Experiments

These experiments are my messy playground, not polished products.

The blog is where I write things up once I think I understand them. This is where things go before that. Some experiments became posts, some didn't, and some are still going.

These tend to be written in roughly the order I worked through them, which isn't always the order that would make sense in retrospect. There usually isn't a linear path through these topics. The canonical starting point of of these topics connects to everything else in too many directions to lay out as a sequence, and reaching the interesting part often means following a related piece of math for a while before doubling back. I group things logically after the fact when I can.

Most are interactive in some form, because there are ideas you can't develop a feel for without moving the parameters yourself. That's the line between an experiment and a post: when prose alone won't transmit the thing, it goes here.

Treat the section like someone's garage. Useful things and half-finished things sitting next to each other, some of it dusty, most of it still where I left it.

NLP Exploration: Topic Models

Using topic models in tomotopy to determine the topics of a corpus, the aspects discussed, and the relationships between topics.

Tags

Nlp Data Math Python

2 Trustworthy Alternatives to Improving Performance Lists in Track & Field

Improving meet performance lists in track & field using the previous performances of the competitors

Tags

Data Forecasting Math Running

Boosting Web Performance: Implementing K-Means Clustering with WebAssembly and Emscripten

Exploring complexities of optimizing web performance by implementing K-Means clustering algorithms using WebAssembly and Emscripten.

Tags

C++ Data Javascript Math

Neural Variational Document Models with PyTorch for Topic Extraction

Discover how Neural Variational Document Models, implemented using PyTorch, improve topic modeling and unsupervised learning in natural language processing. Learn about the architecture, training process, and applications of these latent variable models for text analysis and beyond.

Tags

Data Math Nlp Python

Streamline Hatching: A Programmer's Attempt at Computational Drawing

Sharing an attempt at computational hatching. Using structure tensors, edge tangent flow, and streamline integration to transform photographs into hand-drawn-looking illustrations.

Tags

Data Drawing Math Svg Webgpu

XDoG: Computation of Aesthetically Pleasing Lines

A deeper dive into modern approaches for generating aesthetically pleasing lines based on difference-of-Gaussians (DoG) edge detections.

Tags

Data Drawing Math Webgl Webgpu