encoder4editing

encoder4editing copied to clipboard

Thanks for your great work! The results are amazing! However, I found when I only use the inference.py with pretrained model , it still needs a GPU there, is it...

jeffryWillam

Hi @omertov, Your research is so awesome and thanks for sharing your code. I am using the Car encoder, but I realize that the training dataset for the car encoder...

qilimk

Could you provide me with the code for calculating SWD?

1

Thank you very much!

wtybest

Releasing pretrained model for cat inversion

1

Thanks for all your hard and wonderful works here. I am wondering if you have plan to release the pretrained model for cat inversion. Thanks!

sr-jeon

hello！Some questions about models/styleGAN2/op extension！

2

I found that the op extension you use here is different from the original one. Is there any difference here? What is the purpose?

ssxxx1a

sample_real_and_fake_latents

2

Thanks for your great work! Sorry for my english But line 428 in coach.py. It should add latent_avg of style gan ? It's right or my mistake ? 

baopmessi

How to use multiple GPUs for training?

1

ssxxx1a

Nice work you did. When I read the code carefully, I found the code in coach.py in line 402 " self.discriminator.zero_grad() r1_final_loss = self.opts.r1 / 2 * r1_loss * self.opts.d_reg_every...

tengshaofeng

…st (batch_size) dimension, and the code breaks when using batch_size =1. pytorch lacks a functionality squeeze multiple dims at once, so using reshape() instead. patched the bug in this commit.

timbektu

Can I use my own pretrained StyleGAN2-ADA?

1

I have trained a StyleGAN2-ADA generator on my own images and now trying different inversion methods. My first question after reading your paper is if I can use my pretrained...

hamediut