style-transfer-quality

style-transfer-quality copied to clipboard

Hello, Is there a way to set alpha - style-content ration in the tensorflow implementation of the algorithm? Example from pytorch version: '--alpha', type=float, default=0.2, style-content balance ratio. larger values...

C00reNUT

I succeeded to run it after changing line 63 to this ```gen_hist.append(F.to_pil_image(gen_img.cpu().squeeze(0)))```

rvencu

I have a question, is it possible to use wgan-gp to calculate the wasserstein distance between the style distributions of a batch? Thanks!

mzprose

I implemented the code in colaboratory and i want to know if this new method is slower and how much slower to the previous 2nd order method and if u...

sneccc

Getting "TypeError: can't convert CUDA tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first."

3

Hi When i run i get thid error "File "main.py", line 81, in run(args) File "main.py", line 62, in run losses_dict, gen_hist = style.transfer(args, gen_img, style_img, model) File "/home/hades/discriminator/style/__init__.py", line...

Colliwomple