A scalable land cover classification workflow aimed to be able to scale to cover the Mediterranean bassin.
A research article describing the methodology followed on this workflow can be found at:
Scalable approach for high-resolution land cover: a case study in the Mediterranean Basin.
Antonio Manuel Burgueño, José F. Aldana-Martín, María Vázquez-Pendón, Cristóbal Barba-González, Yaiza Jiménez Gómez, Virginia García Millán & Ismael Navas-Delgado
Journal of Big Data 10, 91 (2023). doi: 10.1186/s40537-023-00770-z
The package is not available on PyPI, so you need to install it from the source code. To do so, you can clone the repository and install it using pip:
git clone https://github.com/KhaosResearch/landcoverpy.git
cd landcoverpy
pip install .For development purposes, you can install the package in editable mode:
pip install -e .An usage example can be found at the main usage notebook. Example input data in different formates can be found at validated_data. Example label mappings can be found at label_mappings.
Most of the configuration is done through environment variables. You should create a .env file following the .env.template file. Default values can be observed in config.py.
This project is licensed under the MIT license. See the LICENSE file for more info.