anananan116 / VQ-VAE-Image-Generation Public

Notifications You must be signed in to change notification settings
Fork 0
Star 0

Image generation using vector quantized autoencoders

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
autoencoders		autoencoders
configs		configs
data_utils		data_utils
trainers		trainers
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
run.py		run.py
visualization.ipynb		visualization.ipynb

Repository files navigation

VQ-VAE-VQ-GAN-Image-Generation

Image generation usinng vector quantized autoencoders

Download data from celebA-HQ here, and put unziped images into ./data_utils/raw_data

It needs about 40-50 GB of RAM(not GRAM) to load the data and transform them. I will implement transformation on the fly so it need less ram to prepare the data in future versions.