This project moved to GitLab: https://gitlab.com/qualikiz-group/QLKNN-4Dkin

This repository contains the 4D-kinetic QuaLiKiz Neural Network. It was used for the paper "Real-time-capable prediction of temperature and density profiles in a tokamak using RAPTOR and a first-principle-based transport model" by F. Felici, J. Citrin et al. http://iopscience.iop.org/article/10.1088/1741-4326/aac8f0.

Note that is a proof-of-concept, and as such, a very limited model. This repository is for reference only. If you are interested in using this model for your own purposes, it it advised to wait for the QuaLiKiz 10D Neural Network by K.L. van de Plassche, currently under development at https://github.com/Karel-van-de-Plassche/QLKNN-develop