Koutilya PNVR

Hi @yzou2, do you mind explaining why the range of [-128, 128] is used and not [-0.5, 0.5]?

This is a good question. I don't understand where in the stable diffusion code does it take into account the random dropping of condition (20%) and more over its not...

https://github.com/cientgu/VQ-Diffusion/blob/fe79083818b47d4d376ab9579ec19cba2a43c3cb/image_synthesis/modeling/transformers/diffusion_transformer.py#L267 More precisely, this is the code line from the Improved_VQ-Diffusion branch. The cf_predict_start function is not defined in the DiffusionTransformer class. Is it the same as the one from...

Also is the reported 93.2% accuracy on validation dataset or on the training dataset?