Generating EMNIST Images: A Comparative Study of Various GAN Models

As part of a school project, I explored the exciting realm of Generative Adversarial Networks (GANs) by experimenting with several models, including DCGAN (Deep Convolutional GAN), CGAN (Conditional GAN), and WGAN (Wasserstein GAN).

My focus was on hypertuning each model to optimize their performance, allowing me to analyze the impact of various hyperparameters on the quality of generated outputs. This in-depth experimentation not only enriched my understanding of GAN architectures but also sharpened my skills in model evaluation and optimization. I am proud to have attained distinctions for this project, which further fueled my passion for advanced machine learning techniques