visonpon / SpectralNormalizationKeras

Spectral Normalization for Keras Dense and Convolution Layers

Spectral Normalization for Keras

The simple Keras implementation of ICLR 2018 paper, Spectral Normalization for Generative Adversarial Networks

CIFAR10 epoch 245 (ResNet architecture)

CIFAR10 epoch 245 (DCGAN architecture)

How to use?

1. Move SpectralNormalizationKeras.py in your dir
2. Import these layer class

from SpectralNormalizationKeras import DenseSN, ConvSN1D, ConvSN2D, ConvSN3D

3. Use these layers in your discriminator as usual

Example notebook

CIFAR10 with DCGAN architecture

CIFAR10 with ResNet architecture

Model Detail

Architecture

DCGAN

Generator

Discriminator

ResNet GAN

Generator

Generator UpSampling ResBlock

Dicriminator

Discriminator DownSampling ResBlock

Discriminator ResBlock

Issue

1. Compare with SELU and WGAN-GP
2. Projection Discriminator