Data

Tools

Indexes

Applications

Experiments

Open Source Science

gaussian-process

Python script used to generate a Gaussian Process fit.

https://github.com/rferguson22/gaussian-process

Science Score: 54.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (9.8%) to scientific vocabulary
Last synced: 8 months ago · JSON representation ·

Repository

Python script used to generate a Gaussian Process fit.

Basic Info
  • Host: GitHub
  • Owner: rferguson22
  • Language: Python
  • Default Branch: master
  • Homepage:
  • Size: 55.7 MB
Statistics
  • Stars: 0
  • Watchers: 2
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created about 1 year ago · Last pushed 9 months ago
Metadata Files
Readme Citation

README.md

Gaussian Process with MCMC Length Scale Optimisation

This Python script generates a Gaussian Process (GP) using Markov Chain Monte Carlo (MCMC) to optimise the length scale from a novel loss function.
See the accompanying paper: Data‑driven Approach for Interpolation of Sparse Data.

File Formatting Assumptions

For an n-dimensional problem, the input file must contain n + 2 columns:

  1. First n columns: Input kinematic variables
  2. The (n+1)th column: Physics quantity of interest
  3. The (n+2)th column: Associated error of the physics quantity

The options.yaml file must also contain a list named resolution with n floats or integers for an n-dimensional problem.

Optional Parameters in options.yaml

| Parameter | Type | Description | Default | |------------------|-----------------|-----------------------------------------------------------------------------------------------------|----------------------------------| | MC_progress | Boolean | Whether to display the MCMC progress. | False | | out_file_name | String | File path and filename for output GP results. | "GP_results.txt" (same folder) | | MC_plotting | Boolean | Whether to plot and save MCMC corner plots, KDE peaks, and silhouette scores. | False | | labels | List of Strings | Labels for columns in the output file. Uses input file labels if available, or generic ones (dim1, dim2, ..., quantity, error). | Input-based or generic |

Citation

If you use this code, please cite:

R.F. Ferguson, D.G. Ireland & B. McKinnon,
Data‑driven Approach for Interpolation of Sparse Data,
arXiv:2505.01473 (2025).

```bibtex @misc{GP_sparse, title={Data-driven Approach for Interpolation of Sparse Data}, author={R. F. Ferguson and D. G. Ireland and B. McKinnon}, year={2025}, eprint={2505.01473}, archivePrefix={arXiv}, primaryClass={physics.data-an}, url={https://arxiv.org/abs/2505.01473}, }

Owner

  • Name: Ryan Ferguson
  • Login: rferguson22
  • Kind: user
  • Company: University of Glasgow

Citation (CITATION.cff) 

cff-version: 1.2.0
message: "If you use this code, please cite the following paper:"
authors:
  - family-names: Ferguson
    given-names: R.F.
  - family-names: Ireland
    given-names: D.G.
  - family-names: McKinnon
    given-names: B.
title: "Data-driven Approach for Interpolation of Sparse Data"
version: 1.0.0
doi: 10.48550/arXiv.2505.01473
date-released: 2025-05-02
url: https://arxiv.org/abs/2505.01473

GitHub Events

Total
  • Delete event: 1
  • Member event: 1
  • Public event: 2
  • Push event: 32
  • Create event: 4
Last Year
  • Delete event: 1
  • Member event: 1
  • Public event: 2
  • Push event: 32
  • Create event: 4

Dependencies

requirements.txt pypi
  • corner *
  • emcee *
  • matplotlib *
  • multiprocessing *
  • numpy *
  • pandas *
  • pathlib *
  • scikit-learn *
  • scipy *
  • smt *
  • yaml *