Comments (7)
This is a known issue and correct
from esrgan-pytorch.
Thanks for your fast answer!
Is there any known workaround or solution for that?
from esrgan-pytorch.
Thanks for your fast answer!
Is there any known workaround or solution for that?
Hello,
You can check here https://github.com/AlexiaJM/RelativisticGAN, which mentioned if you use BCELoss (which is used in ESRGAN), you need to add sigmoid as the last layer of the Discriminator (which is not included here). However, BCEWithLogitsLoss is recommended as it will do sigmoid implicitly.
from esrgan-pytorch.
Thanks for your fast answer!
Is there any known workaround or solution for that?Hello,
You can check here https://github.com/AlexiaJM/RelativisticGAN, which mentioned if you use BCELoss (which is used in ESRGAN), you need to add sigmoid as the last layer of the Discriminator (which is not included here). However, BCEWithLogitsLoss is recommended as it will do sigmoid implicitly.
Hello,
"add sigmoid as the last layer of the Discriminator" can't solve the problem, because "hr_output" will still larger than "sr_output" sometime.
from esrgan-pytorch.
This is a known issue and correct
Hello,
Is this problem really corrected? i still get the same problem.
from esrgan-pytorch.
If you use BCEWithLogits, it will be more stable than SIgmoid+BCELoss,
from esrgan-pytorch.
This point is not mentioned in the original paper, but it is actually a key factor in the successful training of GAN
from esrgan-pytorch.
Related Issues (20)
- resume训练模型时不太对劲 HOT 6
- Keep ratio input image not upscale HOT 4
- How to train the ESRGAN on my own dataset HOT 31
- How to train and test in google colab ? HOT 1
- Can you reproduce the psnr? HOT 5
- 请问是否尝试在ESRGAN上用更大的缩放倍数?比如X8? X16? HOT 3
- 相对GAN的损失函数 HOT 2
- Pretrained weights HOT 1
- Model Weights HOT 2
- Version Requirements not specified in Readme. HOT 1
- Test Spatial size? HOT 1
- convert to onnx
- Inquiries on AMP HOT 1
- Run test.py encounter TypeErrpr: 'type' object is not subscriptable HOT 1
- How to test on rrdbnet? HOT 1
- only support windows,very not friendly HOT 1
- a small bug in train_gan.py ? HOT 1
- 关于 train_gan.py
- 修改重建倍数 HOT 3
- 修改features.34 HOT 1
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 esrgan-pytorch.