/thindeepgps

Reference implementation of Thin and Deep Gaussian Processes (NeurIPS 2023)

Primary LanguagePythonApache License 2.0Apache-2.0

Install

This library was implemented against Python 3.10.8. To install it and its dependencies, please do:

pip install -r requirements.txt
pip install --no-deps uq360==0.2
python setup.py develop

Run:

To reproduce the plots and results of our synthetic experiment, please run:

python 'synthetic experiment.py'

This script will generate the plots used in paper alongside additional plots. This script is also formatted as a percent format notebook that can also be open as a jupytext notebook or in many IDEs (Spyder, VS Code, PyCharm, etc).