justinpinkney / awesome-pretrained-stylegan3 Goto Github PK

I would like to share our datasets and models...

should i write an issue for each individually, or is it ok like that?
much love,
ed

• Name: Lucid Layers Stylegan3 Datasets and Models collection
• Source URL (e.g. blog post, tweet etc.): https://github.com/edstoica/lucid_stylegan3_datasets_models/blob/main/README.md
• Resolution: 256-1024
• Config (T or R): T
• Notes (e.g. training details, gamma, FID, kimg, anything else): default, no mirror

If you'd like to some sample image please attach them to the issue.

I have been experimenting with using IPFS for caching large models so I don't go over the quota limits of Google Drive or Mega.

The idea is that models are often shared by researchers via Google Drive links which have transfer limits and are not reliably online. IPFS provides a great decentralized solution to hosting data that can be downloaded via regular web links.

I pinned the two models in the repository to my server if there is interest in using IPFS links.

https://ipfs.io/ipfs/QmZkrYwEUnykVQJfJw3opTj1HfdNUCm87amsR3LHp1QnuV/wikiart-1024-stylegan3-t-17.2Mimg.pkl
https://ipfs.io/ipfs/QmZkrYwEUnykVQJfJw3opTj1HfdNUCm87amsR3LHp1QnuV/lhq-256-stylegan3-t-25Mimg.pkl

https://shkspr.mobi/blog/2022/04/this-bench-does-not-exist/

I've been following your updates and I'd like to ask a few questions about GANs，1.How to train a model with large resolution, I see it is 256x256 2. I want to edit some properties（like rain, snow, forests, mountains, etc.） of the Landscapes，Which project can be used to achieve

Looking forward to your reply, thank you again