TF-GAN is a lightweight library for training and evaluating Generative Adversarial Networks (GANs).
pip
using pip install tensorflow-gan
, and usedimport tensorflow_gan as tfgan
TF-GAN is composed of several parts, which are designed to exist independently:
Inception Score
, Frechet Distance
,Kernel Distance
with a pretrained Inception network to evaluate yourNumerous projects inside Google. The following are some published papers that use TF-GAN:
The framework Compare GAN uses TF-GAN, especially the evaluation metrics. Their papers use TF-GAN to ensure consistent and comparable evaluation metrics. Some of those papers are:
Training in TF-GAN typically consists of the following steps:
GANModel
.GANLoss
.GANTrainOps
.At each stage, you can either use TF-GAN's convenience functions, or you can perform the step manually for fine-grained control.
There are various types of GAN setup. For instance, you can train a generator to sample unconditionally from a learned distribution, or you can condition on extra information such as a class label. TF-GAN is compatible with many setups, and we demonstrate in the well-tested examples directory