This is an implementation of the paper Auto-Encoding with Variational Bayes by D.P.Kingma and M.Welling. This code follows pytorch/examples/vae.
pip install -r requirements.txt
python main.py
If you run the code by above command, you can get some generated images:
and reconstructions of test set digitis:
The test set ELBO after 10 epochs should be around 106
.