faltysad / bachelor-thesis-code

Tensorflow implementation of VAE model for MNIST generative modeling.

Variational autoencoder MNIST generative model

save folder contains pre-trained model weights after 500 training epochs. Delete it if you want to run your own training phase.

In order to perform model training phase, run:

python vae_mnist_train.py

In order to visualise trained model, run:

jupyter notebook eval.ipynb