InfoGAN-tensorflow

InfoGAN implementation using tensorflow

[paper]

Installation

Tensorflow(>=1.4)
OpenCV
NumPy

How to run?

For training:

sh train.sh

For testing:

sh test.sh

The results will be saved in /path/to/model_dir/test

Results

Modify categorical latent code from 0 to 9(top to bottom) and
continuous latent code(c1(width) or c2(rotation)) from -2 to 2(left to right)

References

https://github.com/burness/tensorflow-101/tree/master/GAN/Info-GAN