LsFlyt

Maybe the labels of font style in your training set are not continuous? What's more, follow this: issue https://github.com/ecnuycxie/DG-Font/issues/11#issuecomment-892472339

And another question, in VQ-VAE, the data are normalized to [-0.5, 0.5], but in the training phase 2, the content image (which feeds to the content encoder) is normalized [-1,...