https://github.com/landscapegeoinformatics/spatial-ml-soc-2024

Code supplement: Spatial autocorrelation in machine learning for modelling soil organic carbon
https://github.com/landscapegeoinformatics/spatial-ml-soc-2024

Last synced: about 1 month ago
JSON representation

Code supplement: Spatial autocorrelation in machine learning for modelling soil organic carbon

• Host: GitHub
• URL: https://github.com/landscapegeoinformatics/spatial-ml-soc-2024
• Owner: LandscapeGeoinformatics
• License: mit
• Created: 2024-11-28T09:20:04.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-11-28T15:19:20.000Z (over 1 year ago)
• Last Synced: 2025-02-28T12:47:07.480Z (over 1 year ago)
• Language: Jupyter Notebook
• Size: 27.3 KB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# spatial-ml-soc-2024

Code supplement: Spatial autocorrelation in machine learning for modelling soil organic carbon

[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.14236923.svg)](https://doi.org/10.5281/zenodo.14236923)

Folders and files:

- `data`: reference to the input/source datasets and the conda environment yaml

- `model_test`: fresh training and (cross-)validation, numbers may slightly vary

- `predict_full`: scripts that were used to predict to the of Estonia as 10m raster

Errata:

In the manuscript the term RFSI (Random Forest Spatial Interpolation) is used, in the scripts and data the RFSI-associated model is named KNN (K-Nearest-Neighbours)