Convolutional Variational Autoencoder on the SVHN Dataset

Data Description

python3 -m venv venv
source venv/bin/activate
pip install requirements.txt

python train.py --gen_images_dir images --num_epochs=100 --batch_size=64

Images will be samples and generated at the end of each epoch in the --gen_images-dir directory.