A multi-fidelity emulator for the matter power spectrum from MP-Gadget

The code is derived from the multi-fidelity emulator code reported in

M-F Ho, S Bird, and C Shelton. A Multi-Fidelity Emulator for the Matter Power Spectrum using Gaussian Processes. arXiv:2105.01081 [astro-ph.CO],

including the matter power spectrum data (z=0) to reproduce the multi-fidelity emulator trained with 50 low-fidelity simulations and 3 high-fidelity simulations.

Requirements:

• Python 3.6+
• numpy
• scipy
• GPy
• pyDOE
• emukit

Email me if there's any issues about versions: mho026-at-ucr.edu

Reproduce the emulation results of the 50LR-3HR multi-fidelity emulator

python -c "from examples.make_results import *; do_benchmark('data/50_LR_3_HR', n_optimization_restarts=20)"

Results will be generated as figures after the above command finished.

Where can I get the power spectrum data?

Simulations are run with MP-Gadget code.

A simulation submission file generator is here: github.com/jibanCat/SimulationRunnerDM. I used this generator to prepare the training/testing data in this repo.

Notebooks

Notebooks will be posted in ./notebooks/ folder.

• Build 50LR-3HR Multi-Fidelity Emulator for Matter Power Spectrum.ipynb: reproduce the results of examples.make_results code.