https://github.com/nashory/progressive-growing-torch
Torch implementation of "PROGRESSIVE GROWING OF GANS FOR IMPROVED QUALITY, STABILITY, AND VARIATION"
https://github.com/nashory/progressive-growing-torch
celeba generative-adversarial-network high-resolution nvidia progressively-growing-gan unsupervised-learning
Last synced: 5 months ago
JSON representation
Torch implementation of "PROGRESSIVE GROWING OF GANS FOR IMPROVED QUALITY, STABILITY, AND VARIATION"
- Host: GitHub
- URL: https://github.com/nashory/progressive-growing-torch
- Owner: nashory
- License: mit
- Created: 2017-10-30T01:48:29.000Z (almost 8 years ago)
- Default Branch: master
- Last Pushed: 2017-11-15T02:57:22.000Z (almost 8 years ago)
- Last Synced: 2025-04-02T20:38:32.694Z (7 months ago)
- Topics: celeba, generative-adversarial-network, high-resolution, nvidia, progressively-growing-gan, unsupervised-learning
- Language: Lua
- Size: 2.53 MB
- Stars: 35
- Watchers: 6
- Forks: 8
- Open Issues: 12
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Progressive Growing of GANs for Improved Quality, Stability, and Variation
-------
[NOTE] This project was not goint well, so I made PyTorch implementation here. :fire: [[pggan-pytorch]](https://github.com/nashory/pggan-pytorch)
-------
Torch implementation of [PROGRESSIVE GROWING OF GANS FOR IMPROVED QUALITY, STABILITY, AND VARIATION](http://research.nvidia.com/sites/default/files/pubs/2017-10_Progressive-Growing-of//karras2017gan-paper.pdf)
__YOUR CONTRIBUTION IS INVALUABLE FOR THIS PROJECT :)__
## NEED HELP
[ ] (1) Implementing Pixel-wise normalization layer
[ ] (2) Implementing pre-layer normalization (for equalized learning rate)
__(I have tried both, but failed to converge. Anyone can help implementing those two custom layers?)__
## Prerequisites
+ [Torch7](http://torch.ch/docs/getting-started.html#_)
+ [display](https://github.com/szym/display)
+ [nninit](https://github.com/Kaixhin/nninit)
## How to use?
__[step 1.] Prepare dataset__
CelebA-HQ dataset is not available yet, so I used 100,000 generated PNGs of CelebA-HQ released by the author.
The quality of the generated image was good enough for training and verifying the preformance of the code.
If the CelebA-HQ dataset is releasted in then near future, I will update the experimental result.
[[download]](https://drive.google.com/open?id=0B4qLcYyJmiz0MUVMVFEyclJnRmc)
+ CAUTION: loading 1024 x 1024 image and resizing every forward process makes training slow. I recommend you to use normal CelebA dataset until the output resolution converges to 256x256.
~~~
---------------------------------------------
The training data folder should look like :
|--classA
|--image1A
|--image2A ...
|--classB
|--image1B
|--image2B ...
---------------------------------------------
~~~
__[step 2.] Run training__
+ edit script/opts.lua to change training parameter. (don't forget to change path to training images)
+ run and enjoy! (Multi-threaded dataloading is supported.)
`$ python run.py`
__[step 3.] Visualization__
+ to start display server:
`$ th server.lua`
+ to check images during training procudure:
`$ : at your browser`
## Experimental results
### Transition experiment: (having trouble with transition from 8x8 -> 16x16 yet.)
## What does the printed log mean?
~~~
(example)
[E:0][T:91][ 91872/202599] errD(real): 0.2820 | errD(fake): 0.1557 | errG: 0.3838 [Res: 4][Trn(G):0.0%][Trn(D):0.0%][Elp(hr):0.2008]
~~~
+ E: epoch / T: ticks (1tick = 1000imgs) / errD,errG: loss of discrminator and generator
+ Res: current resolution of output
+ Trn: transition progress (if 100%, in training phase. if less than 100%, in transition phase using fade-in layer.)
+ first Trn : Transition of fade-in layer in generator.
+ second Trn : Transition of fade-in layer in discriminator.
+ Elp(hr): Elapsed Time (Hour)
## To-Do List (will be implemented soon)
- [X] Equalized learning rate (weight normalization)
- [X] Support WGAN-GP loss
## Compatability
+ cuda v8.0
+ Tesla P40 (you may need more than 12GB Memory. If not, please adjust the batch_table in `pggan.lua`)
+ python 2.7 / Torch7
## Acknowledgement
+ [tkarras/progressive_growing_of_gans](https://github.com/tkarras/progressive_growing_of_gans)
+ [soumith/dcgan.torch](https://github.com/soumith/dcgan.torch)
## Author
MinchulShin, [@nashory](https://github.com/nashory)
