Source: https://github.com/hlud6646/storeSimul3
Dashboard: https://hlud.xyz/store

In an actual application, there’s a good chance that the interaction with the database will take place in a format which is different to simply executing statements inside a console. Like, it’s all well and …

Disclaimer: Do NOT use anything you find in here to help you choose a mushroom!

Source: https://github.com/hlud6646/data-practice/tree/main/07-Mushrooms

This notebook explores the tradeoff between type 1 (false positive) and type 2 (false negative) errors. In this example, we are attempting to …

Source: https://github.com/hlud6646/us-accidents

An interviewer asked me ‘What kind of exposure to SQL have you had?’ and I felt that my answer of ‘one uni subject and some hacker-rank practice problems’ was insufficient. Next time I will tell them about this.

I wanted to do …

Source: https://github.com/hlud6646/data-practice/tree/main/04-Neural-Networks

I’m filling in some gaps from my masters and one of the side-quests that I never completed was coding up a neural network from scratch. These models have the reputation of being a ‘black box’ where data …

It’s well known that vectorised operations should be preferred in numpy/pandas for reasons of performance. A less common reason is that it actually makes the code easier to read. For a simple example imagine that you are doing a multivariate linear regression to transform an $n$ dimensional …

Source code: https://github.com/hlud6646/grep-haskell

Grep seems like such an innocent tool, but it does a lot. Rebuilding it was a hard project for me and I am proud of what I have learned along the way. This post is a summary of those things. From a bird’s eye perspective:

• Developing an …

There are only two hard things in Computer Science: cache invalidation and naming things. — Phil Karlton

Say what you want about Haskell, Hoogle is an amazing tool. If you don’t know, it’s simply a search engine for functions where your search term is a function signature. That …

This notebook demonstrates the advantage of working with a transformed parameter when making a quadratic approximation (i.e. an approximation to the probability density function by a Gaussian distribution). If, for example, the parameter to estimate is a proportion and therefore bound to the …

I took a course on multivariate statistics and the instructor was very committed to the idea of demonstrating the quality of any claimed asymptotic results by running simulations. He did everything in R which I do not like and I took the opportunity to see what Julia was like for this kind of work. …

A young student of mine told me that they had started a new topic in class, namely symmetries of plane figures. These were translations, reflections and rotations of polygons and some irregular shapes. This was a good opportunity to do an exercise involving symmetries that I’ve done a few …