This repo contains the submission for CW2 of Deep Learning module from Department of Computing. It achieved a score of 99%.
The core knowledge tested was the implementation of generative models, including Variational Autoencoders (VAE) and Deep Convolutional Generative Adversarial Networks (DCGAN).
Please refer to the jupyter notebook to find the full experimentation and techniques employed (Top-K algorithm, One-sided label smoothing etc.)