NIFTy.re

What is NIFTy.re?

NIFTy, “Numerical Information Field Theory“, is a Bayesian imaging library. It is designed to infer the million to billion dimensional posterior distribution in the image space from noisy input data. At the core of NIFTy lies a set of powerful Gaussian Process (GP) models and accurate Variational Inference (VI) algorithms. This document gives a short summary on the NIFTy.re implementation. In the following, an online version of the latest publication describing NIFTy.re is provided. The publication gives a quick overview of the core concepts powering NIFTy.re. Furthermore we showcases simple examples, and demonstrate the Correlated field model in NIFTy.re.

• Publication
• Introductory Notebooks
• Spatio-spectral plotting