roakum/thunderstorm-ghana-ml

Drivers of Thunderstorm Variability over Ghana # thunderstorm-ghana-ml Thunderstorm Variability over Ghana – ML Analysis Python scripts supporting the manuscript: **“Thunderstorm Drivers and Trends in Ghana, West Africa: An Interpretable Machine Learning Study”** This repository provides the core modeling, interpretability, and diagnostic codes used in the study. **Repository Contents** 1. LightGBM Model 2. SHAP Analysis 3. Correlation Matrix 4. Box plots 5. Q–Q plots **Data** Raw observational data are not included due to data sharing restrictions. Scripts assume preprocessed inputs consistent with the Methods section of the paper. **Reproducibility** Written in Python Standard scientific and ML libraries used Workflow follows the sequence described in the manuscript File paths and data loading sections may need adaptation by users. **Author** Robert Ayueboning Akum, PhD Climate Scientist **Citation** If you use this code, please cite the associated manuscript.