Comments (2)
Sorry for bringing this up after such a long time, but did you ever figure out why this is the case? I tried to change normalizing LR and HR images to [-1,1] in my own implementation to maybe improve performance, but I couldn't figure out the additional changes needed to the various preprocess and deprocess calls to get consistent results.
Right now it seems like I'm having issues with this line.
Line 143 in 3a75d69
I'm thinking I can remove the preprocess/deprocess calls here, if both LR and HR footage is normalized to [-1,1], but I get weird color shifts/ flickering in the WarpPreGen tensorboard summaries, when I do.
I'm using my own paired LR data, not the default gaussian downsample from HR training data, if that changes anything.
from tecogan.
It seems that two different normalizing strategies originate from historic reason and don't have huge impact to the performance.
The color shifts/flickering may be caused by something else, in my opinion.
from tecogan.
Related Issues (20)
- How do I continue the training from the point where it stopped? I encountered a sudden shutdown HOT 3
- How to run the training on multiple GPUs?
- can you guys run TG with your Nvidia 30 series GPU? HOT 3
- What's the difference between input_dir_LR and input_dir_HR? HOT 1
- Specify latest supported Tensorflow version HOT 1
- Error training own dataset - LR + HR HOT 2
- successful operation but no output HOT 1
- dataPrepare.py don't work
- Evaluation Metrics
- Low gpu usage
- testing test case 1
- Can anyone share the TG-Z415 model?
- self._handle = _dlopen(self._name, mode)
- Error when running runGan.py 1 on colab
- does not run python 3 runGan.py 0
- Any advice on Postprocessing
- Fleje
- Video
- Training the TecoGAN model into REDS Dataset
- ValueError: Moving average already computed for: generator_loss/content_loss/Mean:0
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 tecogan.