-
Did you try to use transfer learning for generator training, ie using FFHQ generator weights as initial state for LSUN datasets training?
-
Hi!
I am a big fan of your work. For another paper I am working on, it would be great to have your pretrained LSUN styleGAN generator. Is it possible to make that publicly available? Thanks!
-
Hi Zach,
Great product!
Possibly very minor and maybe nothing to do with your code, but when trying to run the `mcmc.ipynb` notebook, I encounter an issue with `corner` in the final block:
`K…
-
Thank you for your awesome work! I'm very interested in the paper. I saw that you planned to publish the pre-trained model, will you upload it soon? I'm very excited to test the models tbh.
Thanks ag…
-
## Steps to Reproduce
Add the google maps plugin and create a widget according to the tutorial.
**Expected results:**
No log mesages, Or logmessages that are relevant to the library user
**…
-
### What target platforms are you seeing this bug on?
Android
### Steps to reproduce
1. Enable Impeller on Android
` `
2. Use native views directly or through plugin
### Expected res…
-
**Describe the bug**
When I try to scan on android, I only get a blank screen in the background color of my app. It does not show a viewfinder or anything else. The same code works perfectly on iOS. …
-
` model = torch.hub.load("Lornatang/WassersteinGAN_DIV-PyTorch", args.arch, progress=True, pretrained=True,
verbose=False)`
Downloading: "https://github.com/Lornatang/W…
-
-
想请问一下博主,我使用LSUN的数据集,初始学习率2e-04,使用adam优化器,为什么会有训练到一半突然loss增大的问题。
期待您的回复。