Aden Kneebone
I'm a fresh QUT graduate with a Bachelor's in Computer Science who has skills in multiple forms of development from desktop software and web applications, to APIs and intelligent systems.
Projects
MDANI
Web Interface
MDANI is a proof of concept application proposed by the Bureau of Meteorology that enables users to write natural language queries to retrieve information from a gridded data lake. Undertaken as our university capstone project, I was responsible for creating the web interface, along with training the Named Entity Recogniser used to extract information out of written queries (see: NLPUtils).
JavaScript React Semantic UI
NLPUtils
spaCy Wrapper
NLPUtils is a piece of desktop software that acts as a wrapper for some of the command line methods found in the spaCy library. The software allows users to create document data sets, train Named Entity Recognisers, easily modify parameters and hyperparameters, and switch workspaces between multiple models.
Python Tkinter spaCyUnown GAN
Reusable REST API and Web Interface for TensorFlow GANs
Unown GAN is a personal investigation into Generative Adversarial Networks, their training processes, and their utilisation as a web service. The project aims to showcase the interpolation between points in the latent space of a GAN and demonstrate the bulk generation of images via a REST API and Web Interface.
Python TensorFlow Keras Flask JavaScript React
