This my simple implementation of Conditional Variational Autoencoder for Neural Machine Translation paper by Artidoro Pagnoni, Kevin Liu and Shangyan Li
The project is built on top of Pytorch 1.3.0
.
To install all requirement packages, run the command:
pip install -r requirements.txt
The Conditional Variational Autoencoder (CVAE) is a an algorithm to generate certain image (e.g. generation of the digit image of given label) from the latent space
The model was taught on MNIST dataset, which can be obtained as:
torchvision.datasets.MNIST
For demonstration 10 random digits were generated