This is a Generative Adversarial Network (GAN) which learns to predict the dynamics of a specific attractor. It is complementary to the other project related to the deep classifier of chaos and order (see the relevant repository on my Github), and is a first step towards the described future directions in the published paper of that project. Such a GAN network can be used towards a deeper understanding of the statistical nature of chaotic dynamics, in particular the discreet time series exhibiting chaos.

The code consistes of only one Jupyter notebook, and can be executed simply by running the notebook's cells. After training, the network is able to predict a realisation of the the (discreet) chaotic dynamics of the specific system.