualization can help you identify clusters in your data.

In this project, I draw lines between data points and as the algorithm runs, it creates pieces of arts which you can move inside of.

• Modified t-SNE code using D3 to draw lines

• JavaScript
• D3
• t-SNE