Trained Conditional Vanilla GAN and DCGAN models on the MNIST dataset to generate digit images based on user-selected input.