-
Mapping UK Street Crime: An Interactive Data Visualisation Using Police Data
Using UK Police crime data, I built an interactive web app that allows users to visualize street-level crime data by clicking on the map or defining custom areas. This post covers data retrieval, processing and visualisation.
-
Predicting Bike Sharing Demand: A Step-by-Step Guide to Building a Regression Model
Using 2 years of weather and bike-sharing data, I built a machine learning model to predict the number of bike rentals in any given hour. This post covers the exploratory data analysis, data preprocessing, feature selection, and model evaluation.
-
When Not to Use Neural Networks
While neural networks are powerful, they aren't always the best tool for the job. This post explores cases where simpler models outperform them, with Python code snippets and experiments.
-
Seven Years Later: Refactoring the Code Behind My First Peer-Reviewed Paper
I revisit the code from my first peer-reviewed paper (and master's thesis), published seven years ago, to modernize it with the skills and knowledge I’ve gained since, improving its performance, readability, and maintainability.
-
Testing the new JaxMARL library
Testing JaxMARL, a cutting-edge library for multi-agent reinforcement learning, built on the high-performance JAX framework for efficient parallel computation.