add baseline GAN models - Githubissues

danmar3 / AssociativeGAN

0 stars 0 forks source link

add baseline GAN models #6

Open danmar3 opened 4 years ago

danmar3 commented 4 years ago

add models to compare the performance of our models

start by incorporating models from tf-gan examples.

[x] DCGAN: added in https://github.com/danmar3/AssociativeGAN/commit/23985819c6c41b04169485aa2c442f0780a27a09 , https://github.com/danmar3/AssociativeGAN/commit/c65fc0a960c394e86aef3ab056a029e7847bdfaa
[ ] Info GAN
[ ] Wasserstain GAN

To consider:

[ ] ACGAN https://arxiv.org/pdf/1610.09585.pdf
[ ] ClusterGAN https://arxiv.org/pdf/1809.03627.pdf
[ ] GMM GAN https://arxiv.org/pdf/1808.10356.pdf
[ ] Progressive growing of gans
[ ] VEEGAN https://arxiv.org/pdf/1705.07761.pdf
[ ] Self-Attention GAN https://arxiv.org/pdf/1805.08318.pdf
[ ] Large scale GAN (res-GAN) https://arxiv.org/pdf/1809.11096.pdf

Morgan243 commented 4 years ago

Facebook research has some GAN tools, but only a few models and metrics: https://github.com/facebookresearch/pytorch_GAN_zoo

They provide the PGAN referenced on PyTorch site: https://pytorch.org/hub/facebookresearch_pytorch-gan-zoo_pgan/
They provide the DCGAN referenced on PyTorch site: https://pytorch.org/hub/facebookresearch_pytorch-gan-zoo_dcgan/

A random guy with a decent library of torch GANs: https://github.com/znxlwm/pytorch-generative-model-collections

Of the ones you listed in the issue, it has InfoGAN, WGAN, ACGAN
He references a tensorflow library you probably already know about: https://github.com/hwalsuklee/tensorflow-generative-model-collections