style-based-gan-pytorch
style-based-gan-pytorch copied to clipboard
rosinality
Implementation A Style-Based Generator Architecture for Generative Adversarial Networks in PyTorch
When I train a new model, the gradient of R1 is 10^-13. If it is a normal value? Thanks!
Hi , what a wonderful project! I successfully run it on my computer！ I just wonder why the following code in Line 106, train.py does not raise any error like...
Dear author: Thanks for your amazing repo, it helped me understanding stylegan more. I got stuck when tracing your code at model.py I didn't understand what blur functions do and...
Hi, when I finished the task of training model using customized datasets and tested with command "python generate.py checkpoint/010000.model", I encountered the error as following: """" generator.load_state_dict(torch.load(args.path)['g_running']) KeyError: "g_running" """"...
Wondering - is there any reason that num_workers is hardcoded to 1, and there's no usage of more workers?
It seems like in line 80 of train.py, the step is changed only when used_sample > phase x 2. This makes sense in resolution >= 16x16, because each step is...
Hey, I am looking to run your code on CelebA dataset and I can get the code to work but it takes really long to train this (given the resources...
In the function forward of class Generator, if the len of styles is more than 2, there will be a bug. Because the inject_index = random.sample(list(range(step)), len(style) - 1) is...
Hi, I think tihis a great work. But I have a quitions about the gen_in1 and gen_in2 , when traning the discriminator, the latent feature is gen_in1 while when traning...
Metadata
Owner
Metadata
Implementation A Style-Based Generator Architecture for Generative Adversarial Networks in PyTorch