Hacknight Challenges

We run hacknights in the Lovell seminar room of the Alan Turing building with free pizza and prizes. The aim is to help people explore different machine learning tooks. Below we list the previously run hacknights along with links and download buttons.

Deep Learning to Predict Hurricanes

In this hacknight we provide an introduction to neural networks using keras. The goal of the hacknight is to predict hurricanes using weather data.

Radar data for Hurricane Maria in Puerto Rico

On the 18th of September 2017 a hurricane made landfall in Puerto Rico. Torrential rain and high winds caused extensive damage to infrastructure and led to widespread flooding https://en.wikipedia.org/wiki/Hurricane_Maria Using radar data provided by the European Space Agency Sentinel 1 Mission, can we detect the damage/flooding which occured?

Greenhouse gas and Dietary choices Open source Toolkit

We'll be developing tools for estimating the contribution of food to greenhouse gas emissions, as part of the Greenhouse gas and Dietary choices Open source Toolkit (GDOT) project funded by N8 Agrifood. We have made a set of tools for you to play with which shows a breakdown of emissions and nutrition for a set of foods you specify e.g. what you ate yesterday, or your favourite recipe, or the average UK diet. You could play with different foods to try to find a nutritious and tasty low carbon diet, work on improving the visualisation of this complex dataset or think of neat optimisation tools.