rinongal

We used the directions from this paper: https://github.com/mit-han-lab/anycost-gan I think they have a direction for eyeglasses as well, but you'll have to extract it from their file and save it...

Hi, Unfortunately it's been almost 3 years and I don't have access to the machine with the original code. We just used the clustering metric code from [Ojha et al](https://github.com/utkarshojha/few-shot-gan-adaptation/blob/main/feat_cluster.py)...

These 4 images are GAN generated. This isn't really a style-transfer method, so there's no real notion of content and style images. It's a zero-shot finetuning method for GANs. If...

On the first error: I see that torchvision was updated to use "value_range" instead of "range", you'll have to either downgrade to an older version of torchvision from before this...

Hi, It's been a while since I touched this code, but I *think* you should be able to get this done by modifying generate_videos.py to load multiple generators and pass...

Hey. Thanks for this. Since it's already pushed I understand it's not urgent. I'll have a look over the weekend and see that nothing in replicate is broken :)

Do you have nvidia docker installed?

Thanks! We did not attempt any other wavelet transformations, but we certainly think that it can be an interesting approach to take! Actually, one of the things we discussed while...

The step_size parameter will control strength when doing multi-attribute editing, yes. If you want to make an adjustment to the single-mode sample you'll need to change this line: https://github.com/yzliu567/sc-gan/blob/610575627fd4732e84df8389722bdafc3531f4ad/pti/conditional_pti.py#L152 so...

For the stable diffusion version you need any of the compvis-based models, so 1.4 and 1.5 will both work fine, as will models tuned from those base models (e.g. protogen)....