27 posts with this tag
Over the past two years, since the initial release of ChatGPT in November 2022, we've been in a hype cycle around large language models, or LLMs. Due to the very human "feeling" of LLM outputs, a lot of people have been convinced that these models …
In our previous blog post we discussed how to implement the Minkowski distance formula in a couple of functions which relied heavily on for loops. On our full data, this lead to a processing time of over an hour. With some simple tricks in NumPy …
In the last blog post, we managed to shave a bit of time off our calculation of the Minkowski distance by using vector subtraction. Instead of calculating the difference between each pair of vectors elementwise using a loop, we were able to take …
In the last blog post, we discussed how to calculate the Manhattan and Euclidean distances from first principles. However, in that post, we did a very manual implementation for a single pair of vectors, which would not generalise well to more than …
If you had to invent a machine learning algorithm from scratch, what would be some of the ways you'd find patterns in your data? One idea that you might have come up with is to assume that data points that are "close" to each other are similar, and …
