Skip to content
/ VAE Public

Saves the code to train a VAE for small images. Details in readme

Notifications You must be signed in to change notification settings

ZeratuuLL/VAE

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 

Repository files navigation

VAE

Full name for VAE is Variational Auto-Encoder. It is a special case of AEVB(Auto-Encoding VB) algorithm when the recognition model. There will also be a short introduction for this in the jupyter notebook. For more details please read the original paper Auto-Encoding Variational Bayes

In the jupyter notebook, I use pytorch to implement this encoder and trained it separately on MNIST and some images collected from Pong game to see the effect. Some illustrations were done there. The next step will be training some RL agent in the encoded space and compare the result which is obtained by training in the original space.

About

Saves the code to train a VAE for small images. Details in readme

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published