Comments (4)
I have the same confuse about the gan loss part... the gan loss in the code seem not do adversarial train
from aot-gan-for-inpainting.
The same question
from aot-gan-for-inpainting.
g_fake_label = torch.ones_like(g_fake).cuda() is wrong for gan, g_fake_label = torch.zeros_like(g_fake).cuda() is right. The author handle the wrong with parser.add_argument('--adv_weight', type=float, default=0.01,help='loss weight for adversarial loss'),no use for adv loss for training for netG.
from aot-gan-for-inpainting.
d_fake_label = gaussian_blur(masks, (self.ksize, self.ksize), (10, 10)).detach().cuda()
d_real_label = torch.zeros_like(d_real).cuda()
# g_fake_label = torch.ones_like(g_fake).cuda()
g_fake_label = torch.zeros_like(g_fake).cuda()
dis_loss = self.loss_fn(d_fake[masks>0.5], d_fake_label[masks>0.5]) + self.loss_fn(d_real, d_real_label)
gen_loss = self.loss_fn(g_fake[masks>0.5], g_fake_label[masks>0.5])
from aot-gan-for-inpainting.
Related Issues (18)
- Loading pre-trained model throws error HOT 3
- 用celeba_256数据集从头训练为什么无法收敛?
- Cannot train the model HOT 2
- c++ HOT 2
- Own dataset HOT 1
- Please update this git with latest version of packages
- Cannot test HOT 1
- training failed HOT 3
- Unable to train HOT 1
- 使用训练权重验证出现问题 HOT 1
- 请问图像和掩膜应该在哪里获取呢
- 分布式训练
- Cannot reproduce results from paper with pre-trained model HOT 1
- Do the traning masks need to be hand-crafted generated? HOT 4
- Bugs in Adversarial Training. HOT 1
- Style Loss is the major player for good results and not Adversarial Loss. HOT 1
- Picking the right model for random use-case
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from aot-gan-for-inpainting.