bloc97

Hi, unfortunately there isn't any reference for the "ad hoc" method I used to compensate for CFG, but I can give a quick explanation, if you have more questions we...

Hi, thanks for catching the mistake! The official repo code was released after mine, and I probably misunderstood this part of the algorithm from the paper... I didn't have time...

@mikekasprzak Our team's game was affected by someone abusing the system and I think we sent you that e-mail. I think the current smart balance system can be abused by...

Is it possible to find who liked our comments so we can ask him to remove those likes? I hope it was unintentional and it would be nice for us...

Hi, after skimming through the paper I think it shouldn't be too difficult to implement the guidance operators described in the paper. If I understood it correctly, they apply transformations...

Hi, the dataset can be found here: https://github.com/bloc97/SYNLA-Plus However, it is used as an placeholder for that example, it is better to use real images for training. That dataset is...

You can use any scheduler that you want. I think LMSD was chosen because it was the one giving the best results back then...

Hi, I'm not too sure about the code difference between my implementation and the original, as this repo's code is not a modification of the authors' code but an independent...

I have to find where I put the script to originally generate the dataset, but the dataset generation is very slow compared to the speed of training. It is usually...

> I noticed that you trained the denoising model with 4 datasets, but trained the upsacler with only SYNLA-Dataset, why? The downscaling operation mainly affects salient features (i.e. edges, textures),...