This is a neural networks from scratch library which has been created for the course “Numerical methods of algorithmic systems and neural networks” which has been taught in the summer semester 2020 at the Leibniz Universität Hannover by Prof. Thomas Wick.
Code examples for a classifier, a denoising autoencoder, a variational autoencoder and a generative adversarial network for the MNIST dataset can be found in the folder "notebooks".
For more information about this project please refer to the online documentation:
https://julianroth.org/documentation/neural_networks/