Colin Fraser

Computing big Fibonacci numbers using the Golden Ratio, abstract algebra, and S3 classes in R

Programming

R

Math

This post is a bit interdisciplinary. It stems from a back and forth that I had on X about ways of computing the Fibonacci numbers. I learned something new in that…

Wrapify

An R package for making API wrappers declaratively

R

R Package

I often find myself needing to write functions to make API calls in R. This tends to involve a lot of boilerplate and repetition. The {wrapify} package is an attempt to…

Estimating how many there are of something when you can’t see them all perfectly

A brief introduction to quantification

Essay

Statistics

I wrote a tweet recently complaining about how it’s hard to estimate how many of what kind of posts there are.

Who are we talking to when we talk to these bots?

Essay

LLMs

An attempt to figure out who exactly ChatGPT is, if indeed it is anyone at all.

Making Decisions With Classifiers

Essay

ML

When you’re building a classifier, you face a tradeoff between precision and recall. You can have a high precision classifier which is usually right when it flags a positive, or you can have a high recall classifier which catches most of the positives, and often there is a spectrum in between. This post is about how to choose a point on that spectrum.