CycleGAN An simple implementation of CycleGan using pytorch. Notes 1. 在做马和斑马的实验时,马->斑马 很容易,但斑马->马却相对困难。 2. 白马很难转换。 2. 在桔子和苹果数据集上效果挺好。 References CycleGAN paper: https://arxiv.org/abs/1703.10593