confifu / tryongan-pytorch Goto Github PK

Hi confifu,
I could successfully run your code and get the same result in the folder "results/poseConditioned/" as yours according to your train notebook and the stylegan2 notebook.

But I'm a little confused with the generated images, because they don't look like normal photos (they are red or green). I was wondering if they are initial generated results.

In the training notebook:
/gdrive/MyDrive/tryongan_models_results/models/poseConditioned
There's only "model_0.pt". I was wondering if it's convenient for you to provide the complete pretrained models?

I could run the train,py and generate the pickle file.

For testing, which file need to be run. Is there a Testing Notebook? I tried to run style_mixing.py. Got error on G.synthesis expecting pose. Made the required changes. After the changes, getting error RuntimeError: Expected object of device type cuda but got device type cpu for argument #1 'self'. Thanks for the help