/pggan-pytorch

:fire::fire: PyTorch implementation of "Progressive growing of GANs (PGGAN)" :fire::fire:

Primary LanguagePythonMIT LicenseMIT

Pytorch Implementation of "Progressive growing GAN (PGGAN)"

PyTorch implementation of PROGRESSIVE GROWING OF GANS FOR IMPROVED QUALITY, STABILITY, AND VARIATION
YOUR CONTRIBUTION IS INVALUABLE FOR THIS PROJECT :)

image

What's different from official paper?

  • original: trans(G)-->trans(D)-->stab / my code: trans(G)-->stab-->transition(D)-->stab
  • no use of NIN layer. The unnecessary layers (like low-resolution blocks) are automatically flushed out and grow.
  • used torch.utils.weight_norm for to_rgb_layer of generator.

How to use?

[step 1.] Prepare dataset
The author of progressive GAN released CelebA-HQ dataset, and I am working on it.
Before then please use CelebA to generate up to 256x256 face images. The CelebA-HQ dataloading would be supported very soon.

---------------------------------------------
The training data folder should look like : 
<train_data_root>
                |--CelebA
                        |--image1
                        |--image2
                        |--image3 ...
---------------------------------------------

[step 2.] Prepare environment using virtualenv

  • you can easily set PyTorch (v0.3) and TensorFlow environment using virtualenv.
  • CAUTION: if you have trouble installing PyTorch, install it mansually using pip. [PyTorch Install]
$ virtualenv --python=python2.7 venv
$ . venv/bin/activate
$ pip install -r requirements.txt
$ conda install pytorch torchvision -c pytorch

[step 3.] Run training

  • edit config.py to change parameters. (don't forget to change path to training images)
  • specify which gpu devices to be used, and change "n_gpu" option in config.py to support Multi-GPU training.
  • run and enjoy!
  (example)
  If using Single-GPU (device_id = 0):
  $ vim config.py   -->   change "n_gpu=1"
  $ CUDA_VISIBLE_DEVICES=0 python trainer.py
  
  If using Multi-GPUs (device id = 1,3,7):
  $ vim config.py   -->   change "n_gpu=3"
  $ CUDA_VISIBLE_DEVICES=1,3,7 python trainer.py

[step 4.] Display on tensorboard

  • you can check the results on tensorboard.

$ tensorboard --logdir repo/tensorboard --port 8888
$ <host_ip>:8888 at your browser.

[step 5.] Generate fake images using linear interpolation

CUDA_VISIBLE_DEVICES=0 python generate_interpolated.py

Experimental results

The result of higher resolution(larger than 256x256) will be updated soon.

Generated Images







Loss Curve

image

To-Do List (will be implemented soon)

  • Support WGAN-GP loss
  • training resuming functionality.
  • loading CelebA-HQ dataset (for 512x512 and 1024x0124 training)

Compatability

  • cuda v8.0
  • Tesla P40 (you may need more than 12GB Memory. If not, please adjust the batch_table in dataloader.py)

Acknowledgement

Author

MinchulShin, @nashory
image