Train CPPNs as a Generative Model, using Generative Adversarial Networks and Variational Autoencoder techniques to produce high resolution images.
Python
Latest commit 12de312 Apr 1, 2016 HaDavid 2nd commit
Permalink
Failed to load latest commit information.
MNIST_data first commit Apr 1, 2016
examples first commit Apr 1, 2016
save first commit Apr 1, 2016
README.md 2nd commit Apr 1, 2016
anim_gif_demo.py first commit Apr 1, 2016
images2gif.py first commit Apr 1, 2016
mnist_data.py first commit Apr 1, 2016
model.py first commit Apr 1, 2016
ops.py first commit Apr 1, 2016
sampler.py first commit Apr 1, 2016
train.py first commit Apr 1, 2016

README.md

cppn-gan-vae tensorflow

Train Compositional Pattern Producing Network as a Generative Model, using Generative Adversarial Networks and Variational Autoencoder techniques to produce high resolution images.

Morphing

Run python train.py from the command line to train from scratch and experiment with different settings.

sampler.py can be used inside IPython to interactively see results from the models being trained.

See my blog post at blog.otoro.net for more details.

I tested the implementation on TensorFlow 0.60.

Used images2gif.py written by Almar Klein, Ant1, Marius van Voorden.

License

BSD - images2gif.py

MIT - everything else