Skip to content
No description, website, or topics provided.
Jupyter Notebook Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
dataset
notes
plots
.gitignore
Calibration.ipynb
ClassifierTest.ipynb
Distribution.ipynb
Distribution2.ipynb
MnistDetection(1).ipynb
PlotElboBounded.ipynb
TractabilityProof.ipynb
Untitled.ipynb
abstract_network.py
elbo_bound.py
elbo_mi.py
elbo_mi_binary.py
elbo_mog.py
eval_inception.py
eval_ll.py
exploding_latent_visualization.py
gibbs_net.py
hierarchical_vae.py
launch_remote.sh
launch_remote2.sh
launch_script.sh
launch_script2.sh
limited_mnist.py
logger.py
lsun_vae.py
matrix_proof.py
mmd_mog.py
mmd_vae-semi-supervised.ipynb
mmd_vae.ipynb
mmd_vae.py
mmd_vae_eval.ipynb
mmd_vae_eval.py
model_bigan.py
model_binary_bigan.py
model_binary_wgan.py
model_vae.py
model_wgan.py
plot.ipynb
plot.pdf
plot.py
plot2.ipynb
readme.md
sample_points_50.zip
sparse_vae.py
tri_gan.py
vae.ipynb

readme.md

Tensorflow Implementation of MMD Variational Autoencoder

Details and motivation are described in this paper or tutorial. For your convenience the same code is provided in both python and ipython.

This implementation trains on MNIST, generating reasonable quality samples after less than one minute of training on a single Titan X

mnist

When latent dimensionality is 2, we can also visualize the distribution of labels in the feature space.

mnist

You can’t perform that action at this time.