Keras/Tensorflow
implementation of the decoder component from: Attention Is All You Need and example usage by means of a language model trained on a small baking cookbook.
Keras/Tensorflow
character-level language model trained on the British National Formulary (BNF) A-Z of clinical medicines and a minigame to challenge it.
Keras/Tensorflow
implementation of the 1cycle learning rate policy as originally described in the paper: Super-Convergence: Very Fast Training of Neural Networks Using Large Learning Rates.
Pytorch
implementation of the Learning Rate Range Test and Cyclical Learning Rates (CLR) as originally described in the paper: Cyclical Learning Rates for Training Neural Networks.
Hackathon: Helping South East Water manage bursts. Working in a team alongside a friend, our solution got a special mention from the judges and we were awarded with a small cash prize - check out our presentation slides.
My solutions to the labs and exercises from the classic textbook: An Introduction to Statistical Learning with applications in R. I follow in Python rather than R, with heavy use of: numpy
, pandas
, sklearn
, matplotlib
, seaborn
, patsy
and statsmodels
.
Numpy
implementations a handful of statistical learning algorithms that appear in the classic textbooks: An Introduction to Statistical Learning and The Elements of Statistical Learning.
My solutions to the assignments from Imperial College's Mathematics for Machine Learning coursera course. FWIW - you get a certificate of completion, mine is here.
A media libray in the cloud for the performing arts.
An open source iOS framework that makes it easy to create apps for medical research or for other research projects. I'm a top 5 contributor.
An events platform for the Institute of Physics.
Stream performances of classical music and interact with them on your tv.
Color in Millie Marotta illustrations on your iPad with your apple pencil.
An app for iPhone and iPad, including score syncing and on-the-fly performance switching.