A course project in deep learning that explored the effects of different architectural ways of constructing a generative adversarial network. The main purpose of this project was to use the model presented in the paper "Semi-Supervised Learning for Face Sketch Synthesis in the Wild" [1] and tweak the construction of the generative model in order to observe the effects the tweaked models had on the sketches that were synthesized.
[1] Chaofeng Chen et al. “Semi-Supervised Learning for Face Sketch Synthesis in the Wild”. In: 2018. arXiv:1812.04929.