search

POSTECH-CVLab / PyTorch-StudioGAN

StudioGAN is a Pytorch library providing implementations of representative Generative Adversarial Networks (GANs) for conditional/unconditional image generation.
https://github.com/MINGUKKANG
Other
3.43k stars 343 forks source link

Question on the checkpoint for Stylegan2 on FQ #167

Open lizaitang opened 2 years ago

lizaitang commented 2 years ago

Excuse me, it seems that the google drive does not contain the checkpoint for StyleGAN on FQ dataset? How can I find it.

Hzzone commented 2 years ago

Same problem here. I remember that I have seen the pre-trained models on FFHQ a few weeks ago.

asagar60 commented 2 years ago

Yes, i can see links to features and moments but not checkpoint

mingukkang commented 2 years ago

Sorry.

We deleted pretrained weights temporarily for internal reasons.

However, I can share some weights for only research purpose.

I'll share a link for StyleGAN2-FFHQ weights tomorrow evening.

Sorry for the inconvenience.

Hzzone commented 2 years ago

@MINGUKKANG Thanks very much for your kindful share!

mingukkang commented 2 years ago

Hi,

Here I share the pretrained StyleGAN2 [weight] on FFHQ 1024x1024.

Thank you.

lizaitang commented 2 years ago

Thank you so much! By the way, can we achieve conditional GAN on FFHQ dataset?

Hzzone commented 2 years ago

Thank you so much! By the way, can we achieve conditional GAN on FFHQ dataset?

No, there are no official human labels for FFHQ dataset, but you can do this by borrowing the labels from other research if available.

lizaitang commented 2 years ago

Thanks, to achieve this do I need to retrain with the labels?

Hzzone @.***>于2022年9月30日 周五11:50写道：

Thank you so much! By the way, can we achieve conditional GAN on FFHQ dataset?

No, there are no official human labels for FFHQ dataset, but you can do this by borrowing the labels from other research if available.

— Reply to this email directly, view it on GitHub https://github.com/POSTECH-CVLab/PyTorch-StudioGAN/issues/167#issuecomment-1263070307, or unsubscribe https://github.com/notifications/unsubscribe-auth/ANWXMBNSOGSHIUGIEFB3Q2DWAZPQ5ANCNFSM6AAAAAAQKA4NPM . You are receiving this because you authored the thread.Message ID: @.***>

mingukkang commented 2 years ago

We provide all checkpoints we used here.

Thank you for your patient.