This library implements some of the most common (Variational) Autoencoder models. In particular it provides the possibility to perform benchmark experiments and comparisons by training the models with the same autoencoding neural network architecture. The feature make your own autoencoder allows you to train any of these models with your own data and own Encoder and Decoder neural networks.

PyPI: https://pypi.org/project/pythae