This project focuses on deep generative models and their ability to model the probability distribution of the data. It consists of two parts: VAE (Variational Autoencoder) and AAE (Adversarial Autoencoder).
The VAE folder contains scripts and README.md related to the Variational Autoencoder. The VAE is a generative model that learns to encode and decode data by approximating the underlying probability distribution. It is trained using a combination of a reconstruction loss and a regularization term.
The README.md file in the VAE folder provides detailed information about the VAE implementation, including instructions on how to run the code and explanations of the key concepts.
The AAE folder contains scripts and README.md related to the Adversarial Autoencoder. The AAE is another type of generative model that combines elements of both autoencoders and generative adversarial networks (GANs). It learns to generate data by training a generator network to fool a discriminator network.
The README.md file in the AAE folder provides detailed information about the AAE implementation, including instructions on how to run the code and explanations of the key concepts.
To get started with this project, please refer to the README.md files in the VAE and AAE folders for specific instructions on running the code and reproducing the results.