Code for the paper "Understanding and Improving Interpolation in Autoencoders via an Adversarial Regularizer" by David Berthelot, Colin Raffel, Aurko Roy, and Ian Goodfellow.

This is not an officially supported Google product.

sudo apt install virtualenv

cd <path_to_code>
virtualenv --system-site-packages env2
. env2/bin/activate
pip install -r requirements.txt

Choose a folder where to save the datasets, for example ~/Data

export AE_DATA=~/Data

python create_datasets.py

CUDA_VISIBLE_DEVICES=0 python acai.py \
--train_dir=TEMP \
--latent=16 --latent_width=2 --depth=16 --dataset=celeba32

All training from the paper can be found in folder runs.

These are the maintained models:

• aae.py
• acai.py
• baseline.py
• denoising.py
• dropout.py
• vae.py
• vqvae.py

• classifier_fc.py: fully connected single layer from raw pixels, see runs/classify.sh for examples.
• Auto-encoder classification is trained at the same as the auto-encoder.
• cluster.py: K-means clustering, see runs/cluster.sh for examples.

• create_datasets.py: see Installing datsets for more info.